It could also be easier to cite policy when referring to it. [[User:P,TO 19104|P,TO 19104]] ([[User talk: P,TO 19104|talk]]) ([[Special:Contributions/P,TO 19104|contributions]]) 22:29, 21 June 2020 (UTC)
It could also be easier to cite policy when referring to it. [[User:P,TO 19104|P,TO 19104]] ([[User talk: P,TO 19104|talk]]) ([[Special:Contributions/P,TO 19104|contributions]]) 22:29, 21 June 2020 (UTC)
:Hi, while your idea is forward looking in concept it has several problems:
:# Any codification is a fork of the Rules/Policies/Guidelines/Generally accepted practices/essays and would quickly become out of date. We're contstantly revising and updating based on experiences we have.
:# very few new editors are going to spend ~10 hours reading through codifications to understand all the nuances of the corpus of behaviors before making their first edit.
:# By codifying the above you invite [[WP:POINT|point making edits]] and having to [[WP:BEANS|explicitly tell editors to not shove beans up their noses]].
: Sorry to be such a downer, but make some edits, get to learn the culture of WP, and remember the [[WP:5P|Five Pillars]] and you should be safe to do things. [[User:Hasteur|Hasteur]] ([[User talk:Hasteur|talk]]) 23:08, 21 June 2020 (UTC)
The policy section of the village pump is used to discuss proposed policies and guidelines and changes to existing policies and guidelines.
If you want to propose something new that is not a policy or guideline, use Village pump (proposals).
If you have a question about how to apply an existing policy or guideline, try one of the many Wikipedia:Noticeboards.
This is not the place to resolve disputes over how a policy should be implemented. Please see Wikipedia:Dispute resolution for how to proceed in such cases.
Please see this FAQ page for a list of frequently rejected or ignored proposals.
RfC: Locality categorization by historical subdivisions
While there is significant support for limiting the categorization of current localities by historical subdivisions in some way, no clear criterion emerges to do so systematically.
The consensus is to stick with the status quo (option C), by using common sense to assess on a case-by-case basis whether the categorization of a page reflects a relevant historical fact rather than a cataloguing attempt.
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Question: What should the general rule/principle/guideline be for categorizing current localities by historical administrative subdivision in Central and Eastern Europe? There are quite a few articles of cities and towns that have been categorized not only in which administrative subdivision they currently are in, but also by the former subdivisions.
A: categorization should be limited - by what? Whether it is referenced in the article? How long the subdivision lasted? How large the subdivision was? To the 1st-level former subdivision? To how recent subdivision was? Etc?
B: categorization should not be allowed (i.e. current localities should be removed from the former subdivision categories; historical information could be preserved in a different venue like a separate list or an addition to the locality article or something similar to the "historical affiliation" box as in Görlitz#History)
C: status quo; no general rules; specific issues with individual categories should be addressed at WP:CfD
22:24, 21 February 2020 (UTC)
Major concerns with such categories:
WP:OR/WP:V: many of the locality articles do not even mention or reference former subdivisions. In Eišiškės example above, only Nowogródek Voivodeship is mentioned in the article body (added by me 12 years ago without a reference). What is the basis to claim it was in the Lidsky Uyezd? An editor looking at a map? Finding out former subdivisions is not always straightforward, particularly for smaller towns or for older subdivisions – some medieval regions did not have well defined borders, while in more recent years administrative border adjustments are frequent.
WP:NONDEF: if many of the articles don't even mention the historical subdivision, it cannot be the defining characteristic (which is the central goals of the categorization system).
Confusion for readers: in the example of Eišiškės above, could you tell which of the 6 categories is for the current and which is for the former subdivision? (this could be somewhat alleviated by better category names)
Clutter/maintainability: Görlitz lists 23 different countries/states (not to mention subdivisions) that it was a part of. Should all of these be represented in a category? If not all, then which ones?
Why this RfC? There were some CfD discussions over the years (ones that I am aware Aug 2015 (delete), Sept 2015 (delete), Oct 2015 (no consensus), Apr 2017 (no consensus)) but they did attract much attention (unlike AfD, CfD rarely attracts outsider attention), yielded inconsistent results, and did not hash out what should be done with these categories in general. And these categories keep proliferating. Therefore, looking for a broader principle-based discussion here, rather than individual consideration of specific categories at CfD.
Side note: some locality articles have "historical affiliation" boxes (example: Görlitz#History), though in some others it was removed as "nightmares" or "LISTCRUFT". And a user got blocked for adding them (and refusing to communicate).
Opinion poll: Locality categorization by historical subdivisions
Please place your !vote here.
A: definitely should be limited to may be current immediate subdivision and may be the historical in which a populated place was established. For the "historical affiliation" box mentioned above for Gorlitz, it should be avoided as a spam as it simply fails the Manual of Style for flags WP:MOSFLAG and infringes on original research WP:OR due to political speculations. Aleksandr Grigoryev (talk) 22:59, 21 February 2020 (UTC)[reply]
Aleksandr Grigoryev: I thought about it, and I don't think it's a workable solution. Many places don't have a specific founding date and they are just mentioned in written sources in year x, or even more broadly in century y. Plus what makes the first subdivision so special? Further, I don't think it's maintainable. If you think about it, it still means that there will have to be categories for all historical subdivisions of that region as localities were founded/mentioned in different times. So, for example, there will have to be a category for Vilnius Voivodeship that contains localities founded/first mentioned in 1413-1795 and for Lidsky Uyezd that contains localities founded/first mentioned in 1795-1915. But then, it's likely that someone will decide that the category on Lidsky Uyezd is not comprehensive and start adding articles purely by geographic location. Renata (talk) 03:15, 24 February 2020 (UTC)[reply]
C. This is far too broad a question and these things badly need to be determined on a case by case basis. Some of the above shouldn't have locality categorisation in this way. Some of them should. The idea that we can answer them on a global basis with reference to a handful of subdivisions in eastern Europe is the sort of discussion that leads to all kinds of ridiculous situations when applied to local situations in places nobody was giving thought to. The Drover's Wife (talk) 00:54, 22 February 2020 (UTC)[reply]
The Drover's Wife: not really looking to write any policy here, but just to get a rough idea/consensus from the wider community on what categories should or should not be present in locality articles. It would be very helpful if you could expand on your comment "Some of the above shouldn't have locality categorisation in this way. Some of them should." -- which should (not) and based on what criteria? Even if just considering the examples listed above. Renata (talk) 01:46, 22 February 2020 (UTC)[reply]
I am woefully under-educated about the history of this specific region and I'd hate to give pronouncements on things I don't understand well enough to have a sensible opinion. I'm just extremely cautious of a discussion like this creating a rule that then gets applied to completely different circumstances in other places. The Drover's Wife (talk) 03:23, 22 February 2020 (UTC)[reply]
B (A if we have to): Limited to current subdivisions only, as has been the long established practice; bio articles relevant to the polity itself are also currently placed in the category named for that polity -- it is Category:People of medieval Wallachia, but notCategory:People from Saac County (i. e. a defunct county in said Wallachia). This avoids a massive overcrowding. I don't see when populated places would be placed even in articles pertaining to those polities, let alone their subdivisons; only nostalgia and irredentism can be the driving factors here, and neither is encyclopedic. Current subdivision also establishes a neutral standard: populated places that were once in Romania are categorized by their current subdivision in Ukraine, but the same standards would apply to localities in Romania that were once in Hungary. Dahn (talk) 05:46, 22 February 2020 (UTC)[reply]
A or B, one could say "A, because we should allow this if a historical subdivision is a defining characteristic of a locality", but in practice it never is a defining characteristic, so A and B are very similar. Marcocapelle (talk) 08:47, 22 February 2020 (UTC)[reply]
A. Current and historical are enough. Historical division/subdivision should at least be mentioned in prose before including it. In addition, as already noted by other editor, the "Historical affiliations", including the mentioned problems, should be removed, because it is unsourced, trivial, and just takes up unnecessary space of the page. – Sabbatino (talk) 11:13, 22 February 2020 (UTC)[reply]
C I'm with The Drover's Wife on this. It's unwise to make policy decision on such a broad front. Examples can be listed of multiple short-lived political entities to which a city may have been attached over many centuries; it would probably be excessive to make the city a child of all of them. Cities changed hands multiple times in the Holy Roman Empire. On the other hand some administrative sub-divisions, while practically defunct, nevertheless remain on the statute books. For example Thurles (civil parish) is in the ancient barony of Eliogarty. While Eliogarty no longer has a practical administrative function, it has never been legally abolished. I would not like to see Thurles being removed from Category:Eliogarty. In summary, such thingsare best decided on a case-by-case, CFD basis. Laurel Lodged (talk) 11:36, 22 February 2020 (UTC)[reply]
@Laurel Lodged: As per your own comment, the barony in question still exists, in some definition, and the first verb in Eliogarty is "is". This is therefore an irrelevant example to this particular discussion, equivalent at best to including cities and towns in their traditional or cultural region. Dahn (talk) 10:03, 23 February 2020 (UTC)[reply]
Comment - CFD Piecemeal Approach A CFD discussion is just as likely to suggest a global approach as this discussion might suggest a case-by-case approach. The area I have concern with is the subcategories of Category:Districts of East Germany, where we categorize literally every populated place that used to be part of the GDR by former region, which doesn't seem remotely defining to me. If I nominated that tree for deletion, it's likely to come up why I'm not nominating the Lithuania examples Renata provided. Does anyone see a difference between those two examples? RevelationDirect (talk) 19:24, 22 February 2020 (UTC)[reply]
I'm not sure why it would come up. It doesn't follow that that what might be appropriate in one situation must be appropriate to another in a completely different geographical, political and historical context because they're both abolished institutions. If you think the German and Lithuanian ones you've both mentioned are equivalent and that they suck, nominating them both is a much better outcome than attempting to make global policy affecting thousands of situations you haven't considered. If you're preferring the few-heads global policy attempt because you think you're going to lose a CfD on the two (I don't know, this is emphatically not my area of knowledge in the world), that should tell you something. The Drover's Wife (talk) 20:23, 22 February 2020 (UTC)[reply]
I'm not sure anyone can name a situation when categorizing by past subnational entity would benefit anyone. Mind you, we're not talking about examples such as "Ancient Greek colonies" or "Former capitals of...", none of which actually refer to a subdivision. We are talking about subdivisions for all purposes defunct, and the type of info one would be able to recover from the article and/or a map. Nobody would benefit from having Places in modern-day Turkey grouped under their former Ottoman vilayets, though the article on both the place and the vilayet should include references to one another, at least once theyre both developed. Dahn (talk) 09:59, 23 February 2020 (UTC)[reply]
...unless someone was trying to find out what happened to the cities that were once within a particular Ottoman vilayets. I'd expect that to be unusual, but I can imagine it happening (at least for larger cities). (That sounds like a great school assignment: "Pick one of the Ottoman vilayets we've been talking about this week, figure out what it's biggest city was, and find out what's happened to that city since then.") WhatamIdoing (talk) 18:59, 25 February 2020 (UTC)[reply]
Whether it's "proper" is going to depend on the context (e.g., age of the students and whether this is meant to be an important assignment or just a few minutes' homework). I do not say that we have to accommodate that reader. I only say that when billions of people have access to Wikipedia, the odds are high that at least one reader would sincerely appreciate whatever seems unimportant to any given editor. WhatamIdoing (talk) 03:02, 2 March 2020 (UTC)[reply]
We should be here to provide every type of encyclopedic information. Some of our tools for doing this are pretty awful at the moment (consider, e.g., the necessity of Category:18th-century British women writers, when it'd be better to have a way to record the simple facts of "18th-century", "British", "women", and "writers" and let the software combine them). The same general type of system could be used for geography: Here is the location, and now give me a list of every relevant Wikipedia article. It'd be clunky to do this with just categories, but I hope that in the future, people will be able to look up any the patch of dirt and see all of its history, from well before being absorbed into the Ottoman empire, through the creation of the province/vilayet system, to the end of the Ottoman empire, and what's happened since then. I think that helping people understand history is consistent with our goals. WhatamIdoing (talk) 06:10, 2 March 2020 (UTC)[reply]
@WhatamIdoing: One can understand the point of having women who lived in the 18th century and practiced a certain trade, and were of a certain nationality, in a standalone category, however: the encyclopedic relevance of having articles placed in defunct administrative divisions is entirely unproven, and unargued -- beyond "it would help hypothetical students perform a hypothetical inane assignment with even more ease". What we do have from the above is your hope that we should all embark on this "patch of dirt" pet project (which, btw, is an immense task you unload on anyone writing articles on such topics, without offering them the option to refuse -- since once this is a standard, everyone will be expected to follow it). Instead of simply dreaming of how interesting it would be to have that goal materialized, you could consider that it has no objective use, while demanding a lot of work from "someone else". Dahn (talk) 06:38, 2 March 2020 (UTC)[reply]
I don't think so. We already put {{coord}} in articles about geographical areas, and Special:Nearby already lets you find articles within a certain distance of your location. Wikivoyage (and other projects) is using Wikidata, Commons, and/or OpenStreetMap to mark territories (e.g., Alpine County#Communities – the region, not just a single point within it). It doesn't seem impossible to take that existing data and using something similar to Special:Nearby to find all the articles that are within that arbitrary shape, rather than all the articles that are within a certain radius of a single point. None of this would require any extra work from editors. WhatamIdoing (talk) 16:30, 2 March 2020 (UTC)[reply]
I'd go with B, with the usual allowance for exceptions in exceptional cases. This is a classic list role. All the problems that afflict using a category for this information would disappear if using a list. A list is also much easier to maintain and add any necessary qualifiers to (as might be needed for example if administrative boundaries shifted during the relevant historical period). As a bonus, a list is also much more likely to attract the attention of contributors with relevant historical expertise. I can see no reason why the approach would be different from one geographical area to the next; the arguments with respect to Central and Eastern Eurperiodically I ope would seem to apply equally well in any other geographical context. -- Visviva (talk) 04:33, 24 February 2020 (UTC)[reply]
C. I'm not sure why this is such a contentious issue. If the town existed in the past as part of a former subdivision, why would it be inappropriate to note that? It actually sounds fairly useful; if I were trying to find out what was the extent of and former municipalities in, such-and-such of a now-defunct province, the categorization of places into such categories seems like a natural way to do that. --Jayron3218:53, 25 February 2020 (UTC)[reply]
@Jayron32: Because it adds a million categories that could be simply replaced by lists in/alongside articles, and because it serves no purpose other than to satisfy dreams of lost glory? Dahn (talk) 14:04, 1 March 2020 (UTC)[reply]
I'm leaning C (no particular rules). I'm not sure that every little village that was once part of the Roman empire should be categorized that way, but Vienna was the capital of multiple empires/nations, and it seems odd to limit its categorization to only the most recent. WhatamIdoing (talk) 19:00, 25 February 2020 (UTC)[reply]
C. Should be treated case-by case basis, and the text must support categorization, with valid refs. In fact it is often important to know who belong where at a particular time, and periodically I am thinking about adding a kind of timeline template to articles about locations. Staszek Lem (talk) 20:07, 25 February 2020 (UTC)[reply]
C (A if we have to): Definitely not B. When talking specifically about Central and Eastern Europe, some places actually have more connection to their former subdivision in terms of historical importance than their current one, so it would be strange not to categorize them by their former subdivision. Ejgreen77 (talk) 11:43, 26 February 2020 (UTC)[reply]
A/C Some of these categorisations (and not only for former east European areas, it goes for the whole of the world) are utterly confusing (at least in my opinion). There are objects that are categorised by current areas where the organisation never existed in that current area (organisations (in the most broad sense of the word) that have been discontinued well before the current area where they would have been if the organisation still existed existed (intentionally confusing sentence)). I had to look, but 1962 Northern Rhodesian general election was once categorised in Category:1962 in Zambia where Northern Rhodesia was renamed in 1964 to Zambia (this one has since been fixed: Wikipedia:Categories_for_discussion/Log/2013_June_18#1935_establishments_in_Zambia; however, there is still Category:Elections in Zambia on the article ...). Within the volatility of the 'countries' in Europe in the past, there are many cases where things happen to an organisation while they are in A, then country changes to B and something else happens, country changes again, to C, and they stop existing, and if they would now still have existed they would now be in D ... Categorisation in these cases should be limited (A) and well thought through (which is basically what should happen now: C). --Dirk BeetstraTC06:24, 2 March 2020 (UTC)[reply]
C I'm with User:Staszek Lem on this one: if referenced text in the article supports the historic categorization (and thus it's presumably appropriate text that does not violate WP:UNDUE), then the cat should stay. But if no referenced article text supports the category, then the categorization is the result of original research and should be removed. UnitedStatesian (talk) 19:16, 6 March 2020 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The behavioural guideline at WP:PE reads:"you are very strongly discouraged from editing affected articles directly;" same as the general WP:COIEDITING guideline. Both are without inline citations/notes, so I am not sure how unlikely it is that anything can be done about it, but I think it should be changed to "you must not edit affected articles directly;" The exemptions are already listed at WP:COIU on the same page. A Partial block from the affected article seems an appropriate remedy to address violations.
Occasionally, paid editors do take the option of not being discouraged by the very strong discouragement, which presents an awkward situation. This would remove all confusion and make handling paid contributions a lot more efficient and straightforward. The community should decide once and for all, whether or not it is okay with paid editors making substantial non-urgent direct edits to affected articles.
Obviously, we'd need an RFC to actually discuss it; first, I wanted to make sure I am not missing something obvious that would end this proposal speedily. Usedtobecool☎️10:48, 9 May 2020 (UTC)[reply]
Clarification One:
The proposal is not to outright prohibit all mainspace contributions from paid editors, just the ones that are both significant && non-urgent. This means they are still allowed to make the edits exempted by WP:COIU, as that section is not being changed, and the new articles would be forced to go through AFC, as creating a new article would be a non-urgent significant mainspace edit. This proposal also leaves the WP:COIEDIT section as is.
So, let me reformulate:
At WP:PE, let's change the bullet-points:
you are very strongly discouraged from editing affected articles directly;
you should put new articles through the Articles for Creation (AfC) process instead of creating them directly;
to:
you must not edit affected articles directly, except as provided by WP:COIU;
you must put new articles through the Articles for Creation (AfC) process instead of creating them directly;
And, optionally, let's add after the last sentence of the last paragraph—You may be technically restricted from editing the affected articles for failing to adhere to these guidelines.
Our policies on paid editing are already a joke, and making them stricter would do nothing but make people feel good about "doing something". After blocking 300 socks in one case at SPI some years ago (yes, 300) I came to the conclusion that simply outlawing paid editing isn't going to work. It is so painfully easy to sockpuppet, to create multiple accounts without getting caught, tightening the rules would only overburden the already overburdened SPI system. Creating 100 sock puppets that are difficult to link to each other does not require much skill, just patience, and a little monetary motivation. "Outlawing" paid editing makes as much sense as the War on Drugs, and has the same effect; makes it more profitable, thus alluring to those looking to make a profit. As I'm sure this will fall on the collective community's deaf ears, I will just leave it at this. Dennis Brown - 2¢19:08, 9 May 2020 (UTC)[reply]
Dennis Brown, I have added a clarification to my original proposal above. I am not proposing outlawing it, but, yes, it seeks to tighten the rules a bit. User-centred policing is less effective, true; and perhaps, UPE is more of a problem than some wikilawyering DPEs, but it's the latter this proposal is concerned with, and I do believe the proposed amendments would clarify the matter a bit without changing the status quo for the good faith DPEs. Regards! Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
I would generally support this. See Special:PermanentLink/946560078#Conflict_of_Interest_Editing for one example of a paid editor arguing that "peer review is not mandatory" because of the current wording. However, it must be clear that removing incorrect information about living persons is always allowed, and that doing so is exempt from both the edit warring policy and the discussed COI guideline bullet point. ~ ToBeFree (talk) 23:40, 9 May 2020 (UTC)[reply]
The last big thing related to COI editing I'm aware of was WP:TOUSL in January, in which the community very resoundingly asked the WMF to take legal action against Status Labs. @Doc James: you mentioned that you planned to share that discussion with the WMF board; if you feel like taking a break from COVID-19 stuff, could you catch us up on how that went? There is also some recent stuff happening at AFC (from Sulfurboy) and the Article Wizard (from me) about paid COI disclosure. I'm not well-informed enough on the policy to make a judgement here, but Dennis Brown's comments above seem reasonable; my general sentiment is that we may be close to reaching the limit of what we can do on-wiki, and that the more productive path may now be through WMF legal. {{u|Sdkb}}talk00:11, 10 May 2020 (UTC)[reply]
Sdkb, I have added a hopefully clearer version of my proposal just below my OP. I don't think there is much else to miss other than the very small few sections linked in the proposal. So, I do believe you can form an informed opinion from reading the proposal alone. Best, Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
Although a lot of paid editing is problematic, paid editing isn't necessarily problematic. I think the current disclosure requirements adequately mitigate the risks, and that our existing policies are as good as they will ever be for handling problematic paid editing (unfortunately, not very effective, but no better than any alternative). --Bsherr (talk) 00:22, 10 May 2020 (UTC)[reply]
I disagree. That is the equivalent of saying "Although a lot of dictatorships are problematic, dictatorships aren't necessarily problematic". We must avoid conflicts of interest even if they may do some good faith edits, just like we must avoid dictatorships even if they may do some good faith policies. --NaBUru38 (talk) 20:15, 10 May 2020 (UTC)[reply]
Bsherr I am not proposing any sweeping changes with any ambition to fix the larger issues. I am seeking to clarify a rather, in my opinion at least, simple point, that nonetheless creates an awkward situation with paid editors who seek to wikilawyer instead of working within general community expectations. I have added a clarification to my original post, hoping it helps. Best, Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
I would endorse a proposal to change wording to unequivocally require paid editors to go through the AfC process when submitting new articles. I'm on the fence as to whether an outright ban on mainspace editing is likely to help much. signed, Rosguilltalk03:37, 10 May 2020 (UTC)[reply]
@Rosguill: I'd support that. Even with NPP, I cannot imagine a situation where I'd be okay with a paid editor creating a page without going through the additional scrutiny of the AfC process. {{u|Sdkb}}talk03:52, 10 May 2020 (UTC)[reply]
I have zero issues with paid editing. I have a lot of issues with paid editors that bring serious WP:COATRACK oder WP:NPOV issues. And since a lot of my time is dedicated to the AfC project, I've become particularly bitter towards said editors. The aforementioned reference by Sdkb was my suggestion that for AfC articles that are created by PEs that there be an easily visible temporary notice on the draft page that would be removed if and when an article is accepted. As it stands now a PE or COI editor only has to disclose on their userpage, and that can go unseen in the typical AfC process. I also have issue with some language in policies, particularly the word "should". See my convo with DESiegel[1]Sulfurboy (talk) 03:41, 10 May 2020 (UTC)[reply]
Sulfurboy it does get tedious having to check the draft talk, then the creator's user and talk pages and then other significant contributors too. Yes, exactly, not all, but I am seeking to change a few of those fuzzy imperatives to definitive ones. Added my proposal Mark II at the end of my OP above. Best, Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
Paid editors who are making apparently good-faith efforts to comply with our disclosure and other policies should not be demonized, and marking their contribution in this way will only increase the pressure to omit disclosure and just hope to avoid being detected. I think it is counterproductive, and I oppose any such marking. I agree that COATRACK and NPOV violations need to be addressed and corrected, but this is true whether the editor is an employee or merely a fan of the subject. Indeed fans may well be more intransigent in POV-pushing. I think that Sulfurboy's proposal if I have understood it correctly, would be a mistake. DES(talk)DESiegel Contribs16:14, 10 May 2020 (UTC)[reply]
DESiegel, in case yours was not wholly a response to Sulfurboy's proposal (indenting), I would like to clarify that I do not seek to change the status quo for good faith paid editors who are already complying with the community expectations. Here, by community, I am largely referring to WP:NPP and WP:AFC folks; it is widely understand in that area that paid editors are supposed to put their articles through AFC and propose substantial non-urgent mainspace edits at the talk page, but because the actual guideline is worded weakly, paid editors who don't want to, don't have to adhere by those expectations. My proposal seeks to eliminate that loophole for wiki-lawyering. I have added a clarification to my original proposal to hopefully make that clearer. Regards! Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
Usedtobecool, It is my viewm that to make an AfC review mandatory, for any editor or group of editors, under any circumstances at all, would be to damage what reputation AfC has left, and to harm its value for its actual purposes, helping inexperienced editors to create valid pages. At hte very lest, paid ewditors should have the option of seeking an individual non-paid editor in good standing to do a one-on-one review of a draft outside the AfC framework as an alternative to going through AfC. Any proposal, which makes AfC mandatory I oppose as strongly as i can. DES(talk)DESiegel Contribs12:19, 12 May 2020 (UTC)[reply]
The only realistic way to remove paid editing and the problems associated therewith is to increase the limit of edits required for new article creation to mainspace. 1500 mainspace edits and 6 months time (above extended) should be used. The number of articles that are being added by ~10 edits and go COI accounts will become zero. There are no VITAL articles that have not been written. So if an article incubates in Draftspace, there is no emergency need to create it at once. It is not like we are lacking the articles on the states of USA that need to be created asap. Most vital work is done and we have prolific article creators. The argument that we need new editors does not work on this scenario as most (99%) of these ~10 edits and go writers just create one article and then leave, so leaving the bar low is not inviting new editors, it is just creating a mess. MistyGraceWhite (talk) 19:33, 10 May 2020 (UTC)[reply]
I can think of no better way to ensure that we cease to get new editors than to adopt the above proposal by MistyGraceWhite. I would strongly oppose it if it were seriously proposed, and I doubt that the Foundation would allow it to be enacted and take effect. It would certainly mean that Edit-a-those, many of which are focused on creating new articles, (and for which I routinely grant confirmed status on the spot) would have to cease or drastically change. DES(talk)DESiegel Contribs00:31, 11 May 2020 (UTC)[reply]
@User:DESiegel I may be wrong. I have not seen any data from editathons. Is there any data which is available for non admin editors to see from editathons. I would like to specifically see the number of new editors with >2000 mainspace edits that have started editing from editathons. I find it odd that an editor MUST get their article onto mainspace at once. What is the rush? Anyhow, can you give out the data from editathons on new editors who have become part of the project? MistyGraceWhite (talk) 00:37, 11 May 2020 (UTC)[reply]
MistyGraceWhite most edit-at-hons, or at least the ones I have attended, (which are linked on my user page) record the list of attendees, and the articles created or changed. But if there is a central compilation of these results I do not know where it is. It would be possible to use individual event reports to compile such data, but rather tedious. So no, I cannot provide such data off-hand. DES(talk)DESiegel Contribs00:45, 11 May 2020 (UTC)[reply]
@User:DESiegel without this data, how can we actually evaluate the impact of editathons on the project? I mean if editathons in their current form are not bringing in any new editors who help wikipedia, then why fight to keep that form intact? MistyGraceWhite (talk) 00:50, 11 May 2020 (UTC)[reply]
That is a different question, MistyGraceWhite, and one more properly addressed to those who organize such events. It is my belief, without having attempted to analyze data, that such events are good for our overall reputation and a better understanding of what Wikipedia is, whether they bring in large numbers of new editors or not. I also suspect that the editors they do bring in are valuable, but again i can't prove this. DES(talk)DESiegel Contribs02:21, 11 May 2020 (UTC)[reply]
I think the question misses the boat in another way as well, by focusing on new-editor-at-editathon retention. Even if zero of those newbie editors ever return, let alone reach several K mainspace edits (lack of gain for our long-term editor pool), and even if there is no reputational aspect (lack of gain for Wikipedia site and movement), edit-a-thons often create viable new articles (gain for our readers). DMacks (talk) 04:20, 11 May 2020 (UTC)[reply]
I disagree in the strongest possible terms about Misty's suggestion. It seems a real case of cutting off your nose to spite your face. Given we'd have to create a new userright, I can't imagine the WMF would be particularly obliging. I wouldn't even support EC. If there was some intermediate option I might consider that, but the community has indicated a firm opposition to expansion of protection/permission levels Nosebagbear (talk) 08:24, 11 May 2020 (UTC)[reply]
I agree completely with every comment DESiegel has made in this discussion. The problem is not paid editing or paid editors, the problem is breaches of the content policies (NPOV, etc). The two groups "editors who are paid to edit" and "editors who make harmful changes to the encyclopaedia" are overlapping sets, one is not a subset of the other (and the same is true for more specific types of harmful edit). We should never prohibit any editor from reverting obvious vandalism, correcting obvious BLP errors, etc. Thryduulf (talk) 10:11, 11 May 2020 (UTC)[reply]
@Thryduulf: I have added a clarification to my original post above. Sorry for the misunderstanding; the proposal doesn't seek to outright prohibit all mainspace paid contributions, just the non-urgent substantial ones. Best, Usedtobecool☎️08:59, 12 May 2020 (UTC)[reply]
@Usedtobecool: I still oppose that. While it is right to discourage paid editors from making substantial contributions to mainspace, outright prohibiting it in all circumstances will not benefit the encyclopaedia. Examples include implementing a discussed consensus, clearly non-controversial changes such as accessibility improvements, reference fixing, etc. that might be regarded as "substantial". Paid editing is not synonymous with POV editing. Thryduulf (talk) 09:56, 12 May 2020 (UTC)[reply]
I would support a wording change from "you are very strongly discouraged from editing affected articles directly" to "you must not edit affected articles directly, except when your edits are unambiguously uncontroversial", which would draw a far clearer distinction between acceptable and unacceptable edits (and there is always WP:IAR for edge cases).
I am strongly opposed to Misty's proposal to raise the editing threshold for article creation, which is far too broad and would cause too much collateral damage to good-faith article creation. However, if it were modified to solely focus on a narrow subject area which proves overwhelmingly likely to contain poor article creations by new editors with a COI, I might support it. For example, if it could be shown that 95 of every 100 articles by new editors on, for instance, live businesses that began operating under three years ago were excessively promotional, then a requirement for AfC review before an article on a topic meeting the criteria could appear in mainspace would make sense. – Teratix₵12:36, 11 May 2020 (UTC)[reply]
Teratix that is exactly what I intended to convey. I am not seeking any changes to WP:COIU. I have added a clarified version of my proposal to the bottom of my original post. Regards! Usedtobecool☎️08:50, 12 May 2020 (UTC)[reply]
The problem with the modified wording is that it gives the impression WP:COIU is a comprehensive list of scenarios where paid editing in mainspace is acceptable, which, as Thryduulf's above examples of acceptable paid editing in mainspace not technically covered by COIU show, is not actually the case. – Teratix₵01:42, 13 May 2020 (UTC)[reply]
I would support both proposals under "Clarification One". The current phrases, "you are very strongly discouraged" and "you should", are functionally equivalent to "you may" when it comes to enforcement. The proposed phrases, "you must not" and "you must", actually have teeth. — Newslingertalk11:27, 12 May 2020 (UTC)[reply]
The current wording is correct: there are circumstances in which they may make those edits, because doing so improves the encyclopaedia. Completely prohibiting all such editing would be to the strong detriment of the project. Thryduulf (talk) 13:23, 12 May 2020 (UTC)[reply]
I disagree with respect to the first bullet point ("...editing articles directly") and strongly disagree with respect to the second bullet point ("...Articles for Creation (AfC) process instead of creating them directly"). {{Request edit}} is available for paid editors who wish to make changes to articles, although Teratix's suggestion ("except when your edits are unambiguously uncontroversial") is reasonable and resolves most to all of the circumstances you're referring to. I see no reason that paid editors should create articles directly in mainspace, instead of having them vetted through AfC. The AfC process works very well for maintaining article quality in the presence of a conflict of interest. — Newslingertalk02:37, 13 May 2020 (UTC)[reply]
Once again though you seem to be conflating "paid editing" and "POV editing". They are not the same thing - while paid editors have a clear COI regarding the person or organisation who paid them, that doesn't necessarily translate to POV editing or that the conflict extends to every edit (e.g. when a paid editor reverts obvious vandalism their and our interests align perfectly). Additionally, I'm not sure where you get the idea that I'm opposed to AfC, given that I've never mentioned it? My opposition is to blanket prohibitions that will not only not help the encyclopaedia and actually hinder it by making it more likely that paid editing will be undisclosed. Deal with the problem content, not the method used to create it. Thryduulf (talk) 11:13, 14 May 2020 (UTC)[reply]
The larger point, and one that seemed to be missing, is the fact that the stricter you make the rules, the more you push paid editing into the shadows, where you have NO control. If anything, it makes it more profitable. No rule or rule change will make paid editing go away, there is simply too much financial interest in it. Again, making rules so we can say "so at least we tried" tend to backfire, badly, and in this case will overload SPI even more than it is. Dennis Brown - 2¢14:27, 12 May 2020 (UTC)[reply]
We can accept that unwelcome practices such as vandalism and sockpuppetry will never be totally suppressed while still endorsing and enforcing policies and guidelines which prohibit such practices. Why not for unacceptable paid editing? – Teratix₵01:42, 13 May 2020 (UTC)[reply]
We already have policies and guidelines prohibiting unacceptable paid editing, nobody is proposing to remove them. The problems we have currently are from those that do not follow the existing rules so making it harder for those to be followed will just mean fewer people follow them while doing nothing about those that already don't - i.e. it would make things worse. Thryduulf (talk) 10:15, 13 May 2020 (UTC)[reply]
Upon reflecting on this more, I think it's chasing after the wrong problem. COI-related difficulties generally come from the editors that don't self-disclose, so changing the rules for the ones that do doesn't help (and I'm not sure paid editors are significantly worse than editors who are self-promoting or writing about their friends). signed, Rosguilltalk03:27, 13 May 2020 (UTC)[reply]
@Rosguill and Dennis Brown: We seem to be on similar wavelengths. I'll highlight in case it got buried that Doc James replied to me above that WMF does not seem to have delivered any response to the 100+ editors who asked nearly unanimously for WMF Legal to take action against Status Labs back in January. How about we head over to our new WMF pump page and start talking about that? {{u|Sdkb}}talk10:48, 17 May 2020 (UTC)[reply]
I strongly oppose MaisyGraceWhite's proposal. First, it will immediately drive new editor retention down to near zero. Second, it goes against the basic principles of Wikipedia as an encyclopedia that anyone can edit, which includes creating new pages. Third, coverage on Wikipedia is still frightfully thin or even nonexistent especially in topics affected by systemic bias. CJK09 (talk) 15:40, 24 May 2020 (UTC)[reply]
I think MistyGraceWhite is on to something, although the parameters are certainly negotiable. We have a substantial problem of sockpuppets and SPAs, and we have basically unlimited flexibility to tweak the degree to which new editors may be required to demonstrate their willingness and ability to contribute to the project before receiving specific tools like the ability to create pages in mainspace and pagemover rights. BD2412T14:21, 27 May 2020 (UTC)[reply]
The "must" proposal I very strongly support. If we just say "shouldn't", every single paid editor will think they're the exception, since, well, they're being paid to think so. "Must not" doesn't have any such wiggle room, while still leaving space for things like reverting obvious vandalism. No one's going to complain if a paid editor reverts the addition of random strings of profanity anyway. I also see the merit in upping the bar for new article creation. Autoconfirmed is a pretty low bar to clear for someone being paid to do so—make ten minor uncontroversial edits and set up a calendar reminder four days later. (This is not hypothetical; when I delete spam articles, this is often clearly exactly what they did.) Raising the bar to even EC would dramatically raise the amount of effort to create socks and sleepers, and if nothing else that would mean the paid editors will have to charge more, operate slower, and lose more when a sock is caught, which may help to discourage the practice. Creating an appropriate mainspace article is hard; the number of editors who could do so after ten edits is probably quite small. SeraphimbladeTalk to me21:00, 1 June 2020 (UTC)[reply]
None of it works. Sorry for being such a skeptic, but it's a joke if you think any of these proposals will stop paid editing. See the following: [2] and [3] - and weep. And that's only one site among many. They even predict that volunteers will be a thing of the past and that paid editing will win out. Money talks. How do you silence it? There's only one way and that's registration with validation. Sorry but that's it in a nutshell, and I haven't seen anything that convinces me otherwise, especially not more failed bureaucracy. AtsmeTalk📧22:36, 1 June 2020 (UTC)[reply]
Honestly, I am getting a lot of Nirvana fallacy vibes from the comments. I didn't expect to solve all our conduct and content issues with regard to promotion with this proposal. I did not identify this as the worst problem we have, or the proposed as the best solution we can have. I identified a very particular problem: A guideline that limits contributions from paid editors in good faith and rewards the ones who have no respect for community norms. Despite what the guidelines say, it is the norm that paid editors are expected to use edit requests for every edit that is not urgent enough or trivial enough to forego peer review from an independent editor. With the guidelines we have, editors who want to respect these norms wait for weeks/months waiting for a response to their edit requests while those who don't, can continue to edit the articles directly, and since AFD is not cleanup, there is nothing preventing them from making the article whatever they wish to make it after sufficiently exhausting any good faith neutral editors that may be watching into quitting. Which one do you think would Wikipedia be better off with, and is that the same one that will win out in the market competition? We should stop acting like a bully, showing our muscles to good faith paid editors and resigning to those that won't be tempered by our very strong suggestions. If one believes paid editing is the future, or an essential evil, it is all the more reason to level the playing field in favour of more-collegial, more cooperative ones. My proposal may not be perfect, it may not be the most urgent, it may not be the superlative of anything, but would we be better off, if marginally? Now, I do take the point about whether those kind of editors would be more likely to prefer UPE over compliance, and if they are, is handling them then likely to be harder or easier than with the status quo? Honestly, that was the discussion I thought would be the most relevant on whether we go forward with this. Usedtobecool☎️21:09, 7 June 2020 (UTC)[reply]
I agree with you when you say we cannot stop paid editing and that we should reward those who play by the rules and not those who don't. The problem is that this proposal will not do that - indeed it will make it harder for those who do play by the rules (in some cases, it will make little to no difference in others) and wont make a jot of difference to those who don't. The way we solve the problems with paid editors is to focus on non-neutral editing by all editors (paid and unpaid, disclosed and undisclosed) rather than spend our energy on ineffectual measures directed at editors. Thryduulf (talk) 18:01, 8 June 2020 (UTC)[reply]
I'm in favour of you must put new articles through the Articles for Creation (AfC) process instead of creating them directly, but not inclined to back banning paid editors editing in mainspace altogether. Creating articles is an incredibly difficult process, and to be honest, I've never seen any paid editor successfully create anything approaching an encyclopedically-neutral article in mainspace - the AfC process offers not only review, but also a process by which advice can be given to those who are genuinely interested in positively contributing. Those who are disinclined to follow our rules will make articles in mainspace anyway that will be speedy deleted, and those who follow them won't mind having to wait a little while for an AfC reviewer. On the other hand, there are cases where paid editors genuinely contribute to encyclopedia articles that already exist in mainspace, and I don't think we should discourage those - especially given that that might perversely encourage people to hide their paid editing, in contravention of the ToU. Naypta ☺ | ✉ talk page | 18:20, 8 June 2020 (UTC)[reply]
I think the whole idea of paid editing has completely subverted Wikipedia. It has turned it, from what should been the vision and idea of an egalitarian project into a two tier project that enforces inequality at every level and delivered 10000's of article that are completely COI and can't be trusted as fact, but are nevertheless defended to the limit by industry. On one side we have beautiful paid articles, mostly non-entities and companies with tenuous notability to the volunteer side, with reams of stubs, from people and events and history, that have utterly changed the world, but will probably never be expanded. One side is increasing in size, while the other is decreasing. Paid editing is a booming industry and they are getting larger, while we are getting smaller. That is the future. I don't think the POV editing and paid editing are synonymous. Occasionally you come across paid articles that have not been looked at in months and they expand them right out, effectively with no barriers. I've seen 60k articles on 12-15 person companies. We are essentially tinkering around the edges, when a wholesale change is needed. More bureaucracy is not needed, although the idea of putting it all through Afc is ideal. I think we need registration. We can track what is being done, otherwise the future is going to be very grim. I think having registration will not drive drive retention down, or the flow of newbs. The number that is coming through the door, is slowly dwindling anyway, the honeymoon period is long over, while the paid crowd are increasing every day. They have made the conceptual move, why don't we. When you look at NPP now, only about one, out of every 6-8 articles are not paid. It is quite hard to find them. Sure, there is a large cadre of folk who still believe in the project and work to improve Wikipedia, but they are in minority now. While industry has power and money to do what it wants, that is their will, we are losing power, while they are gaining it. scope_creepTalk08:20, 18 June 2020 (UTC)[reply]
It has 7-8 years now since wee started talking about this, and nothing has been done. The idea of WP:PAID is a failure as it is largely ignored. Its a catch-22 or catchata-22 situation, round and round in a circle we go, with no end in sight. We should be looking for innovative solutions. They are always going to be here, so we should use them: e.g. perhaps to support and maintain portals (with educational input) and all policy applied. Or if you want that article, then expand 1 stub first, or we have difficult article, write it up for us. We could have the "1 for 1" rule. 1 new article in an area that is not serviced properly for one paid article, or 1 expanded article for 1 paid article. There is so much good work that could be done with such a rule. It has a number of advantages, and deals in realpolitik, instead of the jaundiced condition we operate in now. It's sightly socialist and mercenary, in its approach, but industry has access to resources we can only dream about and would even the balance, somewhat. We could engender the idea of a company Wikipedian, who is paid to write complex articles and could be made competitive, with quality as the metric. They've longed talked having a historian in places like Microsoft, to document the company's path. Almost overnight the idea of stubs would disappear. There is articles I known of artiles that don't get written, because they are too complex. One of them is Forfeited Estates Commission. In Scotland, after the rebellion, the commission came in and sold a bunch of stuff, estates mostly. There is 400 documents about it, apparently, and very little on it. Hence there is no article. Huge articles like that, that could be made into a series, and happily done by a paid editor. scope_creepTalk09:20, 18 June 2020 (UTC)[reply]
If paid editors can't edit then the concept becomes self-contradictory. As Wikipedia is billed as the encyclopedia that anyone can edit, this would not be sensible. Consider some recent examples:
I myself had another article at DYK recently: fever hospital. This used an image from the Wellcome Library which has paid staff working on the release of such images and generally supporting Wikipedia. Following my activity, I noticed mainstream media picking up my crop of the image and using it in their pandemic coverage.
Another current discussion is about the Internet Archive. It appears that this is not content with snapshots of the entire internet but now wants all the world's books too. The issue is a bot which is being run by a paid editor and there seem to be other paid editors active in the discussion too. In this case, I tend to oppose the activity but the matter is debatable.
These examples show that the paid editor issue is not just a matter of the puff pieces which come to the attention of the NPP. Crude and draconian policies are not appropriate. Andrew🐉(talk) 09:33, 18 June 2020 (UTC)[reply]
They are all academics and/or in the public sector, which is by far the minority, and welded to the belief of the common good. Certainly in the UK, the public ethic is based on that belief. The value to paid editors is only google rankings and that wouldn't change. We don't have control of Google, but their mantra is quality of content, and their doubling down on it for the last few years, so that won't change. I just wish we could try something practical instead of the roundabout conversations that lead nowhere, while they are singing and celebrating, at our expense. The thing that worries me most is that we have less and less people looking at them, as time goes by. scope_creepTalk10:51, 18 June 2020 (UTC)[reply]
We will never make progress on this issue so long as we continue to allow the good-hand sockpuppets of paid editors to stymie every effort to control them. The pig does not vote to abolish the trough. —Cryptic10:52, 18 June 2020 (UTC)[reply]
The problem with these progressivly more restrictive versions of the paid editing policy is that it is making no impact on the amount of paid editing - only on the amount of visible paid editing. There continues to be very high demand for paid edits - a quick scan of Freelancer.com gives 348 open jobs related to Wikipedia, and while a minority of those are not related to direct editing, the vast majority of them are. Mostly we have three categories of paid editors: people working for a company who could be better seen as COI editors; people who are knowledgeable about Wikipedia's policies and try to engage in disclosed paid editing; and those who are hired by a company specifically to edit Wikipedia but engage in undisclosed paid editing. The first group is normally unaware of our policies and generally can be convinced to accept the "should not" as equivalent to "must not". The second group do care, but there are very few of them. The third group doesn't give a damn about our polices, are very hard to detect, and will ignore our policies no matter what we change them to. The problem is that as we continue to make it harder and harder for the second group to exist, the demand for paid editing doesn't decrease - it simply moves to the third group. Insisting on AFC is a decent move, but banning edits to mainspace will significantly increase the demand for undisclosed paid editors. We've already made disclosed paid editing almost impossible; this would make it that much harder. For me, the discussion should be around how do we target the third group - what techniques can we use to identify socks, close down undisclosed paid editors, and either reduce demand or make paid editing visible enough that we can effectively manage the problems that come from it. - Bilby (talk) 10:39, 18 June 2020 (UTC)[reply]
My main concern about banning paid editing outright is that it will encourage those editors to try to hide their identification, making potential CoI issues harder to detect. As for increasing the requirement to create a page to 1500 edits, that seems a bit excessive - extended confirmation should be enough. Elliot321 (talk | contribs) 05:11, 19 June 2020 (UTC)[reply]
I strongly disagree with this proposal for a variety of reasons and I can easily wax on about how we need to incentivize corporations to use legitimate means to contribute to Wikipedia or the stellar contributions by legitimate paid editors such as User:CorporateM. However, as written this proposal would effectively ban all Wikipedians in residence as well as WMF employees acting in their official capacity (what if a Wiki Ed coordinator needs to edit an assignment that's been mainspaced?) from editing articles in a fashion apart from the WP:COIU exceptions list. This proposal is currently untenable without significant amendments to clarify the status of WIRs and WMF employees with respect to this new absolute policy. Chess(talk) (please use {{ping|Chess}} on reply) 08:05, 19 June 2020 (UTC)[reply]
The lede of WP:Notability needs work. It's widely accepted that "notability" is not a measure of a subject's worthiness, but rather a direct consequence of the nature of Wikipedia. It is simply impossible to write a high-quality Wikipedia article on a subject that isn't "notable." Yet the first line of the lede says:
On Wikipedia, notability is a test used by editors to decide whether a given topic warrants its own article.
The phrasing here is misleading, and may contribute to the myth that Wikipedia determines the suitability of article topics on whether a subject is deserving of attention or respect. I suggest something more along the lines of:
On Wikipedia, notability is the term used by editors for whether it's possible to write an encyclopedia article about a given topic.
I am not proposing this specific phrasing, but rather offering it as a starting point for discussion.
The lede goes on to state that "Article and list topics must be notable, or "worthy of notice"." I suggest we trim this back to just "Article and list topics must be notable," as again, notability is not a measure of worthiness. This line, as it stands, only further conflates the different concepts of notability on Wikipedia and elsewhere.
Later, it says: "A topic is presumed to merit an article if..." Again, the use of the term "merit" is misleading, as we're not judging goodness or worthiness. It might be better to say something like: "It is possible to write an article on a topic if..."
I think we can and should make this much clearer for those who aren't already familiar with the concept of notability on Wikipedia.
On the second point, there has been long discussions over and over that boil down to the fact that "notability" on Wikipedia is not the same as how it is usually treated in the rest of the world, and ways to fix that have been proposed multiple times but no good solution that gets away from the intent of the guideline has been found.That reason comes back to the first point and the issue of merit: the guideline is about allowing for standalone articles or not for a topic, and thus showing via quality of sourcing or the types of information in that sourcing that we would presume that we can build out that article more. That creates the presumption of notability that allows for the article to be kept or created, and allows for it to be challenged later should it be found impossible to actually build out beyond a few sentences (that is, after one proves no more sources are possible via WP:BEFORE). Notability doesn't tell us if it is possible that an encyclopedic article can be created but if we presume one can be. --Masem (t) 23:47, 5 June 2020 (UTC)[reply]
(And now that I see where this comment came from, in response to issues about how we cover underrepresented topics, that has been also discussed at length multiple multiple times at WP:N and most solutions would require us to weaken the application of WP:V to meet that, which is not going to happen). --Masem (t) 23:49, 5 June 2020 (UTC)[reply]
What if, instead of "possible," we said "likely to be possible," to make it clear that notability is a presumption and not a certainty? Also, I scanned the archives of WT:Notability and didn't see anyone making these particular points about the lede, though I didn't make an exhaustive search and might have missed it.
I do not suggest or support changing the notability guidelines to cover underrepresented topics, but rather to clarify the notability guideline with the hope of reducing the number of spurious claims of discrimination. Mysterious Whisper (talk) 00:04, 6 June 2020 (UTC)[reply]
Yes, and I acknowledged that in my response, didn't I? I'm responding to the concerns over the use of the word "possible" in my suggested phrasing. Mysterious Whisper (talk) 00:19, 6 June 2020 (UTC)[reply]
[edit conflict] The point of "warrants" is that it's not about whether we can have an article about the subject (we can, and do, have many horribly sourced and unverifiable articles about topics that do not meet our notability tests), but whether we should have an article. Your rewording is bad because it ignores that distinction, and talks about things that are clearly possible (writing bad articles about non-notable subjects) as if they were impossible. —David Eppstein (talk) 23:49, 5 June 2020 (UTC)[reply]
The original draft would have said "...possible to write a high-quality encyclopedia article..." I omitted "high-quality" for brevity and because any "article" that isn't based on reliable sources isn't actually an "article" suitable for inclusion in the encyclopedia. Mysterious Whisper (talk) 00:04, 6 June 2020 (UTC)[reply]
It's like If a tree falls in a forest, except the question is "if there is no reliable sourcing, is a person not notable?" It is well established that some people such as African Americans or female academics were turned down for articles because there was not enough sourcing, but that does not mean that they are non-notable figures. There needs to be an acceptance that WP:GNG can lead to this type of problem, which is a form of systemic bias.--♦IanMacM♦(talk to me)04:57, 6 June 2020 (UTC)[reply]
I'm actually becoming more convinced WP:GNG is not the problem, but rather how WP:GNG is interpreted. If a topic fails WP:GNG, we cannot have an article on them. (Now, if you're specifically referring to academics, that's a separate problem, since academia apparently doesn't need to pass WP:GNG.) But often users downweight sources they don't understand or from countries they're not familiar with even though those sources perfectly contribute to WP:GNG. SportingFlyerT·C05:43, 6 June 2020 (UTC)[reply]
Your last sentence there is an important concern. I've run into far too many AfD nominations for subjects relating to non-English-speaking countries where it's clear that the nominator didn't even look in non-English sources. For many languages, with the amount that Google Translate has improved, there's no excuse for not doing even a quick search in the relevant language(s) to see what appears. I think it's worth adding a note to the GNG that "significant coverage in multiple reliable sources that are independent of the topic" encompasses all reliable sources, not just English-language sources for which the full text does not appear online. −−−CactusJack 🌵20:27, 7 June 2020 (UTC)[reply]
@CactusJack: This is already in WP:GNG - Sources do not have to be available online or written in English. I would support it being added to the {{nutshell}} at the top of WP:N, as well as having Likewise, search for native-language sources if the subject has a name in a non-Latin alphabet (such as Japanese or Greek), which is often in the lede clarified at WP:BEFORE to include any situation in which there is a credible indication that there may be non-English sources available - not just those in non-Latin alphabets. Naypta ☺ | ✉ talk page | 18:26, 8 June 2020 (UTC)[reply]
There is a huge problem, and it's not limited to articles about people. If one looks for sources on say, public schools in the united states, for instance, there are usually a plethora of sources easily found with a google search - so long as the school is one that educates white people. The newspapers of the segregated time often didn't acknowledge the existence of schools for African-American, Latino or Native Americans. Yet even middle-school sports often get regular mentions. The sources that exist are often difficult to access because they are rarely online. Editors have - until recently - been willing to give a free pass to high schools, but in almost every case, when integration occurred, the shitty facilities provided to African-Americans were downgraded to middle schools, which often were very quickly replaced. Their history is not recorded at all in the same way as white schools - it is simple to go online and find who won championships in every sport throughout history in most states....but the records of the black schools are not included, and are not available. Are these schools not as notable because their mention was suppressed by racists? By the current standards, they mostly are, making WP complicit.Jacona (talk) 13:19, 6 June 2020 (UTC)[reply]
Which is why routine coverage is something that we do not consider as part of notability. Routine coverage doesn't tells us anything why a topic is important from the rest of the world's standpoint. We want coverage about the topic, not that name-drop the topic or are linked to the topic, so this would not actually happen. When it comes to something like schools, one thing we do have is the fact that we do consider all government recognized towns as worthwhile for an article as part of a function as a gazetteer - this is sorta above and beyond our notability principle - but this means any public schools should have at least one mention within these town articles as part of the education system there. Some schools may have unique standalone notability beyond that. --Masem (t) 13:53, 6 June 2020 (UTC)[reply]
My feeling is that this idea of "routine" coverage exists purely to promote bias. It allows AfD commenters to dismiss any topic they deem unworthy of an article as non-notable, because the coverage is routine. If the problem is that some forms of coverage that would otherwise be considered reliable independent and in-depth are too easy for insignificant topics to generate, the solution is not to magically pretend that it doesn't count because it's easy, the solution is to consider that maybe depth of coverage is not the right way to measure significance and use SNGs that can be more accurately tuned to the real markers of success in some area. —David Eppstein (talk) 18:47, 7 June 2020 (UTC)[reply]
Exclusion of topics that are only covered by routine coverage is essential to the fundamental part of WP:NOT - We are not an indiscriminate collection of information. There is so much data out there today that we have to have a better standard, in certain fields and areas, that just name dropping and common coverage (such as high school sports) that we can expect. Defining what routine coverage is in some areas to make sure that AFDs do not incorrectly label sources as routine, or vice versa, can be done. This is probably where NCORP is the example to be clear of what are routine sources that are insufficient to demonstrate notability for organizations and businesses as to avoid self-promotion. --Masem (t) 19:01, 7 June 2020 (UTC)[reply]
I'm not suggesting that routine coverage should be allowed for notability. I'm suggesting that our determination of what is routine is an obvious place for editors' biases to creep in, and that they do creep in. More, the idea of "routine" coverage is basically a patch for the fact that some unimportant topics get far more press coverage than some important ones. And the fact that we can't use press coverage without patching it in this way suggests that measuring importance by press coverage is fundamentally wrongheaded. In the old days, the philosophy was that anything that *could* be covered, *should* be covered. That has (rightly) stopped being the case, and the way we made it stop being the case was to arbitrarily declare that some coverage that obviously *could* be used as the basis of an article *should* not be used because it was too routine to imply any significance to the topic. But my feeling is that this leads to arbitrariness and bias. If we want to judge significance, beyond the mere existence of sources, we should do so explicitly, by clear standards that fit the topic at hand, not by editors' biases that too often amount to "I don't think that a high school female athlete (or whatever similar subject) could ever be notable so I'm going to declare that all the coverage we actually do have is routine", where "routine" is really a shorthand for WP:IDONTLIKEIT rather than any meaningful and reproducible classification of sources. —David Eppstein (talk) 19:54, 7 June 2020 (UTC)[reply]
Maybe we shouldn't call it "routine coverage". After all, coverage of subjects such as elections, sporting contests, and cultural events is "routine" to the point that some fans can predict which tropes are likely to be invoked in that coverage before the events happen, but we don't reject that. Maybe we need a different term for this – something that differentiates between the routine-and-expected news story about spring gardening and the routine-and-expected news story about the spring elections. WhatamIdoing (talk) 21:16, 7 June 2020 (UTC)[reply]
A minor correction: In fact we do reject coverage of elections, that goes in-depth into the candidate's background and opinions, as "routine" when the candidate loses, but not when they win. I think that's a clear tell that our definition of routine is really an excuse to judge the significance of a topic rather than something inherent in the sources themselves. —David Eppstein (talk) 22:24, 7 June 2020 (UTC)[reply]
I like this idea. In terms of wording, I'd suggest that it say …is possible to write a neutral and verifiable encyclopedia article…". I could turn almost any CEO/scientist/activist's webpage into something that looks like an encyclopedic biography, but writing an article WP:based upon nothing except what the subject says about itself isn't neutral. WhatamIdoing (talk) 19:36, 6 June 2020 (UTC)[reply]
Biographies based solely on personal web pages, LinkedIn profiles etc have been turned down in the past because of WP:PRIMARY and WP:COPYVIO. The trouble is that journalists are not interested in writing about some people unless they die, win the Nobel Prize or commit a mass murder. So there are inevitably gaps in the sourcing as defined by WP:GNG.--♦IanMacM♦(talk to me)08:05, 7 June 2020 (UTC)[reply]
You're conflating notability and verifiability, which are two separate things. "Notability" is a measure of a subject's worthiness for Wikipedia. Verifiability is whether it's possible to write an article. Chess(talk) (please use {{ping|Chess}} on reply) 08:08, 19 June 2020 (UTC)[reply]
Get help writing your RFC questions
The "regulars" at WT:RFC have been talking about some common problems we see in RFCs, and we are going to try a bit of an experiment this month. This month, you can ask for help with writing your RFC question at WT:RFC.
This is not required, but it may be helpful. We are hoping that by providing a little experienced advice, you will be more likely to end up with a clear result. I particularly recommend this when:
a group of editors is already in conflict or someone is saying that a proposed RFC question isn't 'neutral' enough,
you're starting a "major" RFC (e.g., significant changes to a policy or to a contentious article), and
you want to hold a vote on what the wording of a sentence should be.
User:WhatamIdoing - Good idea for a service. I think it can be a next step for some of the disputes at DRN. (Right now at DRN there is an IP who seems to be arguing against using an RFC because that would be an escalation from DRN, but it isn't clear what the IP wants, other than to write at length.) (Maybe the Dispute Resolution policies need to be revised to clarify that using an RFC is not an escalation. Going to WP:ANI is an escalation.) Thank you. Robert McClenon (talk) 18:23, 7 June 2020 (UTC)[reply]
Wikipedia:Not compatible with a collaborative project
Hi all, a bit of shameless plugging here, but I just wrote Wikipedia:Not compatible with a collaborative project/WP:NOTCOMPATIBLE, which I was hoping would be a summary of something that is sometimes said at ANI/ArbCom/block declines, but hasn't really been summarized in one place. Something like WP:NOTHERE, but for people who are here in good faith but just aren't a fit for Wikipedia and unfortunately end up blocked. Anyway, thoughts on it either on the (as of yet non-existent) talk page, my talk page, or here would be welcome. I normally don't really try to promote essays, but I thought this one might be relevant to more than just my talk page stalkers (and apologies for having the ego to be willing to think that ) TonyBallioni (talk) 02:07, 9 June 2020 (UTC)[reply]
I suggest renaming the essay, because with the namespace and colon, it reads like it is announcing that Wikipedia is not compatible with a collaborative project. :-) Maybe something like "Not everyone is compatible with a collaborative project". isaacl (talk) 03:12, 9 June 2020 (UTC)[reply]
I personally think the current name is fine (obviously, I wrote it ) but I’m not necessarily opposed to a different name, but I’d oppose both of those. Don’t like the first on stylistic grounds and the essay is more meant to focus on individuals who act this way rather than specific behaviour so I don’t think the second fits. TonyBallioni (talk) 06:08, 12 June 2020 (UTC)[reply]
Nutshell added. Not convinced on a need to rename: our other essays and supplements here follow short declarative phrases (e.g. “Not here to build an encyclopedia”, “competence is required”) and the suggestions seem pretty clunky to me. It’s in project space though, so up to everyone. I just don’t see the colon as being any more confusing than any other page like this, and don’t really like the alternatives above as they’re pretty clunky in my view. TonyBallioni (talk) 06:03, 12 June 2020 (UTC)[reply]
I am basically the only user currently interested in systematically improving articles on Russian localities and administrative divisions. We used to have a set of infoboxes customized to our needs. A couple of years ago, all these infoboxes were taken to TfD and, against my objections, converted into wrappers. (All other users interested in Russia retired by the time). The group of users who spearheaded this was not really interested in the opinion of people actually using the template, they just outvoted everybody. We are now going into some technical details which I do not understand very well, but Template:Infobox Russian inhabited locality now automatically calls some data from Wikidata. Now, today I discovered (at Taldom and Ozyory, Moscow Oblast) that these data is unsourced and likely wrong. For example, it shows the area, but it does not specify the area of what it shows, and Wikidata sources it to the Russian Wikipedia (which is not an acceptable reference). For Taldom, I went to Wikidata and just removed the unsourced data [4]. (I left of for Ozyory to show how the template works). However, I can not do it for every Wikidata item, as this data is added by bot, and I can not compete with bots in several thousand items. It looks like my options are:
Convince Wikidata that unsourced and badly sourced data must be deleted (not even deprecated) and never readded — would be a great solution but unlikely plausible in the current situation;
Keep removing bad data from Wikidata — sorry, I am not a bot, and I just can not do work of this volume, even assuming nobody accuses me in vandalism there;
Revert the template to the previous version - this would amount to ignoring consensus;
Remove infoboxes - they would be probably routinely re-added even if I add a notice in the article;
Leave it as it is - this is probably what I am going to do, but we now have several thousand articles which are not compliant with WP:V and some of them likely contain false info (I only noticed the problem because there was a reform of administrative division, and Taldom and Ozyory were vastly expanded, probably by a factor of hundred, in the area, and I have no idea what the current numbers mean);
Find a wizard who can modify the template to block import of certain fields from Wikidata - this would probably solved the immediate issue but the problem of course is people who voted for deletion of the template had no interest in fixing it, and I do not know how to fix it myself.
There are some wizards around who can help pretty easily to fix the issue of unverified/badly sourced data coming from Wikidata. Leave a comment at WT:Wikidata that you're looking for help. --Izno (talk) 17:33, 9 June 2020 (UTC)[reply]
This seems to boil down to the problem that Wikidata does not have the same sourcing requirements as the English Wikipedia. We can fix the one particular issue, but the general issue is much less tractable. Phil Bridger (talk) 18:04, 9 June 2020 (UTC)[reply]
Yes, this is a long-standing problem, and I once almost lost my admin flag over it. I personally would be happy with the decision that Wikipedia only may import statements with also simultaneously importing their source, and statements sourced to Wikipedia in any language may not be imported (assuming this is technically feasible). However, unfortunately, in every Wikidata discussion users with extreme points of view dominate, and they are not interested in finding any middle ground. In the past, I used to to participate in such discussions and to explain how Wikidata works, but I stopped doing this a long time ago. It is useless and does not lead to any good effect.--Ymblanter (talk) 18:18, 9 June 2020 (UTC)[reply]
But this is exactly what we had until a couple of years ago. Old {{Infobox Russian inhabited locality}} was only importing the commons category, which I have no issues with. It was taken to TfD and deleted. None of the people who voted delete were active in articles on the topic.--Ymblanter (talk) 18:02, 10 June 2020 (UTC)[reply]
Deletion discussions remain one of the most hotly controversial parts of the project, but the bar for participation is lower than most other controversial parts of the project.
Bad faith nominations are a common form of harassment or POV-pushing, and while such nominations are rarely successful, there are no protections in place to prevent it from taking a toll on the victim (in cases of harassment) or taking a large amount of volunteer time (for harassment or for POV-pushing). Starting a deletion nominated currently requires autoconfirmed status (4 days + 10 edits).
Once the nomination is started, it's common for people associated with the subject to use social media channels to influence the discussion (whether to support or oppose deletion). New users who sign up just to advocate a position in a deletion discussion rarely take the time to familiarize themselves with Wikipedia's deletion-related policies and guidelines, leading to large numbers of low quality !votes that complicate discussions. In very rare cases, after discussions are already severely affected by canvassing, we semi-protect them. Canvassing creates a lot of drama, rarely helps a deletion discussion, and wastes a huge amount of time and energy.
Is it time to place greater restrictions on AfD? Three inter-related questions for the community. Please note that this is not a proposal, but a discussion to see if a proposal makes sense.
1. Should there be stricter requirements to start a deletion discussion?
2. Should deletion discussions be semi-protected by default?
3. If yes to either of the above, what is the best way to allow new users to participate productively (for example, using AfD talk pages)?
Some context for why I started this thread: For years I've participated at AfD and have seen the problems caused by canvassing over and over again. So question #2 has long been on my mind. What has me thinking about question #1 took place over the weekend: a Wikipedian created an illegitimate sock puppet for the sole purpose of nominating for deletion three related articles: Anna Gifty Opoku-Agyeman, Corina Newsome, and Earyn McGee. It's not the first time I've seen people use AfD to nominate groups of related articles in bad faith, nor the first time I've seen it used to to target biographies of women or people of color in particular. It didn't take long to cause a stir on Twitter, etc., perceived as yet another example of systemic bias on Wikipedia. Of course, those of us insiders know that this was actually an example of process working in the end -- that this was just one illegitimate sock puppet causing trouble, and the articles had little chance of being deleted because it's "not a vote" and whatnot. Here's the thing, though: it's still damaging. Bad faith nominations are not only a huge time sink to the community, requiring people to make sure process does win out; it's also a terrible experience for the article creator/editors, it's a terrible experience for the article subject, and it's a terrible experience for anyone else who looks in and cannot be expected to see what we see. They see Wikipedia working on deleting a topic they care about, and cannot be expected to understand the "don't worry, it's not a vote, and process will win out" part that we might say to ourselves while grumbling. So I, for one, do think it's time to raise the bar a bit. How much to raise it is the big question as far as I'm concerned. — Rhododendritestalk \\ 22:10, 9 June 2020 (UTC)[reply]
On the last two of those, unless I'm missing something, that was an empty nomination ? I would say that admins should feel empowered to close empty or bad faith nominations, especially if they believe they may draw external involvement (Which should be taken as a given for any BLP for anyone of an underpresented minority on WP). If an experienced editor believes the article does merit deletion, let them open a fresh deletion discussion with proper rational (and there should be no penalty here if that's opened even the same day as the rapid closure of the previous one). We may not catch all the bad faith ones, if they are nominated with a reasonable cause (as the first of your three appears to be on a first quick read), but at least we shouldn't let the clear bad ones linger for the 7 required days and cause long term problems --Masem (t) 22:26, 9 June 2020 (UTC)[reply]
Right. Like I said, process usually wins out, but why is this permitted to begin with? How often do you see successful, good faith, policy-based nominations from new users, as compared to the kind of problems caused in this example? How many of those positive examples could be handled through other means (e.g. requesting an AfD at WT:AFD, PROD, etc.)? My central point about question #1 is about new users' nominations being a net negative, and that the negative effects probably reach further than most people would think, because we tend to think of AfDs as being behind-the-scenes projectspace business. — Rhododendritestalk \\ 22:32, 9 June 2020 (UTC)[reply]
I think the deep problem is identifying bad faith nominations. I worry that closing empty nominations is not a robust solution to the problem, because it doesn't take much for disruptive editors to learn how to give the appearance of a rationale. Just quickly looking back over the (all presumably good faith) AfDs I've participated in this year, the modal deletion rationale is typically one sentence along the lines of "This article does not meet the notability guidelines", and (very reasonably) nobody blinks an eye when that's written by an editor with a few hundred edits who stacked a dozen pages in AfD in one afternoon with identical rationales -- most of the time, that sort of deletion is just a user who spent a few hours helping to build the encylopedia by patrolling for non-notable pages, and decided they found several. So I worry that resting everything on an idea like "admins should delete any rapid string of AfDs by a new editor with empty/totally trivial deletion rationales" just moves the problem to a question of how to tell the difference between good faith (but perhaps rather lazy) tagging on the one hand, and disruptive trolling on the other. In this situation, for example, it seems reasonable to guess that with a bit more effort the person who started this AfD might have been able to write a persuasive appearance of a sincere deletion rationale, since they openly admitted to being a sockpuppet during the AfD (as was noted at AN). And that same AfD but with a policy-motivated deletion rationale would still have been subject to all the same canvassing, spam, and trolling. - Astrophobe (talk) 23:06, 9 June 2020 (UTC)[reply]
Empty nominations are easy. And given that most experienced editors know of the BEFORE process and how to nominate, I could see that when we have a sub-par nomination (no sign of prior research, maybe just claimed "person is non-notable", and a quick check of the target AFD page shows 20+ sources with clear reliable sources being used, they can do this rapid close and add something in their close "Any experienced editor, believing this was a valid AFD, may reopen/restart this". Heck, that's even better, just have the rapid close if the admin thinks it is a bad faith AFD, but if an experienced editor thinks it is valid, they can ask to have the nomination opened again on the admin's talk page, mimicking the process one uses to question the standard admin closure process.--Masem (t) 23:49, 9 June 2020 (UTC)[reply]
I would never base a "bad faith nom" on the basis of the account only, unless I know that editor has some type of block/warning or the like specific on using AFD in that topic area or in general. (eg, someone that I know has a AP2 DS on them that they are not to make any edits in that area, and they nominate a topic clearly in the AP2 topic area, that's a bad faith). Barring knowledge of that, the only assessment of "bad faith" is the nature of the nomination and the actual sate of the page - is there a massive disconnect that indicates that this may be a POINTy or nonsense AFD that AFD doesn't need to waste its time with. I agree we should not judge the editor - IP, new editor, or experienced - otherwise in evaluating whether an AFD is good or bad faith normally. --Masem (t) 14:07, 10 June 2020 (UTC)[reply]
That is good to know, but I felt like having to emphasize it because the general conclusion in WP:ANB seems to have been along the lines of "three AfDs were started by sockpuppets accounts with a vengeance," i.e. not worth being taken seriously (I'm quoting Silver_seren specificly, but it was more or less the general opinion). --bender235 (talk) 16:20, 10 June 2020 (UTC)[reply]
I mean, if this solution I suggest is implemented, and one finds that a single user has been submitting several AFDs in a row that have been quick closed as these bad faith noms and suspects possible sock activity, by all mean then check to see if the editor is a sock. But the editor should not be pre-judged outside of any known DS/bans attached specifically to that editor's name if we all for evaluating bad faith noms. --Masem (t) 16:28, 10 June 2020 (UTC)[reply]
First off, I would be against semi-protection by default because many articles listed for deletion are from new editors, and they should be able to participate in the deletion discussions of their articles. While this doesn't always wind up for the best, I imagine locking them out of the discussion or bunting them to an unseen talk page would have even worse outcomes. However, I would be in favor of raising the bar for filing a deletion to extended confirmed, as virtually all new page patrollers will meet that standard easily, and it will create a significantly higher hurdle for bad-faith actors. This won't stop PROD or CSD tagging - but that's a feature, not a bug. Both are easily removed in cases of abuse, and let people that are not extended confirmed and still want to help address the worst new articles. The Squirrel Conspiracy (talk) 23:03, 9 June 2020 (UTC)[reply]
This is a good point. I actually intended to be less specific than "semi-protection by default" in order to allow for that one exception (article creators/editors weighing in), but forgot when it came time to hit save. — Rhododendritestalk \\ 23:35, 9 June 2020 (UTC)[reply]
Thanks for starting this discussion Rhododendrites, and I think it's worth reinforcing that this problem of targeted and high-profile bad faith deletion spam is not at all new, and that with the growth of conversations about Wikipedia across other major platforms, I expect this sort of canvassing to only grow more severe. Per the opening paragraph of the discussion and per The Squirrel Conspiracy, I expect that I would agree with a proposal to require a higher bar to begin AfDs. But requiring a higher bar to contribute to AfDs is, to my mind, much more complicated. On the one hand, I am really sympathetic to the argument that it would be dangerously discouraging to new editors. I remember vividly my early experience editing Wikipedia: I believed that about of whatever you do on this website will get rapidly undone for completely opaque reasons, with lots of giant paragraphs full of incomprehensible acronyms and links and all sorts of emphatic italics about how astonishingly bone-headed you must have been to write that content (I'm not saying that's the impression people were trying to give, just that that's how it often feels to very new editors). People absolutely should be encouraged to WP:BB from their very first edit, including writing pages from scratch, and if their page comes up for deletion they should be allowed to participate in the discussion on it. From personal experience I believe that good faith participation in AfDs by brand new editors who don't yet have a clue is a huge net good for the project, especially as a hugely important (if often unpleasant) learning experience for them. Nothing motivates you to wade deep into notability policy like trying to come up with an argument for why your afternoon of work shouldn't be undone. Having said all of that, not raising the bar for AfD participation leaves half of the problem we're talking about unaddressed: it means that canvassing good faith and constructive deletion discussions is still just as easy, whether you're trying to sway the discussion towards keep or delete. It's very easy to imagine a good faith editor questioning the value of a page about someone with tens of thousands of twitter followers and that person reacting by canvassing support, just as happened in this instance, in which case we would be in the same exact position that we're in now. So I would be very interested in discussing further policies that would allow people with a sincere connection to the page to participate, while ruling out the kind of canvassing that is already a very serious problem and that looks like it will only get more serious over the next few months and years. - Astrophobe (talk) 23:44, 9 June 2020 (UTC)[reply]
I honestly don't see it as a problem. Most of the time canvassing is obvious and the topics are notable. I actually got stuck into the project because I wasn't specifically canvassed, but I read something about whether something should be on Wikipedia off of Wikipedia. Not being able to participate may create a "walled garden" effect for the entire community. That being said, there is a bad faith nomination issue, it was obvious in the cases you mentioned, and we need to do a better job of a community of not defaulting to "no consensus" when a deletion discussion goes off the canvassing rails, but I don't really support increasing the standard threshold. For instance, this should be very unlikely, but there may be instances where a low profile BLP realises there's an attack page written about them here and needs to deal with it. I might be willing to support a specific action item, though, such as a flag when a non-extended confirmed user starts an AfD. SportingFlyerT·C01:47, 10 June 2020 (UTC)[reply]
I would oppose these type of changes. Its hard enough to delete an article as it is. Think about it, it only takes one person to create a bad article, but many to have it deleted. And when we can't agree and the AfD is closed as "no consensus", it gets kept by default. This actually contradicts WP:ONUS where the person adding the material must get consensus, not the person proposing deletion. As for sockpuppets, that is not a issue exclusive to AfDs. They can show up in any discussion anywhere.--Rusf10 (talk) 02:07, 10 June 2020 (UTC)[reply]
can show up in any discussion anywhere sure, but in structured discussions they can be more disruptive. it's also a place where it's much less likely they'll be able to contribute positively. in an article talk page, there's at least an argument from the perspective of knowing the subject; arguing about notability is a bit more, well, technical from a procedural standpoint. — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
Right, canvassing for keeps is a huge problem. And isn't that what Rhododendrites's suggested point 2 addresses? The way I read it, the problem you describe is a big motivation for that remedy. - Astrophobe (talk) 04:52, 10 June 2020 (UTC)[reply]
I would even equate off-line canvassing by a subject or by a connected contributor to COI editing. Point 2 of the proposal would not take care of the issue (most of these accounts were autoconfirmed, just dormant), and probably has zero chances to pass at any RfC anyway.--Ymblanter (talk) 05:51, 10 June 2020 (UTC)[reply]
Bender, are you suggesting that the three articles in question should have been deleted? That they are "Wikipedia shrine[s] for their personal vanity" in your book? Axem Titanium (talk) 07:39, 10 June 2020 (UTC)[reply]
I do agree with you that offline canvassing is essentially just COI editing. What I do not think is clearly true that most of the accounts in the recent spate of canvassed AfDs were autoconfirmed editors -- look at the two rapidly closed (and therefore actually readable) AfDs at Corina Newsome and Earyn McGee. Both of them were absolutely overrun by IPs and single-purpose accounts. It's easy enough to say that the suggestion as written here so far wouldn't perfectly solve the problem or is pretty much guaranteed to fail at RfC, both of which I agree with. More interesting is asking how we can tweak AfD to make it robust to these sorts of multi-front attacks from the outside, which have already been seriously disruptive and I believe will only grow more severe. It could very well be that the answer is there is no possible reform and we just have to live with this issue, but I don't think that's possible to conclude without some more discussion. - Astrophobe (talk) 08:14, 10 June 2020 (UTC)[reply]
I think if every instance of canvassing on twitter would result in a COI template appearing on top of the article, and potentially in an appearance of a paragraph explaining how he subject was canvassing on twitter then they will start thinking twice before starting canvassing. I agree that semi-protecting AfD would generally help (though not entirely) to this issue, however, it is not really desired from other points of view, and this discussion so far shows a clear opposition to this proposal. In addition, I have no idea what to do if (i) a Wikipedia editor canvasses other sympatheric Wikipedia editors outside Wikipedia (which happend a lot and in the past resulted in keeping clearly non-notable articles) and (ii) people are showing upat AfD and it is clear that they are correlated but the source of canvassing could be found. --Ymblanter (talk) 10:02, 10 June 2020 (UTC)[reply]
The really harmful perception that encourages canvassing seems to be that it's a straight up or down vote that everyone in the world has an inherent right to participate in. We already have Template:Not a ballot, but it's clearly highly ignorable for motivated people. I wonder if there is a template that is garish and intimidating enough to actually persuade people that canvassed votes won't work. Maybe a pop-up like some web sites have to discourage ad blockers, and an mp3 that autoplays a siren noise when you load the page ;) - Astrophobe (talk) 18:31, 10 June 2020 (UTC)[reply]
@Astrophobe: as Ymblanter correctly pointed out, a lot of the Wikipedians canvassed to those AfDs were inactive but established accounts. A rule limiting the participation of newly created accounts therefore wouldn't help. Of course, generally restricting sporadically active Wikipedians for !voting isn't a viable solution, either. After all, we are a project of volunteers and clearly not everybody finds time and means to contribute on a regular basis. It's just that in those particular three AfDs the canvassing was so blatantly obvious, with person after person basically copy-pasting the same rational referencing the rarely cited WP:BASIC over and over again. When I was looking for fellow veteran Wikipedians to intervene on the evening when all of this unfolded, Sulfurboy reassured me that "any admin worth their salt will see past meat and spa votes." Unfortunately that never happened. --bender235 (talk) 13:05, 10 June 2020 (UTC)[reply]
As I see it, there are two separate issues. One is people who have (almost-)never used Wikipedia before, who have no investment whatsoever in the site, either being unaware of our rules about things like canvassing and COI or having no reason to care about them. Anecdotally I think this is one of the most far-reaching problems confronting Wikipedia. I know that when I've tried to explain things like "you shouldn't write a page for your dad" to people in my life who don't edit here, the most common response is something along the lines of: nobody cares about Facebook's terms of service or Twitter's terms of service or The New York Times's terms of service, why should I care about Wikipedia's terms of service? It's easy to imagine that a bunch of the first-time editors who were canvassed into that discussion would tell you that the principle they were following in voting keep is more noble than abiding by Wikipedia's policies would be. That's the issue that I think there's room to fix. The second, separate, issue is people who actually are editors here, or who have been active editors in the past, who may or may not be breaking COI/Canvassing rules. I'm not interested in accusing anyone of anything so I'll just assume for the moment that is a problem that exists in the abstract. It's hard to imagine a policy-based solution to that problem other than sanctioning the user, because if somebody has an investment in the website and is ignoring policy anyways, then I definitely agree that we shouldn't adopt a suboptimal global policy to handle that; we have a whole other set of rules for user misbehaviour. So I see the former as worth trying to fix with a policy change, and the latter as just a matter of users breaking rules and all the various policies we have to deal with that. - Astrophobe (talk) 18:47, 10 June 2020 (UTC)[reply]
XfD is already biased too heavily towards indiscriminate inclusionism. We smile benevolently on keep vote canvassing, and allow personal attacks on nominators to pass without comment. Now here is a proposal to skew the conversation even further away from discussion of article subjects and contents and further towards lawyerly rules about who is allowed to talk. I am not in favour. ReykYO!08:42, 10 June 2020 (UTC)[reply]
I'm not sure I've ever seen an indiciation of AfD being a hotbed of indiscriminate inclusionism, nor indeed community encouragement (or at least, lack of notice or reticence) on canvassing of Keep votes. Nosebagbear (talk)
This suggests that there are areas where pretty much everything nominated gets kept, does not matter whether or not material is compliant with WP:GNG. Not that I strongly oppose this, and in some areas (such as localities) it probably makes sense, but this definitely backs up the inclusionism claims.--Ymblanter (talk) 09:51, 10 June 2020 (UTC)[reply]
Well, by its nature XfD tends to attract pages that ought to be deleted so in that sense it leans towards deletionism. But what I mean is that the lenience we show to misbehaving editors correlates directly with whether they voted keep. For instance, I once objected when some pretty blatant keep vote canvassing was allowed to determine the outcome of an AfD/DRV. All I got in response was blank looks and a (hopefully not serious) suggestion to counter-canvass if it bothered me so much. That's not advice I intend to take because, even if I felt like being unethical, a delete voter could never get away with it. I could give other examples of keep voters free to make insulting personal commentary and delete voters getting in trouble for backchat but of course if I did it would only be dismissed as a list of personal grievances. ReykYO!10:26, 10 June 2020 (UTC)[reply]
I could understand this position regarding #1, but #2 is much more likely to apply to canvassed keep !votes than canvassed delete !votes (at least in my experience). — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
So, I get both the base concern(s), but also the issue with SP - that the creator in particular is disadvantaged. On the thought of Extended-Confirmed to start an AfD - does anyone know what % of good-faith AfDs are started by non-EC users? That seems relevant here. It's a shame we don't have PC2 - this would be a great area for it. Nosebagbear (talk) 09:44, 10 June 2020 (UTC)[reply]
AfD needs improvement, for certain. Some kind of competency requirement for nominating articles could help, so could a quicker closing process for bad nominations. Unfortunately, it's hard for me to come up with a good way of accomplishing...Jacona (talk) 12:58, 10 June 2020 (UTC)[reply]
AFD will never be perfect and it is 100x more "friendly" than it was 10 years ago. Participation is lower as well. I see no benefit to suppressing participation any further and that is what more rules will do. Dennis Brown - 2¢13:48, 10 June 2020 (UTC)[reply]
This would be solved if we had more mechanical and less subjective notability rules. Then it wouldn't matter who was being canvassed. We should repeal WP:N altogether and just amend WP:V to require two independent, in depth, reliable secondary sources for every article. Then AFDs will just be about whether there are two qualifying sources or not. If there are, it can have a stand alone page. If there aren't, no stand alone page. Simple and no need to discuss whether or not something is "notable". No SNGs to argue about. Basically, make GNG a hard policy and be done with it. Levivich[dubious – discuss]16:21, 10 June 2020 (UTC)[reply]
We've seen editors try to work with more mechanical/objective application of notability "rules" , claiming things like "I have three sources, that's enough", but this makes things worse because now you have people gaming the system worse than what we see now. Also, this underminds the purpose of notability on WP, which is to reflect topics that are likely to be able to be fleshed out to fuller articles but need sourcing work to help get there, and because we have no DEADLINE, require the flexibility of judging what sourcing exists at AFD rather than rote rules to keep them. --Masem (t) 16:34, 10 June 2020 (UTC)[reply]
@Levivich: I disagree, because that would only further muddy the distinction between what's verifiable and what's notable. Those two are not the same, and while the existence of reliable sources (i.e., verifiable facts) is necessary for someone or something to be notable, they aren't sufficient. --bender235 (talk) 16:50, 10 June 2020 (UTC)[reply]
@Masem, Bender235, and Axem Titanium: Thanks for your comments. To clarify, I am indeed proposing something radical: far more than "blurring the lines" between V and N, I'm talking about getting rid of that line altogether. That's why I'm not worried about "the purpose of notability", because I advocate getting rid of the entire concept of "notability". Let's face facts: 6,000,000 articles, and they're not all about important topics. We have hundreds of thousands of articles about athletes, songs, Pokemon characters, and all the rest. If the purpose of notability is to reflect topics that are likely to be fleshed out, well, then WP:N has failed miserably at that purpose.It's the entire concept of "notability" that is to blame: the notion that a topic has some property, "notable", that determines whether or not it should be in the encyclopedia, and we, as editors, are tasked with examining the topic and determining if it has this property or not. We act like notability is something we discover. It's not. It's something we invent. "Notability" is whatever we say it is; literally, whatever we agree to write at WP:N. If the purpose is to identify topics that can be fully fleshed out, there is no better way to do that than to identify if there are two good sources that we can use in the article. If there are two good sources, we can write an article about it that complies with V, NPOV, and NOR. If there aren't, we can't. This is the principle behind GNG, WP:THREE, WP:42, etc.We should embrace the fact that an AFD is not about a topic's inherent property of notability, but really just about whether to have a stand-alone page or not. We should have a stand alone page if we have the sources to support it. By making the "notability" simply a matter of "sufficient available sourcing: yes or no" and not about anything else, it will be harder, not easier, to game. Every keep !vote, to "count", would have to identify the two sources, and the entire discussion would be about whether the two sources meet WP:GNG criteria. The current system is already being gamed, and has been gamed, for a long time. Gaming is what led to this thread in the first place. Restricting the conversation to just be about the quality of sourcing and nothing else, will lead to less gaming, nor more. Levivich[dubious – discuss]17:30, 10 June 2020 (UTC)[reply]
@Levivich: that's a radical idea, to put it mildly, and I'm afraid that completely eliminating the notion and threshold of notability would turn Wikipedia into somewhat of a repository for everything that was ever written, and every person that ever existed. I mean, I might be able to find a census entry and a birth announcement (two reliable sources!) of some 19th-century John Smith of Iowa, but what's the point of writing up an article recounting his dull biography of plowing the corn field from the cradle to the grave? At some point we have to be firm and say Wikipedia is just not the place for this. --bender235 (talk) 17:39, 10 June 2020 (UTC)[reply]
Bender235, 6,000,000 articles says to me that Wikipedia already is a repository for everything that was ever written. Please note that I didn't say "two reliable sources", I said "two independent, in-depth, reliable secondary sources" (in other words, same as WP:GNG), so no, a census entry and birth announcement wouldn't cut it. Requiring two GNG sources for every article will reduce, not increase, the number of stand-alone pages. Of this much, I'm sure. What makes my proposal radical is that if it were implemented, millions of articles would be eligible for deletion, which are not currently eligible for deletion, because meeting GNG isn't currently universally seen as a requirement. Levivich[dubious – discuss]17:57, 10 June 2020 (UTC)[reply]
No matter how you slice it, WP:PROF almost certainly needs to remain a standalone rule. Many academics are worth having an article written about them despite never having appeared in a newspaper. Significant coverage in secondary sources is not a requirement; merely having one's research (a primary source, albeit a reliable one due to peer review) cited heavily by other papers is sufficient to meet the bar. And we can write an article on their work using mostly those primary sources, with the reassurance that they are reliable because they have been thoroughly vetted by the academic community. -- King of ♥ ♦ ♣ ♠19:01, 10 June 2020 (UTC)[reply]
King of Hearts, if we did this, I would support having exceptions (specifically to the "independent" and "secondary" requirements), including PROF exception, as well as for other specific areas where there is a lack of independent or secondary sourcing, but where the community feels non-independent or non-secondary sourcing is nonetheless reliable enough to satisfy V.
Instead of asking, at an AFD, "is it notable?", we ask, "is there enough verifiable information to support a stand-alone page?" A statement, to be verified, needs to come from a reliable source (a source with a reputation for accuracy), it needs to come from an independent source (or else there's a bias concern, usually), and it needs to come from a secondary source (to avoid OR interpretation of primary sources). For an entire page to be verified (or in other words, for a topic to be verified), we also need in-depth sources: enough content to fill a page.
Even if the community adopts this view of verification, it can still decide that there are some topics, like PROF, where a "reliable source" need not be independent or secondary, and so exceptions could be made. This is also the sort of exception that could be made to address under-coverage of historically marginalized people and topics. Thinking of whether to have a stand-alone page as a matter of V instead of N is a better framework all around. And then, in AfD discussions, the only keep !vote that would count would look like "keep - [source 1] [source 2]", and it wouldn't matter if people were canvassed or IP editors or socks or whatever, because instead of counting votes, or assessing votes, we would just be counting sources and confirming that they meet "the test" and that there's two of them. Levivich[dubious – discuss]20:26, 10 June 2020 (UTC)[reply]
I agree with those who argue that an uninterpreted WP:V is not enough of a basis for deletion policy, but I agree that notability has not served us well. The problem that deletion policy is there to solve is that there are forces out there that aim to undermine the encyclopedia, so we need to choose the ground that we can defend. The notability criterion is a solution: it says the topics we should have articles for are those on which good articles could be written. I have thought since 2006 this is wrong: the criterion we should apply is maintainability, not notability, and we should deal with articles as they are, not as they might be (although I am all for editors who transform bad articles into good ones during the AfD process). The events that convinced me of this led to this AfD: Wikipedia:Articles for deletion/Stephen Schwartz (journalist), an example of something then unmaintainable that I thought should have been deleted, although the subject was notable. More recently I have been bothered by how the WP:INHERIT criterion has frequently been used to delete high-quality, well-maintained, encyclopediac content; cf. Wikipedia:Deletion review/Log/2020 June 8 for the most recent example I aware of; there are have been better exmples. We should drop the abstract ideal of notability as the criterion we use and adopt the pragmatic criterion of maintainability that I think in time would lead to a more intuitive deletion policy. — Charles Stewart(talk)06:45, 11 June 2020 (UTC)[reply]
Speaking as a proud simpleton who loves hyper-minimalist rules, I would support this, but it seems like a different (though of course related) proposal. - Astrophobe (talk) 18:37, 10 June 2020 (UTC)[reply]
I think one of the issues is that AFD, unlike a lot of other Wikipedia processes we think of as happening "in the background", slaps a big red notice on top of an article in articlespace. I'm not suggesting that we change this at all, but it is worth keeping this fact in mind when we discuss solutions. The notice demands your attention when you're on an article and even invites you (yes, you!) to participate in the deletion discussion. You can imagine that a new/IP user would feel confused if they're not allowed to participate at all at this point. Axem Titanium (talk) 16:57, 10 June 2020 (UTC)[reply]
This is a good point, thanks. If we enacted some sort of restriction like this, at very least the wording of that notice should be changed, but I'm not sure in what way. — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
I'm sure that banner is responsible for a good chunk of the hollow "keep" votes that show up for pop-culture articles. It stands to reason that if you're looking up the article for a particular thing, you beleive that particular thing should have an article. (Even if there's no particular policy-based reason for it to.) ApLundell (talk) 14:52, 18 June 2020 (UTC)[reply]
Personally I would support automatically adding extended confirmed protection (30 days/500 edits) to all AfDs as they are created. Positive contributions from editors not meeting these criteria are incredibly rare IMO, and it would stop SPAs, socks, IPs called from social media etc. Number5717:46, 10 June 2020 (UTC)[reply]
I agree with this suggestion. Article creators who are not EC can make their case on the talk page (along with other non-EC editors). EC editors can read those talk page arguments and take them into consideration in their AFD !votes. The closer can also take into consideration arguments made on the talk page. But it'll help keep the discussion more focused if only EC editors participated on the AFD page. Frankly, non-EC editors do not have the experience necessary to meaningfully contribute at an AFD, even if they wrote the article. And I say this as an editor who participated in AFDs before I was EC (and I shouldn't have, because I had no understanding of notability guidelines then [or now really]). Levivich[dubious – discuss]18:01, 10 June 2020 (UTC)[reply]
I would like to see some facts here. How many AFDs in the past month/year can we reasonably classify as being disruptive in the senses concerned in this proposal? I would say that if that number is less than 5 or 10%, I don't see a need for systemic change. --Izno (talk) 18:13, 10 June 2020 (UTC)[reply]
Under these suggestions anyone who is extended confirmed could nominate AFDs. Personally I would support limiting nominations and participation to extended confirmed users and the article creator because it would give more time to block the sockpuppets who seem to zero in at AFD whether they are nominating or voting, imv Atlantic306 (talk) 00:20, 11 June 2020 (UTC)[reply]
Under normal circumstances I think our AfD process works fine and don't need adjustment - usually no or minimal disruption and no need to protect the AfD until something problematic happens. For example, I recently dealt with an case where an article creator was blocked during an AfD of their article and suddenly brand-new accounts showed up to !vote. SPI, checkuser, semi-protected just because of the sockpuppetry, bam - dealt with. What we need to have a process for is cases like these, which are the exception rather than the rule - demonstrable and widespread off-wiki canvassing that turns the AfD into chaos (a flood of mostly-new users using non-policy-based arguments). I think semi-protection is the right call in most cases, but what do we do if there's demonstrable canvassing of experienced editors, for example? creffett (talk) 00:45, 11 June 2020 (UTC)[reply]
Some of the examples listed above are quite extraordinary, I agree with bender235 that “something” needs to be done before this becomes the normal. Whilst it would not deal with bad-faith nominations and canvassed inactive users, perhaps upon presentation of evidence of off-Wiki advocacy, !votes be restricted to extended confirmed users and !votes already cast by non-XCON users be struck/deleted. Cavalryman (talk) 03:42, 11 June 2020 (UTC).[reply]
@Cavalryman: actually what upset me the most in these Twitter canvassing campaigns is the piggybacking on a social justice cause. People weren't just told to vouch for the notability of some hashtag activists, they were sent here to fight supposedly systematic sexism and racism in Wikipedia and its entire community (see [5], [6], [7]). And sure enough the majority of canvassed !voters came waltzing in crying racism right away without even bothering to consider the arguments presented up to this point. That's what concerns me the most. Apart from slandering the Wikipedia community unjustly, it makes certain subjects and topics toxic to a point where our usual (bureaucratic) processes can no longer be applied. Who wants to be the Wikipedian permanently branded as a racist in the Twitterverse simply for questioning the notability of a social media starlet? The nominator of AfD/Anna Gifty Opoku-Agyeman stated that he/she created a sock puppet rather than use his/her established account to avoid online harassment, and perusing the comments and replies of the self-righteous Twitter mob above, I don't think that was a stretch. To me, this whole incident and its likely future copy-cat versions are worrisome. (And just to show that I am not exaggerating, here is a now-deleted tweet by MethanoJen singling me out by name, simply for questioning whether her newly created Category:Black geoscientists doesn't fit our existing category pattern.) --bender235 (talk) 16:39, 11 June 2020 (UTC)[reply]
Please write to T&S about the tweet, this is a cleart wiki-harassment. I have warned her in the morning (qand may be this is why the tweet has been deleted), but if I have seen this I might have indeffed.--Ymblanter (talk) 17:41, 11 June 2020 (UTC)[reply]
bender235, I agree completely, aside from the utterly appalling conduct of that editor the broader trend in identity politics is to brand anyone who presents a rational and articulate counterpoint a racist/bigot/Nazi etc, thankfully not a common issue in the dog articles I tend to edit and to be honest one of the reasons I usually give anything political on Wikipedia a very wide berth. I tend towards supporting the idea of BLUELOCK for AfD discussions (less article creators), I suspect SILVERLOCK would be no impediment. One of the reasons I proposed a middle ground above is to protect closers from the inevitable social media targeting that would follow from a close that went against canvassed IP & SPA opinion. Kind regards, Cavalryman (talk) 22:48, 11 June 2020 (UTC).[reply]
Wikipedia has already strayed too far from being the encyclopedia anyone can edit. Keeping that in mind, further restrictions on editing abilities for newer users should only be implemented when absolutely necessary. I think our admins are pretty good at recognizing canvassing and meatpuppetry by SPAs and the like. Since AfD isn't a vote, closing admins are expected to throw out !votes that are frivolous and/or not based in our policies and guidelines. Even if that weren't possible, I'm not convinced it happens often enough to justify such drastic action. We also have to consider the effect this would have on editor retention. Wikipedia is already confusing enough to newbies, with its byzantine policies, litany of jargony acronyms, and Kafkaesque bureaucracy-that-isn't-a-bureaucracy. I'm convinced this would be a net negative. AfD has far more pressing concerns to deal with anyway. The biggest two that come to mind are careless nominations where WP:BEFORE clearly didn't happen (especially wrt non-English sources), and nationalistic or politically motivated bloc voting by established editors. Established editors know how to make their !vote look like a valid policy-based rationale even when their real motivations are ILIKEIT/IDONTLIKEIT. Freshly recruited meatpuppets don't know how to do this, and so closing admins can safely disregard them per WP:NOTAVOTE. In particularly extreme cases, admins should semi-protect the page as they sometimes do now. That's far better than the current proposal which throws the baby out with the bathwater. −−−CactusJack 🌵05:11, 11 June 2020 (UTC)[reply]
While I appreciate the argument on principle, I don't know if being an encyclopedia that anyone can edit is the same as being "the encyclopedia that anyone can edit, and anyone can jump into the behind-the-encyclopedia technical processes without spending time learning about them first". As for WP:BEFORE, I don't necessarily disagree; don't you think that requiring more experience would make it more likely that someone is familiar with (and follows) that guidance? — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
We need to remember that AFD is just one of our deletion processes, if we were to restrict people from filing AFDs we shouldn't be surprised if they tag more articles for Speedy deletion or simply draftify them. That said I'm OK with the idea that we restrict some people from deletion generally. Over the years I have seen a number of editors who didn't realise they were overly deletionist until they ran an RFA and had their deletion tagging checked and criticised. So I would be OK with 6 month bans from the deletion process where people were only allowed to participate in the deletion process re articles that they had started. I really really don't like the idea of restricting people from a deletion debate where it is their work that we are considering deleting. So restrict the people who have been making mistakes in their deletion tagging, not a blanket restriction on new or newish editors. ϢereSpielChequers13:32, 11 June 2020 (UTC)[reply]
I don't know how we've escalated past the early suggestions of SP for participation to ECP, which is seriously, seriously OTT. While their average edit quality is certainly worse, I've seen many legitimate !votes from editors in that group. Shifting to talk page for all of them brings multiple issues: that's fiddly to spot, so some just won't note to participate there (that is, they'll know the TP exists, but not that they'll be read) & also massively drops that chances that every person in the AfD will read the !votes or comments, which disrupts and weights the discussion inappropriately. We also should be using the least disruptive method, and disrupted AfDs are relatively rare. We aren't implementing a "have more experienced participation" restriction. Nosebagbear (talk) 14:49, 11 June 2020 (UTC)[reply]
AfD is infested with socks. You can see it with old AfDs (a few years back) and seeing how many participants have a strike-through (with that userscript installed). They get busted eventually, somewhere else, and leave behind fossil evidence. I would support reasonable moves to address this problem. -- GreenC16:51, 11 June 2020 (UTC)[reply]
Yes, there is a problem, but it is only a problem with a very small percentage of AfD discussions. Far more discussions have the problem of a lack of participation, which this proposal would only exacerbate. Maybe we need to encourage people to be less tolerant of canvassing, or of other abuses of this process, but I don't think this is the right way to go about it. I remember that my very first logged-in edit to Wikipedia 13 years ago was made to an AfD discussion in response to canvassing on another site, but it was not supportive of the canvasser. Phil Bridger (talk) 19:50, 11 June 2020 (UTC)[reply]
I wonder how typical my own experience is. I began editing in any serious way in 2013. I think it was late 2017 or early 2018 before I even knew AfD existed. Once I discovered it I spent many weeks just observing it before I commented. It was months before I put my first article up for deletion. How many people in this discussion have a completely different experience? I simply don’t assume good faith for ‘new’ editors who show up and are busily nominating articles for deletion in the first couple of weeks. There are all kinds of productive ways new editors can contribute to the project but sitting on your hands for a while before you start nominating articles for deletion seems entirely appropriate. Mccapra (talk) 20:02, 11 June 2020 (UTC)[reply]
@Mccapra: I had a similar experience. My first participation in a AfD was in AfD/Jamaal Anderson in 2007, only after having made hundreds of contributions over the years. AfDs—or any Wikipedia backroom bureaucracy—are almost naturally intimidating to the uninitiated, due to the various cryptic acronyms that are casually thrown around by the regulars. Unfamiliar with these, inexperienced or canvassed editors tend to copy-paste these acronyms in AfDs without actually understanding them, which makes them easy to spot for the trained admin eye. To his credit, creffettimmediately spotted the unusual frequency of WP:BASIC citations in AfD/Anna Gifty Opoku-Agyeman. --bender235 (talk) 20:37, 11 June 2020 (UTC)[reply]
A couple of points worth mentioning here. Other language wikis It would be great to both explicitly encourage editors to look at other language wikipedias for sources and to encourage editors from other language wikis to participate in AfD's, especially in situations where there is the likelihood of sources being in non-English languages. Draft namespace Draft namespace is relatively new compared to AfD and moving good-faith contributions to draft to enable relatively slow-moving editing to occur should be encouraged, particularly for topical subjects. SPAs and paid editing my feeling is that a large numbers of the SPAs involved in articles that end up at AfD are undeclared paid editors. This is a larger issue than AfD, but it may be worth thinking about what can be done in this specific context. The best I can think of is a bot that creates a table on the talk page listing all the AfD participants and editors involved in the article and gives edit counts, how many are related to the issue at hand, and also scans for their names in sockpuppet investigations and other administrative actions. Stuartyeates (talk) 21:21, 11 June 2020 (UTC)[reply]
I am very against default ECP. I'm open to reasonable suggestions for how to resolve the identified problems, but ECP is not one of those. To comment on Rhododendrite's questions: (1) I think this is reasonable. We have technical restrictions on who can move pages, so I think it's reasonable to have slightly more stringent requirements to nominate for deletion. (2) I'm not a fan, but am open to it. I would prefer the first option and see how that goes before default protection. Perhaps more practical is expanding the protection policy to allow protecting AFD discussions for sock/meatpuppetry or obvious canvassing. (3) I think just encouraging use of the talk page by everyone would work, but why have newbies go to talk just to be ignored? We're basically telling them to send their emails to /dev/null. — Wug·a·po·des23:31, 11 June 2020 (UTC)[reply]
2. Should deletion discussions be semi-protected by default? Can I suggest some alternatives?
Grouped edit notice for Template:Editnotices/Group/WIkipedia:Articles for Deletion
Some type of edit filter warning for non-(auto)confirmed users
Something similar to Wikipedia:GettingStarted, but it pops up when you enter WIkipedia:Articles for Deletion/***** for the first time; if for non-(auto)confirmed users, it pops up something similar to {{Not a ballot}}.
If the above aren't going to work in any circumstances, okay then go ahead and semi- protect it and hope those SPAa and canvassed users don't gets 10 edits after 4 days. This will prevent new users from participating, but 99.999% of the time, they think it's a ballot. Nobody uses the AfD talk pages, so let's direct them there. With that comes with more very complicated ideas, like moving policy based votes into the actual AfD by experienced AfDers, and considering if the talk page will be additionally used to addresss the consensus outcome. {{reply to|Can I Log In}}'s talk page! 05:34, 12 June 2020 (UTC)[reply]
I'm certainly not against less restrictive interventions like these; I'm just pessimistic they would be helpful. I've seen {{notavote}} added to lots of AfDs, and [just based on anecdote of course] I've not really seen it help much. Call me cynical, but when I add it, I'm really just trying to signal to other experienced editors (and the closer) that there may be canvassing/SPAs going on here. — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
Thanks for starting this, Rhododendrites. Something that deletion discussions and move requests have in common is that, because they have a mandatory period and appear to readers, they can do damage when bad ones are launched. For move requests, we're trying to help by making the notice less prominent, but for deletions, it needs to be prominent. I wish there was a way we could signal to readers "this article is currently nominated for deletion, but it's very unlikely to pass", but we can't exactly just have it display the running !vote total (either technically or editorially). Still, there might be some changes we could make to Template:Article for deletion to help make it clearer what being up for deletion actually means. {{u|Sdkb}}talk08:06, 12 June 2020 (UTC)[reply]
We already have strict requirements for starting an AfD: WP:BEFORE. The problem is that they are not enforced. From what Uncle G has said, AfD was deliberately made rather difficult as a barrier to frivolous nominations. The tool Twinkle has subverted this to make a deletion nomination much easier than other, more productive actions such as searching for sources, working on the article or starting a discussion on its talk page.
Another problem is that the readership tends to be excluded from these discussions. An article may be read hundreds or thousands of times while it is at AfD but we rarely see these readers joining the discussion. I myself got started on Wikipedia when I saw a deletion notice on an article that I had been reading. Perhaps I have more aptitude for the Wikipedia interface than the average reader but if there had been greater barriers in place, then I too might not be here now and the hundreds of articles that I have started might not have been written. Wikipedia is the encyclopedia that anyone can edit and so we should freely accept comments rather than engaging in voter suppression by restricting discussions to a dwindling number of incumbents and insiders.
It strikes me as odd that we have concluded that the project is best served by creating some filters around article creation for new editors (I don’t know, maybe that’s still controversial?) but we continue to treat AfD as a free for all. It’s true that the best barrier we have is WP:BEFORE but I guess we’re having this discussion because it doesn’t seem as effective as it once was. Mccapra (talk) 11:15, 12 June 2020 (UTC)[reply]
Quite frankly, I think concerns over this matter are completely overblown. People have constantly been predicting doom and gloom over small problems, but it seems to me that AfD is more sturdy and capable of dealing with sockpuppeteering and canvassing than many give it credit for. New editors do not find AfD and immediately start making bad edits. It takes a lot of time for the average editor to even build up the confidence to start making proper edits, never mind contributing to AfD. There are really only two ways in which a new editor will even get exposed to AfD, either an article they created was nominated, or they were canvassed there. The former is an important learning experience, and being able to contribute gives a new editor valuable insight into how the process works. Stopping these editors from contributing will just further the image of Wikipedia as a bureaucratic nightmare where decisions are made by elitists in ivory towers. In the second case, such instances are isolated and so painfully obvious that dealing with it really does not require pre-emptive punitive measures. This sort of goes without saying, but default ECP is a terrible idea and I am opposed to it in the strongest possible terms. Devonian Wombat (talk) 05:02, 13 June 2020 (UTC)[reply]
Not doom and gloom. More about tons of wasted time, harassment, and possible external influence on our process (whether in good or bad faith). One of my original points was that we're typically able to deal with this, but there's so little benefit in forcing good faith participants to do so. Lots of wasted time, lots of attempts to influence the outcome. I don't disagree that the article creators/editors themselves should be allowed to participate, though. — Rhododendritestalk \\ 23:52, 13 June 2020 (UTC)[reply]
Follow-up - apologies to start this thread and only come back a few days later. Several good points here. If these were actual proposals, it seems like were firmly in "no consensus" territory here, at this point. One thing that I think would make sense for me (or someone else) to do if formally proposing these measures would be to gather some data. My perception is that, putting aside the article creators/contributors themselves, new users almost never make valuable, policy-based contributions to AfD. That applies to nominations especially, and !votes slightly less. But I appreciate that not everyone may have the same perception. One open question for me is how to allow article creators/contributors to participate while preventing other new users? Maybe the only way is to direct them to the talk page, and to rework the notifications to be very clear about how to do so (i.e. to do everything we can to encourage participating there). Not sure. — Rhododendritestalk \\ 23:58, 13 June 2020 (UTC)[reply]
As somebody said above (or elsewhere, I do not remember), there are only three categories of IP / new users taking part in AfD discussions: (i) they have been affected (created or significantly contributed to an article being discussed, typically by getting a template on their talk page); (ii) they have been externally canvassed to the discussion; (iii) they evade a block. If this correct (and research probably could be made about this - canvassing is difficult to detect but it must be visible by clusterization of new votes in the same discussion), then these issues probably should be separated -canvassing is not just about new users, and we certainly want creators of the articles participate in the AfD on the articles they created.--Ymblanter (talk) 09:15, 14 June 2020 (UTC)[reply]
I strongly support reforms to streamline the AfD process. In my time I have closed well over a thousand XfD discussions, and have observed two key patterns of sockpuppet manipulation. The first is where novices desperate to keep an article create numerous obvious sockpuppet accounts; the second is more sophisticated, typically connected to paid advertising, where the sockpuppet accounts are crafted with a veneer of legitimacy through the creation of nominal user space pages and through perhaps commenting on a handful of other AfDs or making a handful of other minor edits before engaging in the AfD of concern. Nevertheless, it should not be possible for an account created after the initiation of an XfD to participate in that XfD. I don't think that this is at all problematic for new users, who should expect that some time and experience is required to obtain certain rights. This should also not be a problem with respect to editors creating new pages. Quite frankly, editors should not be able to create new pages at all until they meet some minimal threshold of activity, so the ability to comment in XfDs should coincide with the ability to create new pages at all. BD2412T19:00, 14 June 2020 (UTC)[reply]
These concerns raised seem pretty well addressed by WP:AFD, WP:AFDEQ, and WP:DISCUSSAFD in that legitimate AfD debate cannot be drowned out by obvious manipulation (sock-puppeting) or poor quality commentary, as AfDs are not a poll (consensus is not based on a tally of votes, but on reasonable, logical, policy-based arguments) and those processing the AfD are required to adhere to this (and if they do wrong, the article does not disappear, and can be recalled through the deletion review process WP:DRV). It also appears some of the generalisations about new users (who may or may not be new, given some new registered users have previously been editing Wikipedia for years as non-registered users) has some undertones not in keeping with the spirit that Jimmy Wales had for Wikipedia (that the value of an editor was not in how long their registered account had been in use, or even how many edits they had made, but in the quality of their contributions to Wikipedia, which may come from registered editors both new and old). Some form of artificial class system based on seniority/tenure/clique would seem contrary to that - even the auto-confirmed class (which has been deliberately set at the low threshold of just a few days) isn’t really such a class system. Like Devonian Wombat stated, New editors do not find AfD and immediately start making bad edits... There are really only two ways in which a new editor will even get exposed to AfD, either an article they created was nominated, or they were canvassed there. Therefore, there does not appear to be an issue with the current framework for AfDs that requires re-invention in my view. Kangaresearch 08:16, 18 June 2020 (UTC)[reply]
What Dennis Brown said: AfD is one part of WP that really works well, participation at AfD is down, newcomers who are there are apt to be defending pieces they personally have a stake in, there is no better way to learn the notability standards than to actually participate in deletion debatese, there would be nothing gained and much risked by tightening standards for participation there. Carrite (talk) 14:08, 18 June 2020 (UTC)[reply]
Perhaps all that's needed is a guideline that says that it's acceptable for an experienced editor to move the banner to the talk page in cases where it doesn't seem likely the AFD will pass. Or perhaps BLP shouldn't have the banner at all? It's not hard to see how it can be percieved as an officialy endorsed slight against the BLP subject. ApLundell (talk) 15:02, 18 June 2020 (UTC)[reply]
RFC:DOY change to pending changes reviewer instructions
There has been an objection to the addition of this language on the grounds that it contradicts WP:RPC, specifically It is not necessary for you to ensure compliance with the content policies on neutral point of view, verifiability and original research before accepting.
Opinions (DOY change to pending changes reviewer instructions)
No - For a number of reasons. While verifiability is important, it is outside scope of pending change in general.
For example the purpose of protecting a pages with pending change is persistent vandalism, BLP violations, and copyright violations. In fact reviewing pending changes says It is not necessary for you to ensure compliance with the content policies on neutral point of view, verifiability and original research before accepting, but of course you are free to uphold them as you would normally with any edit you happen to notice. For example, in case of additions for which you can find no reference in the article but estimate unlikely to be vandalism, treat them as you would treat any such edit: do nothing, tag as needing citation, provide an appropriate citation, or revert – depending on the situation at hand. Reviewing pending changes is not to curate content or check sources which this new directive would require lest you have your review rights striped.
We could also look at pending changesThe process of reviewing is intended as a quick check to ensure edits don't contain vandalism, violations of the policy on living people, copyright violations, or other obviously inappropriate content and Acceptance of an edit by a reviewer is not an endorsement of the edit. It merely indicates that the edit has been checked for obvious problems as listed above which if reviewers are required to verify sources and content this would be a problem.
In the end I think it is not what pending change was setup to do. If a reviewer wishes to follow that guildline and either add a source or revert that is their choice, but it should not be something that reviewers are forced to do. Now I understand what they are going for, and verifiability is something that everyone should strive for but this is not the way to do it. I am also not comfortable making an exception for a specific niche. Generally I am not a fan of slippery slope arguments, but one could be made here. I think it is also telling when that guideline is not required for DOYs sister project WP:YEARS. PackMecEng (talk) 15:23, 13 June 2020 (UTC)[reply]
Yes. I view the change as requiring minor effort on the part of reviewers in the interest of improving day-of-year articles for readers. Checking for BLP violations or copyright asks more of reviewers than this check, a simple "no citation, don't accept" decision. I don't interpret the language as requiring reviewers to verify the information in the source, just to see that a source is included in the edit. The DOY articles have accumulated a lot of cruft over the years; I added cites to events on a DOY article and had to remove a number of the old entries because the target article didn't support the statement or the target article lacked a reference to support the statement. The effort to make those articles verifiable directly supports WP:5P. Editors are slogging through each of the DOY pages to verify each and every item; it's a slow, tedious process and is made even more difficult when more unsourced content keeps getting added. When a DOY article has pending changes protection, I think it's reasonable to ask that PC reviewers reject an edit that doesn't include a source. Schazjmd(talk)15:30, 13 June 2020 (UTC)[reply]
Yes - Normally, when people add a new entry without a citation and are told they should have added a citation, they say "Sorry, I didn't know that". But even with a warning message on each date page, contributors still do it because they haven't noticed the message. Occasionally, however, the response is "But X accepted the change, so it must be all right." When you go to the reviewer and point this out, they also respond "Sorry, I didn't know that." There are two points PackMecEng has made that need to be addressed. One is that the instruction to check for citations in a DOTY article (and other types of article will gradually follow) does not, in my opinion, go against the statement: "It is not necessary for you to ensure compliance with the content policies on neutral point of view, verifiability and original research", which applies to specific policies which are not at issue here. The second is that the reason all DOTY articles have the requirement for review in the first place is to protect them from vandalism, which typically consists of anons adding entries such as "10 July 1963 - my brother's birthday". Or "6 January 1975 - Lil Young Rapper, the greatest singer of all time". This happens every day and sometimes one article will have such an entry repeatedly inserted until the person doing it is blocked. There is little point protecting an article if reviewers decide that there is no need to prevent entries like these. I would have expected that reviewers would be grateful to be reminded about guidelines they hadn't known existed. Deb (talk) 15:38, 13 June 2020 (UTC)[reply]
No. - I am inclined to concur with PackMecEng. PC specifically refutes areas like this from reviewers' tasks, and verificability is fairly core to that. I also feel that stating that requiring this source check is just part of vandalism reduction - that's very dubious tying up. Reviewers would prevent edits like "my brother's birthday" etc etc, but filtering out vandalism like that does not require ensuring an inline source, because the categories "vandalism" and "does not have an inline source" only have a minority overlap. Nosebagbear (talk) 15:56, 13 June 2020 (UTC)[reply]
Yes - Like Deb, I have been a major contributor to the DOY articles and pending changes on them. She's summed it up nicely. There's an additional factor though. Those pages were cesspools of incorrect information without any sources. Publishing incorrect information is about the worst thing we can do here as an encyclopedia. In addition to the vandalism, we're trying to prevent more incorrect information being added while we clean the pages up. Like the vandalism that Deb referred to, hardly a day goes by without me or the other active members of the project finding incorrect, unsourced information being added to these pages. Asking change patrollers to help with this very problematic area is not a heavy lift. Nobody is forcing folks to patrol these pages, but if you do, please keep in mind that this is a known, chronic problem. If you're going to be active on these pages (like accepting pending changes), it would seem wise not to contribute to problem. Toddst1 (talk) 16:54, 13 June 2020 (UTC)[reply]
No. When I'm handling pending changes, I'll almost always do my best to verify the changes being made as well, but it shouldn't be a requirement - which is what this is making it. That fundamentally changes the role of pending change-protected articles, as PackMecEng quite rightly points out. There's no reason to make days of the year articles different in this regard to any other article, at least that I can see. Naypta ☺ | ✉ talk page | 17:20, 13 June 2020 (UTC)[reply]
The section you are quoting from WP:DOYCITE is a warning to the user and is accurate. It doesn't ask the reviewer to do anything so I don't see the relevance here. Deb (talk) 19:40, 14 June 2020 (UTC)[reply]
The relevence is the section from that you want to add to reviewers is asking them to comply with DOYCITE. To comply with DOYCITE you have to verify the source. Make sense now? PackMecEng (talk) 21:51, 14 June 2020 (UTC)[reply]
No, because that's not what the instruction says. It just asks reviewers to note that a citation is required, rather than accepting something that blatantly goes against DOYCITE because you feel it's too much work to look at the change before accepting it. Deb (talk) 07:24, 15 June 2020 (UTC)[reply]
Again, when it says "require a direct citation" I take that as it requires a direct citation. How do you know if it is a direct citation? You check the citation to see if it supports the info. You said "rather than accepting something that blatantly goes against DOYCITE" well how do you accept something that is not against DOYCITE if you do not check the source? You cannot. PackMecEng (talk) 16:18, 15 June 2020 (UTC)[reply]
Wikipedia is not a reliable source, yes. But the reliable source in that Wikipedia article can be used. Just accept that DOY pending change if it's verified in that article. If you don't like not having a direct reference, then copy and paste it from the article. {{reply to|Can I Log In}}'s talk page! 18:07, 14 June 2020 (UTC)[reply]
@Can I Log In: I think we may be saying the same thing. All the person who adds the event to the DOY page has to do is use a decent source. It can very easily be the source in the linked article, but it must be present on the DOY page. Toddst1 (talk) 17:34, 15 June 2020 (UTC)[reply]
@Can I Log In:, in that case, the DOY folks will just go and revert the addition as soon as its noticed being unsourced. Better off not accepting it - which is exactly what we're discussing. Toddst1 (talk) 18:52, 15 June 2020 (UTC)[reply]
Maybe I'm misunderstanding you, but you seem to be suggesting that reviewers do additional work by going to the linked article to check if there's a reference there. Wouldn't it be easier if you just checked that the new entry has a citation included? Deb (talk) 19:43, 14 June 2020 (UTC)[reply]
I'm sorry, but I don't understand what you are trying to say. It seems to me like you've got the wrong end of the stick. Deb (talk) 07:24, 15 June 2020 (UTC)[reply]
I was right, you have got the wrong end of the stick. You're opposing on the grounds that you are being asked to do something you are specifically not being asked to do.Deb (talk) 11:59, 17 June 2020 (UTC)[reply]
Up to semi-protection and be done with it seems to be the best solution. Pending changes almost always creates more work and controversy than needed, and I have never once seen a circumstance where it is superior to semi-protection in achieving anything. If these really are as bad as @Toddst1 and Deb: say the solution is semi-protection, not pending changes. The whole point of pending changes was that it would allow good faith contributors to contribute while protecting articles. It doesn't actually do that. It just creates more work for people. If you have non-confirmed user+PC reviewer+person reverting PC reviewer because they don't follow the citation guidelines, you're having three people do the work of one person. That's just idiotic. TonyBallioni (talk) 22:34, 14 June 2020 (UTC)[reply]
I totally agree. I'm constantly protecting individual articles for that very reason, but so far I've only gone up to a 3-month protection. Deb (talk) 07:24, 15 June 2020 (UTC)[reply]
I don't think continual standard SP is needed, but SP that runs for the week before and after (where we get the vast avalanche of twaddle) would avoid most of the issues. Nosebagbear (talk) 12:09, 17 June 2020 (UTC)[reply]
You probably haven't seen our always massive semi-protected edit request backlog. About 250 request are answered after 4 days (with about a backlog net change of +/−10 each day), and I think people still answer edit request wrong, and about 500 pending changes are accepted after 6 days, and about 20 of them are DOY, and yet the pending change backlog is fine. What major (dis)avantage does it have? {{reply to|Can I Log In}}'s talk page! 20:17, 17 June 2020 (UTC)[reply]
If you really want to know: it’s a nightmare from a technical angle. It will regularly make it so not even sysops can edit the page due to the way it stores potential edits. To edit in these circumstances you have to use twinkle to revert back to the last edit before it got stuck in pending changes, usually reverting several edits by established users actually improving the article in the process.It permanently stores the revision in the live history of the page, which on things like BLPs is an issue (DOY has this problem on recent deaths and birth years.) This may require either revdel or suppression.From a non-technical standpoint we give out PCR like candy (my standard for granting is if the person has a pulse), which means that ordinarily the people most active in reviewing are the least familiar with Wikipedia policies and guidelines since it’s functionally the first user right anyone gets because admins don’t care about it since all it does is restore to individuals the ability to answer an edit request. It also is easier to use than semi-protection so instead of getting rid of junk edits you instead create work that ordinarily wouldn’t exist. Anyway, most admins use pending changes rarely because of the issues with it, but if there’s a place where most of the pages are already under PC, they can just be switched to semi-protection. If there’s enough disruption to merit pending changes there’s enough to merit semi. If people don’t think there’s enough to merit semi-protection then PC likely should just be removed since it’s more difficult to deal with than actual live edits. TonyBallioni (talk) 00:57, 19 June 2020 (UTC)[reply]
No, the PendingChanges instructions should not be over-complicated by adding advice (not a guideline!) from a single group of editors about a small subset of articles. It might be good advice, but WikiProjects do not get to make up their own rules about what needs to be cited and what doesn't, and then impose their rules on the rest of the community. Unlike DOY's non-binding advice on what needs to be cited, the rule against WikiProjects setting rules for "their" content actually is one of our rules, written down in an actual guideline. WhatamIdoing (talk) 07:12, 18 June 2020 (UTC)[reply]
No - after reading this discussion, I think it's better if checking for a source were encouraged, but not required, for reviewers. I don't see checking for "a source" as being really helpful. It has to be a reliable source. Who cares if a sentence is unsourced or sourced to a blog? It's functionally the same. In fact, I think sourced to a blog is worse than unsourced, because it gives readers a false sense that the statement is properly sourced. So requiring reviewers to ensure there's a ref tag seems pointless, and requiring reviewers to verify there is an actual reliable source is too much. Hence, I think the best is to encourage, but not require, reviewers to verify sources, not just ensure that there is one. Levivich[dubious – discuss]21:14, 18 June 2020 (UTC)[reply]
Discussion (DOY change to pending changes reviewer instructions)
Oh dear, this is my fault. I added that to the page as a newish user after realizing all the edits I accepted were getting reverted, and asking Deb if it would be a good idea to add. A few months on and I realize it's not as simple as that. Some thoughts:
There's the issue of WP:LOCALCONSENSUS. The consensus that DOY pages require a citation was reached on the DOY wikiproject, not in a forum with a project-wide scope. I read ANI archives sometimes when I'm bored at work, and I believe I saw a case where someone requested a user be sanctioned for repeatedly adding unsourced entries to DOY pages, but the request was declined because WP:DOYCITE was only a local consensus. (Having trouble finding that case now though).
That said, it's my understanding that the consensus to require a citation (as well as to PC protect the pages) arose in response to a large amount of incorrect entries. While the PC guidelines say It is not necessary for you to ensure compliance with the content policies on... verifiability... before accepting, they also say reviewers should take special consideration of the reason given for protection and attempt to uphold it. We can accept unsourced content, but in this case, should we? Lots of vandals like fiddling with dates, lots of people get their info from dodgy sources that might have the wrong date, and an incorrect date for anything involving a living or recently deceased person is a BLP violation and forbidden by the PC guidelines.
There are a number of users who patrol DOY pages and revert unsourced entries or add sources if possible. So unsourced entries aren't going to stay there for long regardless, and removing the advice to PC reviewers just passes the work on to a handful of users from WP:DOY when it could have been dealt with right out of the gate.
Softening the language might be an option. "Required" is probably too strong for something with only local consensus. Perhaps something along the lines of Note that the Days of the Year WikiProject reached a consensus in 2017 that all new entries to days of the year pages should have an inline citation.
These are just some half-formed thoughts, I don't feel strongly enough to weigh in on the actual RfC at the moment. Spicy (talk) 15:53, 13 June 2020 (UTC)[reply]
While an optional part added to the "but good if you this etc" would certainly be more preferable, I'm a little concerned at the logical expansion from this. Local consensus (or even project consensus on things like MOS etc) requires quite a few things that could be handled by reviewers, but would start stacking up really heavily. Nosebagbear (talk) 15:58, 13 June 2020 (UTC)[reply]
I find the resistance a bit disheartening. Rejecting new entries to DOY without a citation (1) asks no extra effort from PCR, (2) is for the benefit of the encyclopedia and the readers, and (3) helps other editors who are working on the daunting task of making DOY articles verifiable. I accepted a number of pending changes on DOY without citations before I noticed the reverts; when I asked for an explanation and learned what the project was attempting to accomplish, I was happy to cooperate because it improves the encyclopedia. WP:NOTBURO, WP:IAR. Schazjmd(talk)14:41, 18 June 2020 (UTC)[reply]
* Each addition now requires a direct citation from a reliable source on this page supporting it. Simply providing a wikilink is not sufficient and additions without direct sources will be removed.
Leaving aside the fact that there is no March 32nd, it's quite straightforward. If it's a new entry, I would undo the change with an appropriate edit summary to inform the editor what is wrong with it. Deb (talk) 18:21, 18 June 2020 (UTC)[reply]
Same as Deb. If we weren't asking for inline citations for new entries, I would check that there's an article and that the specific information was cited to a reliable source, so in this example I would still reject the edit because YouTube is not a reliable source. Schazjmd(talk)18:28, 18 June 2020 (UTC)[reply]
SchazjmdFor demonstration purposes, you must pretend everything is true and complies with Wikipedia:Policies and guidelines. For failure to do so, you will not be sued of your life-savings.
John, a wikilink is not a reliable source. I would undo the edit to the DOY page with a note saying the entry needed a direct source. Some of us have gone through and cleaned up existing pages like May 11, but we have so much backlog in the other pages, most of the DOY participants are not cleaning up new additions that lack a source, just reverting them. Toddst1 (talk) 00:25, 19 June 2020 (UTC)[reply]
AGAIN, WE GET IT. A wikilink is not a reliable source. Have you noticed that in the wikilink, it contains let's pretend that it'sa reliable source? That means it's a verified statement. Either you could 1) accept the pending game; or 2) copy the reference over. Reverting it is not an option; it's the equivalent of vandalism, removing sourced content. {{reply to|Can I Log In}}'s talk page! 02:29, 19 June 2020 (UTC)[reply]
I fear that you DON'T get it. If it doesn't cite a source, it is not sourced content. If Napoleon's birth date was sourced in an article on the Duke of Wellington, you would still expect a citation in the Napoleon article, wouldn't you?Deb (talk) 11:31, 19 June 2020 (UTC)[reply]
No no no you don't get it. This is not a bureaucracy. If you do think so, then start fork of WIkipedia that is a bureaucracy. How many newcomers have you all bitten? A record ammount? It is possible that you have caused us to have less productive editors just by reverting on DOY articles. If there is a problem in a DOY article, then JUST FIX THE PROBLEM IF YOU CAN! * Doing a quick search for sources and adding a citation yourself. Since apparently Deb wants to be a John Quiñones, fine, I'll do this What Would You Do? scenario. So no citation for Napoleon's birthdate in that article, but there is one in Duke of Wellington. We'll unlike you all who are always on a reverting rampage, I would FIX THE PROBLEM and copy the reference over to Napoleon's article. Problem solved. {{reply to|Can I Log In}}'s talk page! 16:29, 19 June 2020 (UTC)[reply]
Let me explain this to you in small, manageable steps. Step 1: You agree that, in the circumstances described, the article on Napoleon should have a citation and that it would not be enough to link to another article and expect the reader to go there to check that it is correct. Step 2: You agree that citations should be added to DOTY articles for all entries, i.e. fixing the problem. This is what we are already working on. Step 3: You agree that this is a time-consuming task. Step 4: Can you now see why we would remove any new entries without citations that appear in DOTY articles while we are working on this? Deb (talk) 17:12, 19 June 2020 (UTC)[reply]
Or I do. It is providing a citation on behalf of who ever added it. Also you need a comma after WP:BURDEN. If you went to school, that is house cause and effect sentence strucutres work. And you used one colon too many. {{reply to|Can I Log In}}'s talk page! 18:36, 20 June 2020 (UTC)[reply]
We have a user right that allows non-admins to send mass messages via the software's MassMessage tool. However, I just ran into a situation in which a user had sent a mass message via their own script. Common sense would seem to say that the user right's existence implies that you need the user right to send any mass messages and can not just ignore the process by writing your own script. There is also a precedent in which a similar mass-messaging script was unanimously deleted at MfD. However, I can't find any actual written statement anywhere that says you're not actually allowed to do that. Is this stated somewhere? If not, should it be? Is this a common sense addition that needs to be made to WP:MMS or do we need to hold an RfA to clarify that users are not allowed to send mass messages in any capacity without the MMS user right? ~Swarm~{sting}22:45, 13 June 2020 (UTC)[reply]
Not that I've ever used it for multiple users, but I know what it does. Simply in the view history tab, select two revisions to compare, and then you can select any users with the selected revisions and in between can be mass messaged.
Unlike MediaWiki MMS, you just have to hope that all the users you want to mass message and in between your selected revisions, and just select the ones you want to message, ignore the other 99%.
I'm very sure the intention was not to bypass the mass message right, like how Twinkle was not intended to bypass the rollback right.
I'm not going to debate over this, but I'll just say, see how it exactly works before calling it something, just like how you shouldn't call someone stupid because the "look" stupid. {{reply to|Can I Log In}}'s talk page! 23:51, 13 June 2020 (UTC)[reply]
I think you're making a problem out of where there is none... This is supposed to be a help in fighting vandalism (much like twinkle, which also semi-automates this); and this particular feature seems to be about giving messages to more than one user who are engaging in edit warring. If it were not done with this, I assume (and I remember doing so) it would be done manually, which is just not helpful when you're dealing with RC patrol... See WP:CREEP. In any case, if you have doubts with this, I suggest you politely ask the script's creator to pass this through the bot-approval policy. RandomCanadian (talk / contribs) 00:08, 14 June 2020 (UTC)[reply]
There is a problem. No matter how big or small you consider this a problem, it was considered, a problem.
Script creator here - only just noted this thread. As Can I Log In and RandomCanadian have said, this tool is not intended to replace the mass message sender right - rate-limiting still exists. The goal of the multiple action tool is to warn multiple users/IPs/socks ext. who have engaged in a vandalism war or other issue (such as notifying good faith pending changes reverts, as if a user didn't check, they might not know what they did wrong). Previously, I'd have to open up the tabs of about 50 users all manually going through and warning them, which is very time-consuming. Perhaps the issue could be mitigated by applying an artificial limit of a small number of users, and only relieving this limit for mass message senders? There is no direct way to add users to the tool at the moment other than via a history page or by reverting a pending change (max users in past 10 revisions). No normal user can access the MAT tool with a custom list. As for why I used the MAT to send out a message, all the users there also have the script installed, and I can use software to determine these users to check for bugs ext. I could easily have sent a message through RedWarn itself, however, that would've caused disruption to the users of the tool via a pop-up dialog, so I have used talk page messages for some time. However, from now on I'll ensure to find an alternative method. Ed6767talk!00:41, 14 June 2020 (UTC)[reply]
TonyBallioni, probably will do that, but a built-in inbox may be useful too, so I'll have a think about it. But I definitely won't send out messages to users via a talk page note anymore. Ed6767talk!00:52, 14 June 2020 (UTC)[reply]
This is not about that tool or the user behind it. Someone pointed out that a user had sent mass-messages via script, as can be viewed here here. Nowhere am I attempting to "go after" the user involved or "make a problem" about that incident, I was just explaining that I ran into that situation at PERM and that's the reason I'm bringing this up. I'm simply asking for the community's feedback on the general policy issue as to whether users are allowed to send mass messages via a script without the Mass Message Sender user right. The merits of the tool are completely irrelevant to that question. ~Swarm~{sting}02:13, 14 June 2020 (UTC)[reply]
Swarm, I don’t think it’s ever been documented anywhere, but with the exception of twinkle/rollback, the general consensus whenever it comes up (one or twice a year) is that scripts should not be used to circumvent a permission the community has in place to restrict access to certain features on the MediaWiki software. This is especially the case for bot-like editing. The MfD you cited on this last year was an example for MMS, there’s the case with the super revert script at ANI a while back (tl;dr basically used rollback to block) and then someone who recently reverse engineered JWB to get around the check page. I don’t think it’s documented anywhere because we almost always deal with it ad hoc, but I think we have enough “precedent” to say there’s a general consensus against using scripts to replicate user rights. TonyBallioni (talk) 02:27, 14 June 2020 (UTC)[reply]
So we can circumvent the rollback right all we want. Yup we got 2 user scripts to do that. Next, to remove the rollback user group and bundle it with the (auto)confirmed user group. Let's just retry this in 5 years. Epic Genius (talk) 03:23, 9 February 2015 (UTC). It's been 5 years and 4 months. You can revert without explanation with rollback-like scripts, particularly TW rollback vandal, but you are not allowed to rollback with generic rollback. There is zero difference, yet its treated differently even though it's the exact same thing. {{reply to|Can I Log In}}'s talk page! 03:37, 14 June 2020 (UTC)[reply]
Rollback’s primary use these days is to prevent people from getting Huggle who shouldn’t have it. If you tried to copy Huggle’s code, alter it to allow you to use it without rollback, and then proceed to use it, you’d pretty much be guaranteed a block. Also, as an aside, all caps is usually considered rude on the internet, even in edit summaries. TonyBallioni (talk) 03:49, 14 June 2020 (UTC)[reply]
There is clearly a lack of clarity and consensus regarding user scripts and their relationship to user permissions, and IMO, (pseudo-)rollback and MMS rights are not comparable. A rollback can easily be reverted via the page history, meanwhile, mass messaging would either require the tedious job of an editor to go into every single talk page that was edited and manually remove the message or the use of another "bot-like" script to remove the message. Ed6767talk!13:39, 14 June 2020 (UTC)[reply]
Per isaacl above, I'd not be in favour of adding a specific line about this to policy anywhere, because there's no need for it. If a BRFA passes for a bot or script to do MMS-style messaging, then it ought to be permitted so to do. If there's no BRFA, it's prohibited already by the bot policy. Naypta ☺ | ✉ talk page | 13:47, 14 June 2020 (UTC)[reply]
So some general comments: there really is nothing specifically wrong with someone using scripts to help them make edits that wouldn't be considered problematic if they were made without a script. And arguments about a particular script are generally weak when other scripts can do the same thing (e.g. messages can be sent with things like AWB as well). Where this generally starts to become a problem is when the edits disrupt things - such as by causing impact to watchlists/recent changes/etc. Using MMS for large distributions avoids much as this, as the edits are flagged as 'bot' by the utility, indicating that they don't need to be reviewed as much. For other technical reasons, it is also preferable to use MMS for large distributions than to use a bot account to make lots and lots of direct edits. — xaosfluxTalk16:12, 14 June 2020 (UTC)[reply]
Another advantage of using the mass messaging bot is that it has an opt-out mechanism for users who don't want any mass messages. I was going to elaborate that scripts that just make it a bit more convenient to make edits that you can do manually aren't generally a concern (the bot policy alludes to this). I agree that the potential for disruption is a key consideration in weighing if a script needs approval. isaacl (talk) 17:00, 14 June 2020 (UTC)[reply]
What would be seen as a mass message? Sending a message to 10 vandals would probably not be considered a "mass message"; sending the same thing to 500 WikiProject members, many of which may not be interested in this, would be considered an overriding of the MMS policy. I remember this deletion discussion about an MMS script, so MMS tools, at least unrestricted ones, should not be allowed. There might be a policy breach if the MMS tool is sending more than ~50 arbitrary notices. However, 20 vandalism notes? I'd call that acceptable.
In any case the Multiple Action Tool, when used on a hist page to notify vandals, is probably acceptable. I believe that users should not be allowed to notify more than 50 users; if possible, the tool should direct such requests to WT:MMS or, if the user has massmessage rights, create a mailing list for them. Eumat114 formerly TLOM (Nachricht) 06:08, 15 June 2020 (UTC)[reply]
There isn't a hard cutoff. As Ed6767 alluded to, a key consideration is how much work you would be giving someone else to do to reverse the effect of any problems that occurred. isaacl (talk) 15:10, 15 June 2020 (UTC)[reply]
I agree with Xaosflux. IMO writing down a "rule" for something that's discussed maybe twice a year – and produces genuine complaints from actually-affected editors less often than that – would be WP:CREEPy. If it's used sensibly, to send relevant messages to small-ish numbers of talk pages, I think it's okay. It should not be wrong to do with a script what's both fine and physically feasible without a script. WhatamIdoing (talk) 06:56, 18 June 2020 (UTC)[reply]
Have been looking at this thread carefully for a bit to think of a good reply.
I have been testing (well at least have installed) RedWarn.
Once again, from my impression, sending mass messages is something that may or may not be a good idea. I have not cared so much for this, but maybe writing the message once on a talk page and pinging everyone involved may be a better approach because the message can be edited to filter errors out by the poster.
I see little point in anyone sending out messages to a large group of people. It may be perceived as spamming, which is not necessarily good. I remember a year or two ago the Arbitration Committee sent out a mass message to admins about securing accounts in error.
About the "Multiple action tool", I have not used it before. I see good reasons to send COI notices to many people, and I have before. But usually, those groups of people are so small that there may be little point to automate it entirely.
May be going on a tangent here, but I think the goal of RedWarn is to be an alternative to WP:TW, but I think Twinkle is never going to go away. Sure, there are some things that Twinkle can do that RW can't do, and vice versa. RedWarn is also pretty much "beta"; features may vary or may show up in places that they should not show up in.
I can argue about sending mass messages with scripts that it is extremely easy to do that with auto wiki browser as well. But once again, there is very little point in it, and it has a lot of potential for spam.
There is inherently a throttle limit in how many actions you can do in a given timeframe; I have tripped it on Wikipedia and on wikiHow before. So my opinion is, no, I do not think we need to automate sending the same message to a bunch of people with the multiple action tool on the English Wikipedia. I don't think it is super useful on any wiki at all, unless if you are sending out a newsletter. There are very few times when you have to do that anyway. Aasim06:59, 17 June 2020 (UTC)[reply]
I can imagine sending out a COI notice to ~10 people, in which case this tool is probably more convenient than opening each talk page and using Twinkle. But @Ed6767: is there a mechanism that enforces the user group restriction (e.g. a non-autoconfirmed user cannot use RW, and a non-XCON user is technically unable to use the MAT? Cheers, Eumat114 formerly TLOM (Nachricht) 01:07, 19 June 2020 (UTC)[reply]
Eumat114, yes (see here for MAT, and here for the script as a whole) - as it's not hosted on Wiki, removing this restriction would be harder than a normal user script, requiring either a custom modified version of RedWarn or an addition user script that overwrites feature restrictions. For normal users without technical knowledge, bypassing the feature restrictions is a non-issue. If more restriction was required, an edit filter could easily be made to restrict MAT actions to extended confirmed users on Wikipedia's side too. Ed6767talk!01:18, 19 June 2020 (UTC)[reply]
(Forgive me if I'm not doing this in exactly the right place; I'm a long-time editor who has somehow never waded into the Pump yet, to the best of my recollection.)
The InternetArchiveBot, perhaps best known for adding archive links for references, has in recent months been aggressively linking mentions of books to copies of those books stored at the Internet Archive. The Internet Archive is widely viewed within the publishing industry as a pirate site when it comes to under-copyright books, and they are currently facing a lawsuit from several of major book publishers over that fact. If we accept the publishers' claims over rights as at least potentially valid, this puts the bot in violation of WP:COPYVIOEL, which tells us "If there is reason to believe that a website has a copy of a work in violation of its copyright, do not link to it."
Additionally, the bot is linking not just in references, but in the body of text (such as this example, which goes against "Wikipedia articles may include links to web pages outside Wikipedia (external links), but they should not normally be placed in the body of an article." (WP:EL)
I recommend that InternetArchiveBot (talk·contribs) be halted and not allowed to run until it is changed to no longer link under-copyright works, or until such time that the ending of the current case deems it to not be a copyright-infringing site. Cyberpower678 (talk·contribs), who is operating the bot as a paid agent of the Internet Archive, should be requested to undo all the edits that have been inserting such links (which I expect can be done largely on an automated basis.) To the degree that it does link books beyond that, the bot should be restricted to doing so within ref tags or in the external links section.
(And to make any concerns about any conflicts that I have clear, yes, I am both a writer and a publisher. I am not currently involved directly in any lawsuit, but the lawsuit does include publishers who have published my works in the past or who have bought out such publishers and may still hold publishing rights, but not for anything that has generated royalties for me in a long time.) --Nat Gertler (talk) 14:52, 14 June 2020 (UTC)[reply]
Among the five people involved in that discussion (counting yourself, a paid agent of the IA), I see absolutely zero consideration to the concern raised about how this comports with our policies regarding copyright. Its approval was based on the claim that "this task is uncontroversial", but clearly it is controversial - if not before the lawsuit, then certainly now. --Nat Gertler (talk) 15:38, 14 June 2020 (UTC)[reply]
Discussion of the legal minutiae can be reserved for Number One Court. This discussion related to how it affects the English Wikipedia. ——Serial#20:33, 16 June 2020 (UTC)[reply]
The following discussion has been closed. Please do not modify it.
Regarding the lawsuit.. it only concerns Controlled Digital Lending which is the lending of a complete book on a 1 to 1 basis. The ability to view limited pages, as being linked here, was settled by Google Books in Authors Guild, Inc. v. Google, Inc.. The publishers lost, and most likely they will loose their case against CDL, also. But it won't be known for many years and in the mean time the legality of CDL is well supported by many reliable legal scholars and institutions. -- GreenC15:48, 14 June 2020 (UTC)[reply]
The pages it links to include the link for "This book can be borrowed for 14 days," so yes, it's taking people to a "Controlled Digital Lending" page, even if it's not directly downloading the book immediately. In the meanwhile, I am going to ping the other people who were involved in the approval discussion.(Jo-Jo Eumerus—TheSandDoctor—Galobtter—GreenC) --Nat Gertler (talk) 16:02, 14 June 2020 (UTC)[reply]
There is an ongoing legal case about CDL, just as there was an ongoing case against Google, during which time we continued linking to Google Books. -- GreenC16:26, 14 June 2020 (UTC)[reply]
Was there substantial discussion about doing so? (I'm honestly asking... although even if so, it's still a bit of a WP:OSE situation.) Should we be rewriting WP:COPYVIOEL so that we're only discouraged from linking to things that have already been proven in court to infringe and yet are still on the web anyway? --Nat Gertler (talk) 16:37, 14 June 2020 (UTC)[reply]
I'm not aware of anyone arguing to remove Google Book links during the years Authors Guild, Inc. v. Google, Inc. was ongoing (2005-2015). And good thing we didn't or it would have been a huge loss to the community and our readers. Internet Archive is a registered public library operating according to a legal position created by legal scholars. No court or judge has determined CDL is illegal or requested it be taken down while the case is ongoing, Internet Archive owns the books it lends on a 1:1 basis (1 book owned = 1 book lent) just like any library. -- GreenC16:59, 14 June 2020 (UTC)[reply]
Yes, the IA has lawyers making its case, and did so even when they took actions that they have since pedaled back from. The publishers also have lawyers who have put forth a legal position. We have never, to the best of my knowledge, waited for a court ruling of infringement before removing links to material that looks to be of legitimate concern. The IA does not work "just like any library", no; libraries are generally loaning the actual items that they purchased or licensed, and not reformatted digital editions that they made without permission of relevant rightsholders. You may think that the IA should emerge victorious in this case, but that is not a given. --Nat Gertler (talk) 17:13, 14 June 2020 (UTC)[reply]
Internet Archive have not "peddled back" CDL .. your thinking of the National Emergency Library (NEL) closing early. NEL is not being challenged by the publishers. IA owns and purchased the physical books in most cases. For the purpose of providing limited page views it is legal to scan books as determined in the Google case and verified by the Supreme Court. The only issue here is CDL and while we may not known the outcome, the question is what should Wikipedia do. Foremost, no judge or court has issued an injunction against Internet Archive despite there being an active case looking at it. Given the precedent of what we did with Google Books during Authors Guild, Inc. v. Google, Inc. we made the right decision [by default] by keeping the links in place, it would have been harmful to Wikipedia to actively remove 100s of thousands of book links only to re-add in the future with all the disruption and errors that would cause plus lack of links to RS helping with V for a lost decade. -- GreenC18:26, 14 June 2020 (UTC)[reply]
NEL is not being challenged by the publishers.??? From the lawsuit: "NATURE OF THE ACTION 1. Plaintiffs Hachette, HarperCollins, Penguin Random House, and Wiley (collectively, “Plaintiffs” or “Publishers”) bring this copyright infringement action against IA in connection with website operations it markets to the public as “Open Library” and/or “National Emergency Library.”" It would be inappropriate to leave the links in place only to have to remove them if and when IA loses this case, having facilitated piracy in the meantime. Links that have been added by bot can be readded by bot, should IA prevail. --Nat Gertler (talk) 18:37, 14 June 2020 (UTC)[reply]
I was misinformed, struck. It is bad faith to call it piracy the legal interpretation is still open, and it wouldn't be piracy after the fact unless they operated in defiance of the law. Linking to a snippit view is completely legal, you are conflating CDL with snippit view and arguing for total removal of links despite snippit view being legal. Also see the RFC it was decided by the community not "five people". -- GreenC19:28, 14 June 2020 (UTC)[reply]
Fair 'nuff; I was citing the discussion you had pointed to earlier, not to this other discussion you now point to. However, I do not see in that discussion the voicing of the concerns I've raised. My calling it "piracy" was specifically in the context of "if and when IA loses this case", i.e., if IA has been judged to have infringed; it may not be the technical legal term, but it is an understood one. I an not conflating "snippit" view with CDL; I am noting that taking them to the snippit page also takes them to the page with a link to follow to download, as it is one and the same page. --Nat Gertler (talk) 19:53, 14 June 2020 (UTC)[reply]
CDL requires signing up for an account, logging-in and the book is available on the shelf. This is why we were requested to include |url-access= in citations because of the entry barrier. There is no downloading, it's an in-browser flip-book of a scanned facsimile. -- GreenC21:06, 14 June 2020 (UTC)[reply]
Yes, for people who are already logged in, it has a link to take you to the book, and for people who are not already logged in, it has links through the process to get you to the book, encouraged by the statement that "This book can be borrowed for 14 days," --Nat Gertler (talk)
What may have precipitated the court case is that during the pandemic "Controlled Digital Lending" was switched to "Uncontrolled Digital Lending" – one printed copy could be lent out digitally to any number of people simultaneously. In the last few days the control has been restored. The removal of control was justified by IA on ethical grounds (people could not go into libraries to borrow books) and not legal grounds. Their legal ground seems to me very shaky or nonexistent and they can be held liable retrospectively. CDL is another matter.[8]Thincat (talk) 17:44, 14 June 2020 (UTC)[reply]
Leaving aside for the moment the legalities of this case, do we need online links to references in any case? If anyone is interested enough to look they can look for the cited text online or in a physical library as they please. I have seen many cases where links are provided to Google Books and the references have been removed because the particular page was not linked at the time that the remover looked, as if the reference was to the link rather than the book itself. Wikipedia should aspire to be more that just what is available on the Internet, and part of that is to educate readers that citations to academic books and papers are valid even if they are not freely available online at all times. Phil Bridger (talk) 17:34, 14 June 2020 (UTC)[reply]
And if the goal is to allow verification of the existence of the book and such publishing particulars as publication date, author, publisher, etc., there is always the fine service of WorldCat. --Nat Gertler (talk) 17:53, 14 June 2020 (UTC)[reply]
You are saying we should force people to use a physical public library by intentionally removing links to online digital books so that they learn a lesson not everything is available online? -- GreenC18:37, 14 June 2020 (UTC)[reply]
Removing links to sources that are (currently) available online does not force people who have internet access to go to a physical library. It may force them to do some online research to find the available online copy of the book. BD2412T18:52, 14 June 2020 (UTC)[reply]
Yes, @GreenC:, your name is there. Given that this discussion is clearly leaning toward disapproval of the bot (particularly among editors not being paid by IA), you should probably pause the bot's activities until when and if a consensus is reach otherwise. --Nat Gertler (talk) 00:39, 15 June 2020 (UTC)[reply]
As long as right now the links are not considered illegal (by the unresolved), our act of linking to them is legal. (I suspect if Wikipedia was based in Europe, this would change greatly due to the newer copyright directive there). So now the question is more the morality of linking to something that may or may not be legal, and which has a possible effect on author some who are editors on WP. I would tend to argue that because we avoid links to sites like scribd, academia, researchgate, etc. all which have "questionable" uses of published materials, not yet proven out in any court, we probably should take the same stance here until at least we have a preliminary decision from the District Court. The current page one gets to via the magic linking of ISBN or similar numbers is fine for current purposes until we are satisfied Open Library is working within US law (and yes, assuming its legal which might disappoint authors as the Google cases had as well, we'd still want to use that resource). --Masem (t) 20:17, 14 June 2020 (UTC)[reply]
Agree with Masem, we already apply the precautionary principle in a number of areas. While the WMF is not going to be held liable in any manner, should a court case go in a certain direction, both editors (who enable the links to the material) and the readers (who follow the links and download the material) could be in for legal troubles. Only in death does duty end (talk) 20:32, 14 June 2020 (UTC)[reply]
Is VPP really the appropriate place to discuss editors' and readers' potential legal liability related to linking to copyrighted material hosted outside of WMF servers? If we have copyright-related legal questions, we should probably request assistance from WMF legal staff. – Jonesey95 (talk) 22:24, 14 June 2020 (UTC)[reply]
Yes, copyright issues fall under policy. For example, we can't linkj to known copyvios outside WP. That's not saying this is the case now, but it is a concern. --Masem (t) 01:36, 15 June 2020 (UTC)[reply]
I already refuse to place links to Google Books in citations because they are a commercial service and we should not give them preference over other providers of books such as Amazon. See the fever hospital DYK for a recent example. What this bot is doing seems to be link spam because it is promoting yet another rival book provider. And the fact that we have paid editors here lobbying for this is outrageous. This matter is certainly controversial. Andrew🐉(talk) 22:44, 14 June 2020 (UTC)[reply]
Perhaps you want to wake up to the fact that IA is also using books that have been scanned by Google, so your refusal there is a bit selective. Eissink (talk) 01:20, 16 June 2020 (UTC).[reply]
What is selective about an editor saying that he will link to neither Google Books nor the Internet Archive? This is treating online sources even-handedly. Linking ISBNs, which links to Special:BookSources, was decided on many moons ago as a neutral way of providing people with the ability to check the content of cited books. If the copyright status is deemed acceptable then the Internet Archive can be included as one of the many sources there. Phil Bridger (talk) 20:20, 16 June 2020 (UTC)[reply]
I apologize, I see now that I have interpreted Andrew incorrect. I totally misread here, thinking that he would only link to IA. I'm sorry. Thanks, Eissink (talk) 22:15, 16 June 2020 (UTC).[reply]
I agree with Masem: while we haven't done anything wrong, we normally err on the side of caution (WP:ELNEVER), and should discontinue the bot linking until the situation is clear. I did raise the issue when the bot run was first approved (as did Masem). I wouldn't be prepared to say whether WMF would be held liable given recent developments in the United States, but this isn't our concern. Hawkeye7(discuss)01:11, 15 June 2020 (UTC)[reply]
Even if we err on the side of caution, there's nothing to err on here - the copyright issue that's being sued over directly relates to the National Emergency Library, not with the actual lending process itself, or the storing of the material digitally. It would make absolutely no sense to stop the bot from linking based on the fact there's a lawsuit pending that's not based on the storage of the material, but rather an action taken by the company during COVID which doesn't change the underlying copyright of the material, but rather would be a violation of the distribution right of the material by the company, which has no impact on us at all. SportingFlyerT·C01:35, 15 June 2020 (UTC)[reply]
That's simply not true. The lawsuit covers the entire "Open Library" project, including the creation and storing of the digital editions ("2. Defendant IA is engaged in willful mass copyright infringement. Without any license or any payment to authors or publishers, IA scans print books, uploads these illegally scanned books to its servers, and distributes verbatim digital copies of the books in whole via public-facing websites. ") --Nat Gertler (talk) 01:46, 15 June 2020 (UTC)[reply]
IANAL but am a person with interest in how law and the Internet interest, and did read through the suit. Keeping in mind Authors Guild v Google , "uploading scans of books to their servers" is considered a possible fair use (though I beleive this suit is outside the Second Circuit so its not clear if that ruling might apply or not), but the complication is the argument that because they have physical copies of those books, they have the right to share a DRM-tied electronic copy of those scans in a lending manner. Part of he suit does content that if you use the web-based interface for your loan, you can then make screencaps of the webbrowser and effectively make a copy of the book that way. The lawsuit documentation is set up to make assertions they believe are truthful (in what OL is doing wrong) but the judge make strike some of those statements. So just because the suit claims it is illegal doesn't mean it is illegal, yet. --Masem (t) 02:33, 15 June 2020 (UTC)[reply]
It's a Second Circuit case, the new one. I'm satisfied the Google case holds until otherwise shown fair use doesn't apply. (As a tangent, if I go to the library and check a book out, I can photocopy or scan the physical copy. And if I scan a physical copy of a book, someone has purchased it at some point in time.) SportingFlyerT·C03:54, 15 June 2020 (UTC)[reply]
Well, keep in mind: with the Google Books, part of the fair use determine was the display of snippet view as a key feature; further in the secondary case that followed with HathiTrust, it was also adding cases of making the whole OCR'd text available to people with disabilities (accessibility) , and for giving copies to academic faculty within the libraries systems that lost their old work but could prove they owned it at one point. The test of whether first sale doctrine applied to fair use is the big question here. --Masem (t) 04:03, 15 June 2020 (UTC)[reply]
If SportingFlyer scans an entire library book then that's an elementary breach of copyright as the entire point of copyright is stop wholesale copying of this kind. Google got away with it because they are using the copies to provide an indexing and search service and the settlement decided that this was fair use. Google does not provide the complete text of the books, if they are in copyright, otherwise no-one would ever need to buy a book again. Andrew🐉(talk) 22:56, 16 June 2020 (UTC)[reply]
The bot should continue linking. All of the concerns raised here appear to be speculative. We shouldn't link to known copyvio, but that's not the case here. I do not work for Internet Archive. Levivich[dubious – discuss]01:41, 15 June 2020 (UTC)[reply]
An important thing to consider is that it's not so much about protecting ourselves (or protecting the writers and publishers) as protecting our downstream users. We can basically defend quite well against any lawsuit that might arise -- we're rich. Our downstream users aren't, and its for that reason that they're possible targets for any unpleasantness.
We basically pledge "You may use our material in your own work, with confidence that you won't be called out for violating anyone's copyrights -- most of the material is either public domain or under a free license, and that that isn't conforms to accepted allowance under fair use; at least, we aim for this and expend much effort on ensuring that it's true". So I mean if there's a question, if there's active litigation... I don't see how we can make that promise and stand behind it, if we include this material. In theory at least we could be endangering our trusting downstream users. Definitely want to be conservative on that account. Herostratus (talk) 01:58, 15 June 2020 (UTC)[reply]
That's why I want to stress that right now the choice if we keep linking should be a moral/ethical one - we know authors don't like that site and some are editors on here, and they've asked us to stop, nicely (no legal threat here). Should we stop? That's the consideration here. --Masem (t) 02:36, 15 June 2020 (UTC)[reply]
Oh, absolutely. The moral/ethical angle is paramount in any decision, in Wikipedia no less than real life.
So... about the authors (and publishers). I mean, I'm not a huge copyright fan, but... sending an e-mail (or text) is so different, so easier, than writing and addressing and stamping and sending a physical letter that they are different things. If you're talking to a colleague and she says "I'll send you Finster's phone number as soon as I get to the office", wouldn't you be surprised to get it in the mail three days later? They're different; they just are, is all.
Similarly, is not possible that going to a website and clicking a button to download a copy of a book is a lot easier than requesting an inter-library loan (for most books), and waiting for the book to be available and shipped to your library, and slogging down to the library, and reading it there or checking it out (and slogging back to return it later)? So much easier that maybe -- maybe -- they are different things? We know that libraries lending physical books is not something that disrupts the author-publisher-bookseller ecosystem to to the degree where that system can't handle it. Can we be certain that an online version of the library paradigm won't?
If the complaints and lawsuits are frivolous, or maliciously self-serving, that's one thing. Are they? Or are authors and publishers feeling real pain, or have reasonable cause to see a realistic prospect of feeling real pain down the road if we continue down this path? They seem to be saying they are. To prove they're wrong you'd have to see the future. Can you? I can't, and I don't want us to be a party to this until we are a lot more sure that we're on the right side here. Herostratus (talk) 09:59, 15 June 2020 (UTC)[reply]
Short version: No they are not frivolous, but they are by no means completely valid either. The argument they are a library had some weight when they were lending books out one at a time. Likewise the argument that as brick and mortar libraries are closed, they need to suspend the one-at-a-time process is a decent excuse as far as it goes. Whats going to be telling is what happens once libraries are open again. Will they go back to one-at-a-time? Will they wait for the court cases to be over? They have used the covid situation to justify their current practices, but the key point of that is they are using the current situation to justify a practice that they know wouldnt fly in ordinary circumstances. Otherwise they would have done it that way all along. This is very much a fishing expedition on their part, and publishers and authors *have* to fight it, lest they be accused of approving of it by not contesting it. Only in death does duty end (talk) 10:16, 15 June 2020 (UTC)[reply]
Extremely Strong oppose As someone who actually knows a lot about copyright and specializes in keeping Wikipedia free of copyrighted text, the Internet Archive; ESPICALLY the books and websites they host, is essential to keeping the copyrights of poorer, less known people safe. The offsite campaign to slander them is Clout chasing nonsense under the guise of "making sure writers/publishers get their money" and should be completely ignored- although I do not fault people for being tricked into supporting this terrible campaign. Moneytrees🌴Talk🌲Help out at CCI!02:59, 15 June 2020 (UTC)[reply]
Got it - let's just ignore writers, publishers, and their concerns about their rights being violated. They (or in my case, we) can't possibly know what they're talking about or what they're going through here; let's just ignore Wikipedia guidelines to stick it to 'em. <sigh> --Nat Gertler (talk) 05:55, 16 June 2020 (UTC)[reply]
I have raised concerns about the mass removal of Google Books links in favor of Internet Archive in February, initially only because Google Books is so much easier to navigate (just scroll). And in the case that brought it to my attention, the book linked to on IA was a copy (a ripoff?) of the book scanned by Google! (Compare this book on Google to this book on IA.) It's a shame that If it ain't broke, don't fix it isn't official policy, because that could stop such nonsens immediately. Nevertheless, I decided to ask what was going on, but to my utter horror, I engaged a terrible attitude of the – paid – bot operator. I have summarized things, short, on Wikipedia_talk:Google_Books_and_Wikipedia#Unbalanced_opinion_and_faulty_arguments, where you'll find more complaints. You will also find my remarks on an IA bot on Wikimedia Commons, performing very questionable edits. And if you think this is all: it isn't. I just saw on the Dutch Wikipedia the discussion on an also very questionable IA bot that replaces working links by the dozens and tagging them as dead while replacing them for an IA link – that bot has been stopped earlier today. For those of you that think WMF might play a roll in resolving possible legal issues surrounding the linking the IA: I don't want to not AGF, but I cannot find any reason for the massive paid edits by IA bots other than that some institution is paying IA for their services, because I think they would otherwise use their resources on different things than on saving Flickr pages, the images of which have already been downloaded to Commons (and of course have the license checked already) – could it be that there is some sort of contract between WMF and IA? If so, it might be in good faith, but things seem to get rather messy at the moment. Greeting, Eissink (talk) 02:07, 16 June 2020 (UTC).[reply]
comment - it's difficult to know everything about copyright - but I don't see how this effects Wikipedia at all. I can't see how we would need to do anything until a verdict was reached. It does seem like the case of "I support this lawsuit (and it potentially effects me off wiki if I'm reading right), so Wikipedia should also support this." Let me know if I'm way off the mark here. Best Wishes, Lee Vilenski(talk • contribs)07:18, 16 June 2020 (UTC)[reply]
Specifically: Wikipedia has a policy (WP:COPYLINK) which prohibits linking to material which is being made available in violation of copyright. The situation with the current 'library/not a library' court case is that publishers and authors are asserting the archive is hosting (and sharing freely) their books digitally in violation of copyright. The argument above is that some editors think as a precaution we shouldnt link to material that is currently being disputed in court until its settled, and other editors think we dont need to worry and should. There is also the added issue that in any situation involving copyright, there is a third group (often the more extreme end of the 'Free knowledge' activists) who think even if it does violate copyright we should do it anyway. Only in death does duty end (talk) 07:40, 16 June 2020 (UTC)[reply]
Do we actually have a policy on if we should change our policies due to something being in litigation? Surely we shouldn't touch it until a result is levied? Best Wishes, Lee Vilenski(talk • contribs)08:10, 16 June 2020 (UTC)[reply]
WP:COPYLINK is our policy on it (WP:ELNEVER is also a relevant guideline). As to the wider question of 'should we do something when its status is in doubt'. No we dont have a firm policy on it. Many adhere to the Precautionary principle. Many do not. Some people like myself adhere to the precautionary principle dependant on the risk and the liability. Here the risk is if a court case goes in a particular way, and we have been deliberately linking to illegally copyrighted materials, we could be held responsible for Contributory copyright infringement - this is mentioned in ELNEVER. The key part of this though, and personally why I favour the precautionary principle, is that the WMF is protected against legal liability in almost every situation under Section 230 of the Communications Decency Act. Individual editors are not. And in fact the WMF is required, and has done in the past, provide details of its users/editors in order that they can be held liable in a court. If it was a case of 'oh we can do it until the court case is settled' and it would be the WMF who would be sued, then I would say fuck it, they have enough money to hire expensive lawyers. As it stands, I am not willing to potentially throw another editor under the bus because of a third party website's ambiguous legal nature. Only in death does duty end (talk) 08:22, 16 June 2020 (UTC)[reply]
@Only in death: (edit conflict) Even assuming the maximum, that both CDL and NEL are found illegal after final appeal, it's incredibly unlikely that they'd go after an individual Wikipedia editor who added links to IA in good faith. This would be a PR nightmare. There is no way that will happen. Large corps like JSTOR show deference to Wikipedia and its editors for their volunteer work... there's just no way that they'd sue any one of us for actions taken before the resolution of the suit. Psiĥedelisto (talk • contribs) please alwaysping!08:33, 16 June 2020 (UTC)[reply]
See my below comment. I agree they would be unlikely to go after individual random editors, however they may decide to go after individuals who have been paid. RE JSTOR - you are barking up the wrong tree with that one if you dont think they will seek legal redress if you start sharing their materials illegally. Only in death does duty end (talk) 08:42, 16 June 2020 (UTC)[reply]
I am indeed familiar with the tragedy of Aaron Schwartz. All I meant to imply is that JSTOR allows our editors to access its materials for free with proper verification. So, given the public relations challenge of lodging cases against volunteers for linking to works they own before it was unquestionably illegal, other publishers are very likely to show amnesty in the same way. Psiĥedelisto (talk • contribs) please alwaysping!08:51, 16 June 2020 (UTC)[reply]
Although to be fair, even in a worst case scenario, the editors pursued would be paid editors of the archive in question who is alleged to be violating copyright. So I wont massively lose any sleep over it. Only in death does duty end (talk) 08:30, 16 June 2020 (UTC)[reply]
Our policy on linking to copyrighted works says "However, if you know or reasonably suspect that an external Web site is carrying a work in violation of the creator's copyright, do not link to that copy of the work." (Emphasis mine.) That's the realm we're in, reasonable suspicion... and that's the realm we're almost always removing such links in. We are not waiting for some legal proof of it; if we see a link to a scan of a book that isn't on the author's or publisher's website, that's reasonably suspect. (Indeed, if we waited until the results of court cases, we'd rarely have to, since the infringing copy should be removed as a result of the ruling.) That's the realm that we are in here, very reasonable suspicion - not only is IA not the authors' or publishers' website, but the publishers of a large portion of the books on the site are saying via their lawsuit that this is an infringment, a stance that many authors have echoed. So we don't even need to reach conclusions over any concern regarding whether editors can be sued; we have the basic question of "are these links within policy?"--Nat Gertler (talk) 12:33, 16 June 2020 (UTC)[reply]
Oppose Is this really within the scope of the community's self-regulatory power, to decide a legal question like this, on how likely this lawsuit is to affect the WMF? Certainly if the WMF thought that this bot, and links to IA books, was a danger to the project, they would invoke their authority and block the bot and inform ArbCom of their decision? I suppose that as the community conceded to the bot, the community can withdraw its consent, but I don't know that it's helpful to try to get that withdrawal by appealing to the WMF, who seems content to allow the situation to continue. Have you tried requesting their intervention? Until then, I find the bot's edits useful, and think it likely that the IA will not face legal consequence from CDL. Psiĥedelisto (talk • contribs) please alwaysping!08:20, 16 June 2020 (UTC)[reply]
Oppose per all the above. Our linking is not violating copyright nor policy, so it would put an unnecessary burden on the reader for no net gain for Wikipedia or the readers. Dennis Brown - 2¢17:53, 16 June 2020 (UTC)[reply]
Do I understand correctly, that this bot basically forces all our articles to replace Google books' links? If that's the case, shut it off until there is an affirmative consensus (widely adverted) that Wikipedia is to be mechanically and universally replacing the links that the article authors (contributors) have provided. -- Alanscottwalker (talk) 21:32, 16 June 2020 (UTC)[reply]
Shut it off. I have commented above, but not "voted" yet. Now I am. I'm not impressed by the "oppose" arguments which mostly seem to come down to variations of "we won't be hurt by doing this, and who cares about anybody else?" or else self-serving analysis that leads to "naw, nobody's hurt if we do this." I don't believe it, and arguments like that me less, rather than more, inclined to wish to join with the people making them. Herostratus (talk) 00:12, 17 June 2020 (UTC)[reply]
Pinging a WMF legal member - @Jrogers (WMF): - just an FYI about this discussion on whether linking to a potentially problematic copyright issue is of note here and if Legal has any opinion, or is happy to leave it to the Community to decide (at least as things stand). Nosebagbear (talk) 12:19, 17 June 2020 (UTC)[reply]
I should have raised earlier that is that these pages that this paid-for-by-IA-bot is linking to are sales pages. IA bought out a major used book source last year, and these pages that they are linking to include a purchase link on them. To some degree, the bot is turning Wikipedia into a sales catalog for their bookstore. --Nat Gertler (talk) 15:39, 17 June 2020 (UTC)[reply]
If true, this is worrisome, as is the conduct of IA bot operators and the edits of their bots on different projects as I have said earlier (see above). Eissink (talk) 19:57, 17 June 2020 (UTC).[reply]
You don't have to trust me on this. Here's an edit by the bot. Click on the link it added, and see the Purchase button right next to Log In And Borrow button. --Nat Gertler (talk) 20:07, 17 June 2020 (UTC) Oh, and when looking for a reference for IA's purchase of Better World Books, it got worse. To quote the Library Journal on the purchase, IA founder and digital librarian Brewster Kahle told LJ’s Lisa Peet that the acquisition—and its pipeline of titles for digitization—would also facilitate broader ongoing efforts at IA to link internet content with relevant, reliable source material. “What we’re trying to do is weave books into the Internet itself, starting with Wikipedia,” Kahle said. [...] "We now have over 120,000 Wikipedia citations pointing to over 40,000 books, but we want to get to millions of links going to millions of books. The way we’re going to get there is by working really closely with Better World Books.” That's pretty blatant. --Nat Gertler (talk) 20:14, 17 June 2020 (UTC)[reply]
While the discussion is split, there looks to me to be enough objection that we cannot say that the Bot has consensus to continue making such edits. Shall we turn the bot off from making further edits until consensus is achieved? We can discuss whether to undo the existing edits separately. --Nat Gertler (talk) 15:54, 18 June 2020 (UTC)[reply]
That isn't how consensus works. If there isn't a consensus, then the status quo is maintained. This is true for all consensus based discussion here, and literally every consensus based community on the planet. Dennis Brown - 2¢17:50, 18 June 2020 (UTC)[reply]
Yes, and the "status quo" is the existing edit. Every new edit this bot makes is a change, not status quo. Reverting the exist edits would be against status quo, thought it's something that should be done. --Nat Gertler (talk) 18:04, 18 June 2020 (UTC)[reply]
Does this need to be escalated to the WMF? I think there needs to be a set of allegations that are investigated and found to be either true or not. What my reading of the situation is, is that (a) that the IA has unilaterally ignored copyright, (b) they have commissioned a bot to replace the current Google Books links, and (c) their links have purchasing options on them. If true this is worrying and the bot should be halted whilst this is investigated further. - Master Of Ninja (talk) 16:45, 18 June 2020 (UTC)[reply]
Oppose per Levivich, Moneytrees, etc. The WMF has a legal team to investigate any potential legal liability, we shouldn't make decisions based on speculative amateur lawyering. The Internet Archive is a legal entity in the same country - the same city even! - as Wikipedia's (for legal purposes) governing organization, so they are governed by the same rules. As long as they are legally open for business, there is no issue here that demands action, much less a profoundly silly one which would make the work we do that much harder. Gamaliel (talk) 16:27, 19 June 2020 (UTC)[reply]
Oppose per most of the above, as far as IA goes, as the copyright debates are a problem for WMF Legal and until or unless they rule, it's a matter of WP policy if this is a legitimate site to which we link, and it is. "Linking to copyrighted materials" is a red herring argument, that occurs every time we link to a newspaper or any other work under copyright that can be read online. GreenC's note that the same issue occurred with Google is well-taken. We can't make the legal decision on that stuff, it's the courts' job to do so. At most, it's WMF's problem, not ours. In general, having bots that can update links to IA, Google Books, Hathi Trust or any of the other services that scan and make books available is very helpful and assists the eternal battle against linkrot. I was appalled to see one editor above say do we need online links to references? The answer, my friend is HELL YES! In general, the issue of linking is critical. "Just look it up" is a time sink, particularly for historic topics. Even if the link is to only an abstract of a journal or a snippet view of a book one has to buy to read in full, LINKS ARE CRITICAL. If you've ever worked on a major featured article with hundreds of sources, (the biggest I ever stuck a toe into was Richard Nixon), it is very difficult to find some types of works, you can't just "google it." Having a link to what might be online is extremely helpful, particularly where there are multiple possible locations. Montanabw(talk)20:44, 19 June 2020 (UTC)[reply]
I have some sympathies with Nat Gertler here. I don't think that the national emergency library is defensible and in effect did amount to mass copyright infringement, but that is over now so it is a moot point. Controlled digital lending is a different issue which I can see arguments for and against, and it is up to the courts to decide their legality. I think linking to old but still technically in copyright books from the early to mid 20th century are fine, but I can see why Nat has issues with linking to more contemporary works where real revenue for authors is at stake. Hemiauchenia (talk) 21:36, 19 June 2020 (UTC)[reply]
Support per the precautionary principle and the principle of least harm; I also have concerns about the effect of this on authors and small/independent booksellers.— Preceding unsigned comment added by CactusJack (talk • contribs) 22:21, 19 June 2020 (UTC)[reply]
Oppose per Levivich and others. We aren't linking to (c) violations; we shouldn't act on unresolved legal challenges (not to mention weak ones); and as Denis notes it would put an extra burden on readers for no net gain. – SJ +02:10, 20 June 2020 (UTC)[reply]
Original Drawings of People Based on Multiple Copyrighted Works?
I need editors and administrators who are very knowledgeable with Wikipedia policies and guidelines to clarify something for me. There are several instances on articles where drawings or simulacrums of living people based on (presumably) copyrighted works have been substituted for a fair-use photo or even no image at all. I want knowledgeable editors to clarify if and how this is acceptable. Specifically, I would think that this falls under original research. There have been instances where original drawings have been deemed to be unacceptable because they copied a copyrighted image too closely and other instances where original drawings have been featured on the main page. There have been other instances where the issue has caused confusion among editors who have raised BLP and OR questions. And more recently I've come across a simulacrum of a living person "based on photographs".
I want to be absolutely clear on how these issues mesh with the policies on no original research, BLP, and usage of copyrighted material. Yes, I'm aware that WP:OI states that "Original images created by a Wikipedian are not considered original research, so long as they do not illustrate or introduce unpublished ideas or arguments, the core reason behind the NOR policy". My contention is: don't hand-drawings always introduce biases and original viewpoints (cf. "unpublished ideas") within the work, no matter how careful and honest the artist is? Blemishes are ignored, asymmetries are corrected (even unconsciously), and the portrait will generally always portray the subject in a flattering light through the artist's personal choices of composition and lighting.
A photograph, on the other hand, has much less chance for this. Now, if you base a drawing or simulacrum on multiple copyrighted photos, does that somehow eliminate the problem of derivative works—as in: to get around the copyright issue, you "use" as many copyrighted works as you can and somehow dilute the problem away? I hope you can see how confusing this is for me. Is it or is it not better to have a low-resolution photograph under fair-use of a living person than an original drawing by a Wikipedian editor trying (in my opinion) to skirt around the copyright or difficulty in photographing problem?
What a person looks like is not copyrightable, it is what the photographer captures that is, and that's the derivative work we want to avoid. (There are places where individuals do have express control on their personality rights, and this is a caution we give to reusers of images, but does not prevent use from creating and offering images) What we want to avoid in a original image of a person is a drawing that is too close to a given photograph in terms of pose, angle, expression, and other factors that would be unique to the photographer (not the person photographed) that could cause the derivative work concern. Eg, tracing over a photo and then creating a painting or similar that otherwise keeps that same pose would be too close as derivative. Ideally, the image created by the user should be of a pose that is not close to any of the ones they're starting from. If one had a full side profile view photo, and a full head-on view photo, creating an image at 3/4ths view would be fine.
The other side is that the end result should be fairly recognizable and not unduly abstract, oversimplified, demeaning, praising, or the like. Taking a person and then making an original drawing but giving them a scowling expression would not be acceptable. Nor would be trying to noticably de-age or age them up. This is not as straightforward to evaluate as the first point but this should be considered. And this is your point about biases that we should avoid. For example, were I to draw a member of the Christian church living today but somehow not able to get a free image, adding even a small halo effect around them would be wrong. --Masem (t) 18:00, 17 June 2020 (UTC)[reply]
Even assuming that we're talking about a drawing that doesn't infringe on the copyright of any of its source materials, I'd question its informational value. It's one thing if it's by a notable artist, but if it's not then we get into very subjective questions about whether it's an accurate depiction. Regardless, the derivative copyright question is so sticky that a prophylactic rule may make sense. postdlf (talk) 19:28, 17 June 2020 (UTC)[reply]
When done skillfully and carefully to avoid copying the copyrightable aspects of the original photo(s), drawing is a perfectly valid way of obtaining free images. Even photos cannot avoid the creative hand of the artist; although one may attempt to achieve a realistic look when editing, colors are certainly subject to variation (e.g. The dress), and the choice of composition affects the relative sizes of elements in the frame in a way that could arguably be misleading. So I wouldn't count out a drawing for the sole reason that it is a drawing; editors should evaluate drawings case by case to ensure that they are faithful to the original photos without being too similar. -- King of ♥ ♦ ♣ ♠21:38, 17 June 2020 (UTC)[reply]
South American couscous with broccoliThere is another problem, one that comes with all images at Wikipedia: If you don't know what the person looks like, how do you know the image is accurate? If someone really wants to troll Wikipedia, images are our weakness because we don't hold them to WP:V or WP:RS the way we do text. I would argue that verification and insisting they come from reliable sources (even if a different set of policies than we have for text) is the only way we can prevent absolute falsehoods with images. I don't have links in front of me, but we have been fooled before. Most of the time, our Fair Use images are from sources that are reliable, so more reliable than many Free works. Hand drawn images that some guy named Bob drew based on some images he saw? No matter how well they are done, I wouldn't like that. That is a Pandora's Box, and that isn't even considering the copyright concerns. Dennis Brown - 2¢00:19, 18 June 2020 (UTC)[reply]
Thanks for that. I have long had a niggling concern that images are not subject to the policies that apply to article text, but the concern has not been strong enough (or I have not been brave enough) for me to open this Pandora's box. We even encourage original research in this area. I also have a concern with the statement by the proposer here, "a photograph, on the other hand, has much less chance for this". Until a few decades ago it required some skill to manipulate a photograph (think of those Soviet photos with Trotsky removed or airbrushed images of models) but these days anyone with a computer or a smartphone can do so pretty easily. Phil Bridger (talk) 07:56, 18 June 2020 (UTC)[reply]
You can upload an image of a fake (troll) mathematical algorithm and it is more likely to stick around than text. Or take an image of one plant, and post it in an article on a similar plant of another genus and it might stay in the article for years before being noticed. It's easier to fact check text, by far. Our policies on images all revolve around copyright concerns, not verification or the source being reliable. In fact, the source for all user generated images are not reliable sources, including the many images I have uploaded. Again, if I wanted to troll Wikipedia and insert a lot of crap information, I would do it with images because they have very little oversight. Dennis Brown - 2¢11:22, 18 June 2020 (UTC)[reply]
I think a hand drawn image is OK if faithful and if the artist has primarily used their own eye sighting of the subject. If the image relied on a copyrighted image or video, it is a derivative work. Relying on many different copyrighted works doesn’t work, that would be fair use of the multiple other images, and the artwork would not then be a study of the subject, but a study of the multiple other images, and the only correct use of the hand drawn image would be to illustrate coverage of the multiple other images. Do you have, can you link, an example of a hand drawn image based on multiple identified copyrighted images, that someone argues is a free image? —SmokeyJoe (talk) 11:41, 18 June 2020 (UTC)[reply]
It is simply not true that any work based on a copyrighted work is subject to copyright of the original copyright holders. For an example, see c:Commons:Fan art#There is no copyright in a commonplace pre-existing element, where HP books.png is clearly based on the Harry Potter series but uses no copyrightable aspects of it. A person's likeness is not copyrightable, and if you combine several photos taken from different angles then all you have left is the likeness. I think a useful test is this. Let's say you're given a drawing like File:Jashodaben Narendrabhai Modi painted.jpg but are not told what its sources are. On one side is the actual list of source photos A, and on the other side is a fake list of source photos B (not fake photos, but real photos of the subject which were not used to create the drawing). Ask people to guess which is the true list of sources. If people can't tell, then I think it is safe to conclude that the drawing has transcended any individual creativity present in the source photos. -- King of ♥ ♦ ♣ ♠18:54, 18 June 2020 (UTC)[reply]
It’s a boundary issue. It is not “simple”. The following statement is simple: “Whenever you take an existing image and modify it to create a different image, you are making a “derivative work.” —SmokeyJoe (talk) 23:06, 18 June 2020 (UTC)[reply]
I guess that User:King of Hearts are talking at slight cross-points? The Wikipedian-made image is a derivative work of a copyrighted work, but I guess that being a derivative work does not necessarily mean that the copyright owner of the source material can make a claim of ownership of the derivative work. User:King of Hearts gives some examples, some things are not copyrightable. --SmokeyJoe (talk) 02:12, 19 June 2020 (UTC)[reply]
Little background about me: I've edited Wikipedia on rare occasions over the last 15 years, usually when I saw something wrong while reading on Wikipedia. People often ask me why I don't create an account. The answer is simply that I don't feel like it. I've never created one and probably never will. There is no reason for me to ever create one. One of the biggest reasons I was drawn to Wikipedia was its vision of creating the greatest encyclopedia by the common people (that has never happened before in human history).
Edits by IPs are much more likely to be wrongly reverted - wrongly assumed to be vandalism or assumed to be incorrect - than editors with an account despite the fact that their edits are correct (improvement to Wikipedia) (with reliable sources cited too). IPs also often get mocked from established editors. I'm only talking about correct edits from IPs. Vandalism and incorrect edits from IPs are not within the scope of this concern. Evidences are shown here: 1, 2, 3. All three cases involve an IP's edit that was reverted multiple times and later was determined to be actually correct. I'm an IP that is familiar with Wikipedia. I know how to navigate and get to the right venues to fight back and keep my edits from being reverted. Most IPs are not familiar with Wikipedia. When they see something obviously wrong, they make an edit to correct it, but soon enough, their edits would be more likely than not to be wrongly reverted by some patrollers. And that would be the end of the story. They don't know any way to argue back and probably would get blocked if they're persistent about it in the wrong venue (and no, arguing in the article talk page doesn't work; I've tried it that way many times and failed). I do wonder how many millions of edits (of IPs) have been wrongly reverted since the beginning of Wikipedia. How much better Wikipedia would be now if this systematic discrimination against IPs never happened? Maybe, we would have a big community of IPs in Wikipedia by now if things were different.
No one has apologized for their mistake in all 3 cases. I feel like this has become a norm here on Wikipedia that people don't feel the need to apologize when they make a wrong assumption about an IP. It is a serious problem that needs to be addressed. This undermines the very principle of Wikipedia, "anyone can edit". With the current status quo, it is more like "anyone can edit but be prepared to get discrimination if you're an IP". IPs are contributors too, and they deserve the same respect and treatment as other editors with an account. Judge IP's edit from its contents, not by the fact that it was edited by an IP. 14.169.157.34 (talk) 15:32, 18 June 2020 (UTC)[reply]
As a matter of principle, yes, every edit should be judged on its merits regardless of who made it. As a matter of practice, yes, IP edits get increased scrutiny because, even excepting vandals and discussing only good faith contributors, IP editors are more likely to be inexperienced, unfamiliar with policy, and make mistakes. As a matter of practicality, if it's that burdensome then register an account. It requires probably 1/30th of the time it took you to post this. It doesn't require an email. It doesn't even require that you actually use the account for more than a single editing session.
Wikipedia is a pragmatic enterprise. So I mean. Sorry. You wan't anonymity then you get it. It's not a perfect world and we can't have all the up sides of what we want without the down sides. You're not practically going to change the organizational culture of 300,000 contributors because of the inconvenience of taking ten seconds to register an account. — Preceding unsigned comment added by GreenMeansGo (talk • contribs) 15:58, 18 June 2020 (UTC)[reply]
I agree IPs should get increased scrutiny. However, increased scrutiny isn't the same as reverting edits that are correct (edits that improve Wikipedia). IP's edits are reverted even though it's correct. I never said inconvenience was the reason for me not creating an account. If I want to, I would do it even if it takes 10 hours. You didn't get my point. And most IPs (sporadic contributors) will never have an account. Even if I have an account, that wouldn't help any other IP. And no, I'm not going to change the IP discriminatory culture of Wikipedia, but big change started from the smallest thing. First thing first, I want to bring this issue to the awareness of as many editors as possible. 14.169.157.34 (talk) 16:25, 18 June 2020 (UTC)[reply]
I mean, we're already aware of it. Having had an account rename a few years ago, even with a fancy pants signature, I can tell you that I got a lot of increased resistance from people I'd worked with for years, just because I was suddenly a name they didn't recognize. I suspect that part of that is likely just a tribal human thing. GMGtalk17:21, 18 June 2020 (UTC)[reply]
One of the principal ways of achieving WP:CONSENSUS on a change on Wikipedia is the WP:BRD (bold-revert-discuss) process. How can you discuss a topic over any period of time, knowing you're talking to the same person, with a user whose broadband router rebooted and they got allocated a new DHCP IP, or they moved out of range of their mobile phone mast and got a new IP, or they're on IPv6 which switches IPs in a /64 range apparently on a whim? Cabayi (talk) 16:11, 18 June 2020 (UTC)[reply]
What do you mean by "how"? I did successfully reach consensus many times over the last 15 years as an IP (the 3 cases above are just examples; consensus was reached in all 3 cases). You're missing the point. I can reach consensus because I'm familiar with the in and out of Wikipedia (which venues to use and such). Most IPs don't, so their correct edits being reverted by patrollers would be permanent. 14.169.157.34 (talk) 16:30, 18 June 2020 (UTC)[reply]
I concur with your observations. I would also add that this type of scrutiny/discrimination also applies to editors without userpages (distinguished by the red links in their signature—like mine). I have over 8,000 edits, but I noticed that the weight or respect of my contributions seemed to slacken by some people after I deleted my userpage. It's unfortunate. -- Veggies (talk) 16:24, 18 June 2020 (UTC)[reply]
@Veggies: you can always just redirect your userpage to your talkpage, if you are just wanting to make a statement by having a redlink - then a side affect of that statement may very well be increased scrutiny because in general no userpage==new editor. — xaosfluxTalk16:45, 18 June 2020 (UTC)[reply]
@Veggies: and xaosflux. I want to make it clear that increased scrutiny for IPs is actually great. However, that's not the same as increased rate of wrongly reverted edits, which is a bad thing. 14.169.157.34 (talk) 16:50, 18 June 2020 (UTC)[reply]
I see you (or someone with a similar IP address to you) casting aspersions on other editors, on Wikipedia, and boasting of your own personal credentials. None of these have any place on Wikipedia, and it seems to me like you are WP:FORUMSHOPping due to not getting your way at WP:ANI. Elizium23 (talk) 17:03, 18 June 2020 (UTC)[reply]
It wasn't an aspersion when someone failed to understand physics and wrongly reverted an edit. Nah, my ANI was closed because I used the wrong venue. I then used Wikipedia talk:WikiProject Physics and got my way many times over there. I'm all good. I already got it my way, so no need to forumshop (false accusation). I'm just trying to spread awareness about IP discrimination. It seems to me that you get salty because you're one of those sysops that like to discriminate against IPs. 14.169.157.34 (talk) 17:13, 18 June 2020 (UTC)[reply]
As has been explained to you, online communities are reputation-based. And you, as a set of octets with no talk page, have zero reputation when I am reviewing my watchlist or recent changes. Now that I've met you, and clicked through on your contributions, your address shows in purple, which for me is a negative reputation. So yes, I discriminate. Discrimination is not always a negative concept. I am sure you discriminate between chocolate and vanilla ice cream, or steak and chicken. You discriminate between night and day. Only unjust discrimination is wrong, and you're not in a protected class.
Someone has a quote about "editing as an IP is like walking around dressed as a moose, in a forest full of moose hunters, during moose season." Well, that is pretty apt. If you don't want to create an account, then it is your choice to remain at zero reputation in this community. We cannot artificially accord to you reputation if your IP keeps changing and you won't create an account. That's just beyond our technical power to do. Sorry, mate. Elizium23 (talk) 17:19, 18 June 2020 (UTC)[reply]
This is not just about me. It's about all IP contributors in Wikipedia. This is unjust discrimination in my opinion when a legitimate improvement to Wikipedia is reverted just because it was made by an IP (it wouldn't be reverted if it was made by editor with an account). Yes, it's beyond technical power to do, but it's not beyond individual power to do. If each and everyone stops the IP discrimination, this problem would be gone. That is not to say it's an easy thing to do. 14.169.157.34 (talk) 17:32, 18 June 2020 (UTC)[reply]
Given the trivial cost of ceasing to be an "IP contributor", it's hard to get to excited about "discrimination". Just register an account. Or not, as you like. But if you see there are advantages to having an account, I don't have too much sympathy if you decide not to take them. --Trovatore (talk) 17:43, 18 June 2020 (UTC)[reply]
This is not just about me. This is also not about trivial cost of creating an account. This is about all IP contributors, and most of them will never create an account due to many different reasons. Stop the discrimination instead of forcing all IPs to get an account, which is impossible to do btw. 14.169.157.34 (talk) 18:21, 18 June 2020 (UTC)[reply]
If they don't, that's fine. There are associated costs and benefits, which are up to them to weigh. You're correct that if an improvement to Wikipedia is reverted because it came from an IP, that's a bad thing — for Wikipedia, not for the editor personally.
There is no "justice" question here whatsoever. There would be no injustice if Wikipedia simply chose to allow edits only from registered users. For practical reasons we don't do that, because even a slight increase in the barriers to entry might discourage someone who would otherwise do a toe-dip, and we might lose out on that person's contributions. But it would not be unfair to that person. --Trovatore (talk) 18:48, 18 June 2020 (UTC)[reply]
It's a bad thing for both Wikipedia and the IPs. Being discriminated against is not fun, and yes, that was a bad thing for me as an IP editor. Discrimination happened to me many times already. It is also a concern because it affects Wikipedia very negatively. Small effects add up over a long period of time. How many millions of constructive edits have been reverted over the years? Your argument on justice is kind of moot. It's like saying everyone is equal under the law, but if you're black, you will get discrimination. 14.169.157.34 (talk) 19:35, 18 June 2020 (UTC)[reply]
The effect on Wikipedia we can discuss. For the effect on you personally I have next to no sympathy. You can register an account, or you can stop complaining. Or you can not register an account and keep complaining, but then you needn't expect me to take your complaints seriously. --Trovatore (talk) 19:41, 18 June 2020 (UTC)[reply]
It's not just about me. It's about all IPs. I didn't ask for your sympathy. I'm asking to stop discrimination. Effects on IPs aside, effects on Wikipedia is quite huge if we're talking about a long period of time. Either way, it's the same problem whether you ignore personal effects or not. 171.253.130.71 (talk) 20:20, 18 June 2020 (UTC)[reply]
It's about all IPs, but they can also register accounts, so I don't have much sympathy for any complaints about "discrimination" from them either. It's not like it's a fundamental characteristic. Just register an account; then you're not an IP anymore.
It's not at all the same question if we're talking about the effect on Wikipedia. That's a discussion we can have, minus this nonsense about justice. --Trovatore (talk) 20:29, 18 June 2020 (UTC)[reply]
They can create an account, yes but it's impossible to force all IPs to create an account. That doesn't solve anything. Plus, this goes directly against Wikipedia's principle of anyone can edit. You can ignore the justice part, whatever. The problem remains exactly the same. Wikipedia is negatively affected by IP discrimination regardless of which effects you want to acknowledge. 171.253.130.71 (talk) 20:40, 18 June 2020 (UTC)[reply]
Who wants to force them to create an account? I certainly don't. I'm just pointing out that they have a very easy remedy. It doesn't at all go against Wikipedia's principle that anyone can edit, because anyone can create an account.
The issues are somewhat different if you want to talk about the effect on Wikipedia. There are costs in terms of good edits lost, but there are also benefits in terms of collaborative editing, which is more effective if you know whom you're talking to (not necessarily in the sense of name and address, but at least in the sense of being able to draw on past interactions rather than having to deal with each edit from square zero). As I say, that's a discussion we can have. But the noise about fairness is going to get in the way, if you keep pushing that angle. --Trovatore (talk) 20:49, 18 June 2020 (UTC)[reply]
Your easy remedy is forcing all IPs to create an account if they don't want the discrimination. I'm not pushing any angle. I'm looking at the problem from an overall point of view. IPs can collaborate with editors on editing, creating an account is not needed to do this. You can talk about the pros and cons of creating an account and its effects on Wikipedia, so far so good. But at the end of the day, it's the exact same problem whether you ignore personal effects caused by IP discrimination or not. 171.253.130.71 (talk) 21:05, 18 June 2020 (UTC)[reply]
Strange, I edited as an IP for months - IPv4 and IPv6 alike - and I did not encounter unjust discrimination from editors. Occasionally I was reverted without explanation; I'd make a case on the talk page and the edit would be reinstated. It was no big deal. I think you're projecting. Elizium23 (talk) 17:49, 18 June 2020 (UTC)[reply]
It's not just about you. Just because you haven't experienced unjust discrimination, that doesn't mean most other IPs haven't. Plus, you're an experienced editor, so you know the in and out of Wikipedia. You know how to raise your concern properly and which venues to use (article talk page doesn't always work on this matter; plus, most IPs don't use article talk page anyway). Most IPs don't, so their improvement edits that were reverted would be permanent (most of them probably don't even check if their edits are reverted afterward). 14.169.157.34 (talk) 18:17, 18 June 2020 (UTC)[reply]
I'm not saying all good (improvement) edits by IPs are reverted, but it happens to a good amount of edits by IPs compared to editors with an account. Some of my edits are not reverted. For the reverted ones, I know which venue to use, so I can get my edits un-reverted. Most IPs wouldn't be able to do what I did, and that is the problem. Reverted edits are permanent even though edits originally by IPs were actually an improvement to Wikipedia. 14.169.157.34 (talk) 18:47, 18 June 2020 (UTC)[reply]
Just to echo what everybody has said, in theory, we judge contributions purely on their merit, blind to who wrote them. That being said, there's no doubt that edits by IPs get greater scrutiny, as do edits by users with redlinked user pages, newly created accounts, etc. It cuts the other way as well. There's some users who I respect greatly. Sometimes I'll see something I disagree with then notice one of those people wrote it. That doesn't mean I'll automatically go along with it, but it certainly means I'll stop and invest more time to understand where they're coming from. This is reality. You're free to edit as an IP, or not have a user page, or whatever. But, recognize that doing those things will affect how people view your contributions. Again, this is reality. -- RoySmith(talk)16:57, 18 June 2020 (UTC)[reply]
RoySmith, greater scrutiny is not the same as reverting correct edits (improvement to Wikipedia). You can have greater scrutiny for IPs and new users AND also not wrongly revert a good correct edit at the same time. 14.169.157.34 (talk) 17:18, 18 June 2020 (UTC)[reply]
As I understand it, the WMF plans to suppress the visibility of IP addresses to protect the privacy of anonymous edits. Such edits will presumably be treated in a generic way and IP editors will have even less of an identity than they do now. If editors want to be taken seriously as individuals then they should create an account. Otherwise, their edits will be ranked alongside all the other anonymous edits. If you want such edits to stick, I suggest that you attach impressive sources and edit summaries to them. Andrew🐉(talk) 18:39, 18 June 2020 (UTC)[reply]
IP addresses or complete anonymity, it doesn't matter. All of them deserve respect and the same treatment as other editors with an account, not discrimination. Sometimes, a good constructive edit with reliable source and a good edit summary still can get reverted by a patroller (it happened to me many times already). 14.169.157.34 (talk) 19:01, 18 June 2020 (UTC)[reply]
They do in theory, but in practice, discrimination happens. By saying they don't, it implies that discrimination against IPs is a good thing to you? 14.169.157.34 (talk) 19:29, 18 June 2020 (UTC)[reply]
They don't "deserve" equal treatment, not even in theory. Discrimination against IPs may be good or bad depending on the effects on Wikipedia. But it is not under any circumstances a "fairness" issue. --Trovatore (talk) 19:38, 18 June 2020 (UTC)[reply]
There is no Wikipedia's official policy to discriminate against IPs, so in theory, they DO deserve equal treatment. In practice, they don't get equal treatment due to discrimination. 14.169.101.188 (talk) 08:17, 19 June 2020 (UTC)[reply]
There is also no official policy not to discriminate against them. In fact there are some areas where they are officially discriminated against (for example, they can't create articles). No, they don't "deserve" equal treatment, not even in theory. If they don't like it, the remedy is simple. --Trovatore (talk) 15:58, 19 June 2020 (UTC)[reply]
There is no official policy against or support discrimination. However, it's common etiquette to not discriminate (except in areas with official policy such as article creation). 14.169.140.90 (talk) 08:45, 20 June 2020 (UTC)[reply]
I agree with you, OP, that Wikipedia has strayed way too far from its mission to be the encyclopedia anyone can edit. Hell, if I were in charge I'd even remove all the restrictions on creation of new articles by IPs and new accounts. Unfortunately, Wikipedia's culture has become increasingly bureaucratic, and as policy creep continues the discrimination against IP edits will only get even worse. I had something similar, but not quite analogous, happen to me a few days ago. I had gotten myself locked out of this account (long story...) so while I waited for the devs to restore my access, I created a temporary account User:CactusJack2. On my main account (this one), for a few months now I've been working on sorting through hundreds of mass-generated microstubs on California "unincorporated communities" that aren't actually communities and never were. Basically, about a decade ago, a few editors mass-created articles for thousands of locations listed on the GNIS database, which in addition to communities contains numerous non-communities including things like bridges, railroad sidings, individual farms, etc. Anyway, I redirected one of these non-communities to an appropriate redirect target, and left an explanation in the edit summary. It was almost immediately reverted by an editor using Huggle. When I explained the edit on that editors's talk page, they did not respond. If that happened to me in my handful of edits on a brand-new account, I'm sure it must happen constantly to constructive edits by IP editors. I think that many Recent Changes patrollers are far too agressive with their reverts. Too many times I've seen good faith, and even constructive edits rolled back or Huggled by someone who likely didn't take more than a brief second to look at the change and the edit summary. It would be nice if we could find a way to tamp that down. Maybe we could start by being stricter about who gets rollback - rollback misuse is rampant from what I've seen. We could even consider restricting the Twinkle 1-click revert buttons to editors with rollback permissions. −−−CactusJack 🌵18:47, 18 June 2020 (UTC)[reply]
This discussion may bear more fruit if we address it from the perspective of overzealous recent changes patrolling instead of framing it as a matter of justice. The former is a very real problem. −−−CactusJack 🌵18:53, 18 June 2020 (UTC)[reply]
Yes, that would be at least potentially a more productive line of discussion. I don't know whether "overzealous recent changes patrolling" is really a big current problem, but it's the kind of thing for which evidence could be produced and evaluated. As far as "anyone can edit" goes — anyone can register an account, so it does no harm to "anyone can edit" to have registered accounts treated differently from IPs. --Trovatore (talk) 19:17, 18 June 2020 (UTC)[reply]
If we required registration, it would be "Anyone can edit. First you have to register." I don't see anything fundamentally wrong with that. I would probably be against it as a practical matter. But there'd be nothing unfair about it. --Trovatore (talk) 19:44, 18 June 2020 (UTC)[reply]
First, Wikipedia's current slogan is "anyone can edit", not "anyone can edit. First you have to register." It would make more sense if they change their slogan to the latter if they want to legitimize IP discrimination. Second, unjust discrimination is always unfair. If you disagree, we can agree to disagree. 171.253.130.71 (talk) 20:31, 18 June 2020 (UTC)[reply]
"Anyone can edit" doesn't exclude "first you have to register". You also first have to click the edit button. Registration is not a lot harder than that. We could ban all non-logged-in edits and still honestly keep the motto "anyone can edit". As for "unjust discrimination is always unfair", of course; that's what "unjust" means. But discrimination against IPs is not unjust. --Trovatore (talk) 20:36, 18 June 2020 (UTC)[reply]
That is your own interpretation. Anyone can twist any phrase into a different phrase that is supportive of their view. It is "unjust" when constructive edits (improvement to Wikipedia) are reverted "just because" it was made by an IP. The same constructive edits wouldn't be reverted if it was made by an editor with an account. 171.253.130.71 (talk) 20:48, 18 June 2020 (UTC)[reply]
It's not unjust to give greater scrutiny to edits just because it was made by an IP. However, greater scrutiny and reverting constructive edits are not the same thing! The first one is a good thing while the latter is a bad thing. 171.253.130.71 (talk) 20:53, 18 June 2020 (UTC)[reply]
Yes, it's one of many possible solutions. The best solution would be all editors judge IP's edit solely on its content regardless of its creator. That probably will never happen. Maybe, setting a higher bar to become rollbackers. And, take away rollback tool if it's being abused. For example, a rollbacker repeatedly reverts constructive edits by IPs (or anyone) should get his/her tool revoked. Maybe, even have a policy against IP discrimination and spread awareness around. IP appreciation month awareness or something like that. 14.169.157.34 (talk) 19:24, 18 June 2020 (UTC)[reply]
No one cares if you're a physicist. If you want to be a tourist here, fine, but if you can't be bothered to create a meatball:SerialIdentity why should anyone bother to trust your expertise? If you aren't willing to stake your reputation on your edits, don't expect others to volunteer their time fact checking an edit they find suspicious. Most reputable academic journals outright refuse to publish anonymous submissions for this reason. You cannot walk into your place of work wearing a mask, hand an anonymous report to your boss, and then get angry when she treats it with suspicion or rejects it outright. If you want to be treated in a professional manner, engage with the community in a professional manner by establishing a serial identity. That should not be a requirement of course---I've argued against it elsewhere---but it's no accident that contributions are viewed with less suspicion when they come from editors who stake their reputation on their work. — Wug·a·po·des23:20, 18 June 2020 (UTC)[reply]
Even if I create an account, that does not nothing to solve the IP discrimination problem. It's not just about me; it's about all the IPs. Forcing all IPs to create an account is an impossible solution. 14.169.101.188 (talk) 08:13, 19 June 2020 (UTC)[reply]
It's not an impossible solution; there are millions of websites where you need to register before participating and they operate just fine. It's not strange to tell someone to create an account. It's not the solution you want, but that doesn't mean it's not a solution. In fact that solution was proposed last November. Though you may view the treatment of IPs as a problem many people do not agree with you, and some even believe we should have harsher sanctions for IPs. Simply asserting your perspective is the correct one and telling us to change the community to align with your perspective is not helpful. We work by consensus here, and you will need to convince a lot of people that our culture of treatment towards IPs needs to change. We're having a hard time doing that to deal with far more important problems like harassment, so it's hard to get hyped about inconvenienced IP editors.If you stuck around, you'd notice that many people are trying to fix this problem and have been working on it for years. You'll notice in the two discussions above I was one of the more staunch defender of IP editing. Personally speaking, I make it a point to welcome every IP that I see making helpful edits in my topic area. Editors and administrators have been admonished for biting new comers, and our rollback policy already states that it can be taken away if repeatedly used to revert helpful edits. The Wikipedia:Welcoming committee works very hard to welcome new editors and encourage more editors to do so. People have been working to improve our welcoming templates, introduction pages, and interface messages so that they are less formal and off-putting for potential editors. Plenty of people are working to change the culture and make Wikipedia more friendly to IPs, but it is hard work that takes more than a disgruntled post at the village pump.When you repeatedly and explicitly refuse to join a community, the community tends to get upset when you tell them how to go about their business. It's especially hard to take this post seriously when one of your examples shows you saying you hate Wikipedia and the OP reads like your personal gripes about WP:BRD. You have not stumbled onto a new problem. As this thread shows, many people are well aware that IPs are not treated the same as registered editors, so forcefully and repeatedly telling us about a problem we're already working on while refusing to join or help is annoying to say the least. Rather than showing up to complain, try helping by offering solutions or new ways of thinking about the problem. Instead of telling us what to do and then leaving, join us and help make the change you want to see. — Wug·a·po·des00:00, 20 June 2020 (UTC)[reply]
I'm trying to spread awareness here, which is a possible solution in itself. I did offer a few other possible solutions somewhere in this long discussion. The best possible solution is of course to stop IP discrimination altogether (impossible solution yea). Registration solution will likely turn away many IPs that edit for the very first time (no one is going to create an account to fix a typo they see while reading; this could be a gateway experience to become long time editor); no consensus happened for a reason. IPs are a part of this community too (I did join the community as an IP). I'm feeling disillusioned with Wikipedia because it has betrayed its own principle of "anyone can edit" (some long time editors have this feeling too due to various different reasons). I said hate editing Wikipedia, not hate Wikipedia (there is a difference). I'll come back to edit from time to time as an IP. 14.169.140.90 (talk) 08:34, 20 June 2020 (UTC)[reply]
I hereby declare Monroe's law: Chance of revert approaches 100% as number of reviewers approaches infinity. From my experience, every single edit by every single editor will be disagreeable to some other editors, as no statement can be pleasing to everyone. Theoretically, that means that any edit will be reverted if it's reviewed by enough editors. Edits by IPs are critically reviewed by more editors than edits by registered users, for the reasons already stated in this discussion. That may explain at least part of the perceived bias against IP edits. While I have no evidence this is the main cause, we also have no evidence it is not a significant factor. --A D Monroe III(talk)01:21, 19 June 2020 (UTC)[reply]
Agree with Wugapodes, this place (like all communities) is reputation- and trust-based. Who can remember a string of ten numbers, assuming they are not a dynamic IP? Where I edit in the WWII Balkans space, at least 90% of IPs are vandals, disruptive and/or abusive, so many established editors in the space revert unsourced additions and any deletions by IPs on sight, and scrutinise sourced additions by IPs very closely. I encourage IPs that make constructive edits to get a username so they can build trust with the other editors that are in the space, but I always suspect that the reason many IPs won't register is because they don't want to be accountable for their edits. Peacemaker67 (click to talk to me) 01:51, 19 June 2020 (UTC)[reply]
"the reason many IPs won't register is because they don't want to be accountable for their edits," you got a wild accusation right there. The vast majority of IPs will never create an account due to many different reasons (one of the possible reasons could be they never thought about creating one). Most IPs edit very sporadically, maybe a few edits per year. It's not much per IP, but it adds up due to a big number of IPs editing over the years. 14.169.101.188 (talk) 08:09, 19 June 2020 (UTC)[reply]
The OP's IP's concern is interesting, but it would be more impressive a concern with examples that are theoretical physics or mathematics. I think there is a philosophical non alignment between historiography (and I assert an encyclopedia is an historiographical document) and theoretical physics and mathematics, with regards to the meaning of "knowledge". --SmokeyJoe (talk) 02:04, 19 June 2020 (UTC)[reply]
The given examples in fact are in theoretical physics. Anyway, it's a problem in ALL topics, not just in theoretical physics or mathematics. 14.169.101.188 (talk) 08:04, 19 June 2020 (UTC)[reply]
I have been a registered Wikipedia user for almost 5 years. My first edit was as an IP about seven years ago. I registered an account so I could make contributions and gain access to additional features.
AFAIK I don't think there is "discrimination" against IP editors and editors with no user pages. Those are signs that the editor may be new, but it does not mean that the editor did anything wrong. Their edits are more commonly misunderstood, so they clarify a bit more in their edit summary when they reinstate their edit. I ask them "Can you maybe explain the edit you made to X" or something like that. Aasim18:41, 19 June 2020 (UTC)[reply]
I haven't been following this discussion closely, but anyone who thinks ips and established registered accounts are on equal footing needs to spend more time editing while logged out. I edit as an anon as a matter of course if my connection is insecure; I've been rolled back without apparent reason more times than I can count, several times been ignored outright on talk pages of semiprotected articles, and once had a (now-former) admin use semiprotection as a weapon of first resort in a trivial content dispute. You pretty much have to accept that you can only make edits you don't much care about. —Cryptic19:36, 19 June 2020 (UTC)[reply]
Cryptic, A suggestion about insecure locations: Make yourself a second account. For example, I have User:RoySmith-Mobile, which use on my phone. I can't recall the last time I logged in from a public terminal, but if I did, I'd use that account as well.
And, I'll echo what you said about equal footing. As an admin, I spend a lot of time evaluating other editor's behavior. I can't tell you everything that goes into forming an initial opinion, because I don't completely understand how my brain works. But, I can tell you that IP is a factor. As is a red-linked user page. Or a low edit count. In an ideal world, I'm not supposed to care about those things, but in the real world, they are factors. I'm human. -- RoySmith(talk)23:10, 19 June 2020 (UTC)[reply]
I haven't been following this discussion closely, either. My perception is that many think of IP discrimination as roughly equivalent to racial, gender, age, etc. discrimination, completely missing the fact that one's race, gender, age, etc. are not personal choices. IP discrimination does exist and will always exist, I don't like being on the receiving end of it, so I avoid IP editing. Problem solved, and I reject all attempts to make things more complicated than that. ―Mandruss☎22:45, 19 June 2020 (UTC)[reply]
There seems to be more at issue than just simple discrimination against IP editors. At Quark the IP, who seems to have been posting from two dynamically-addressing ISPs (IPv4 and v6), edit warred even after starting a talk page discussion. The article was protected from an "IP-jumping edit warrior", who also was cautioned at least twice for their incivility on the talk page (their edit summaries containing some gibes, as well). Dhtwiki (talk) 04:03, 20 June 2020 (UTC)[reply]
Why IPs get all the blame from an edit war? Everyone involved shares responsibility. And, there was taunting from both sides. Dynamical IP addresses have nothing to do with this. In the end, consensus was reached in favor of IP edit contents. Resistance to the original IP constructive edit was most likely due to discrimination against IP. Would the edit be reverted if it was from an editor (with an account)? Probably not. You also cherry-picked 1 example out of many. 14.169.140.90 (talk) 08:55, 20 June 2020 (UTC)[reply]
You get the blame because you kept trying to change the article after a discussion started (which you did start, but after being asked to do so, after your first couple of attempts to change the article) but before agreement was reached, which is in violation of the WP:BRD cycle we try to follow (although often not that well). Your vitriol was greater and was probably first, or you would not have received the cautions you did. I think that dynamic addressing has a lot to do, because it sometimes makes it infuriatingly difficult to get in touch with disruptive editors, as opposed to an IP editor whose address doesn't change, and can easily seem devious. At issue was whether the article stated with sufficient precision whether quarks were found other than as constituents of hadrons, which was not a major flaw in the article, although you seemed to think it so (I didn't see an Aha! or D'oh! moment when others realized how badly things had been misstated; in other words, I didn't see you winning some major point). I took the first example you gave; were you less prone to doing something similar in the other articles? I didn't look at them all. Dhtwiki (talk) 12:39, 21 June 2020 (UTC)[reply]
It wasn't about winning. It was about making an improvement to the article. The issue was more than that; it seems that you didn't read all of the discussion. In the end, 90% of the points I raised were incorporated into the article. There was violation from both sides. There were a lot of false accusations against me too. For example, they accused me of using unreliable sources; I later proved them wrong. Anyway, that was beside the point. The point is that consensus was reached, and IP discrimination has occurred. If you want to play the blame game, I could have said that they started the edit war first because of IP discrimination. Your point about dynamic addressing is moot since I wasn't a disruptive editor. I received the caution "only" because I was an IP, which was a discrimination. Others have experienced discrimination as an IP too, read up their stories somewhere in this thread. 14.169.248.131 (talk) 15:28, 21 June 2020 (UTC)[reply]
I sometimes edit as an IP, if I am in an unsecure environment and all I want to do is fix some typos then it makes sense in security terms and I don't get problems doing such edits. However there is a (small) community overhead to my not logging in, we have a reputation based system and my account has earned a sufficient reputation that my edits are unlikely to be seen by hugglers et al. That lack of scrutiny reduces my chances of being accidentally reverted, it doesn't eliminate the risk, and I know it has happened at least once in my last few thousand logged in edits. By choosing not to create an account, or not to use an account, IP editors use more of the time of their fellow volunteers, and will experience a greater number of human errors by those patrollers than if they had created an account and built a reputation for doing good edits. This is not discrimination, and is a feature not a bug. If we as a community were discriminating against IPs, why would 31 hour blocks be mostly for IPs and indefinite blocks usually for accounts? ϢereSpielChequers13:29, 20 June 2020 (UTC)[reply]
You can dress it up as a feature, but it is still discrimination. Most blocked editors are not indefinite, which is only used for the most destructive editors (they're rare, not common). There is a good reason to not block IPs indefinitely; an IP can be shared by many different people, especially in public location. 14.169.248.131 (talk) 06:18, 21 June 2020 (UTC)[reply]
One reason that IPs get reverted more frequently than registered users is that IPs are significantly more likely to add information without including a citation to support it. Blueboar (talk) 19:37, 20 June 2020 (UTC)[reply]
There is a big difference between good reverts and bad reverts. Bad revert is reverting constructive edits (with reliable source) simply because it was made by an IP. 14.169.248.131 (talk) 06:21, 21 June 2020 (UTC)[reply]
So here is the banning/blocking policy on making edits behalf of such user, WP:PROXY.
Wikipedians in turn are not permitted to post or edit material at the direction of a banned or blocked editor (sometimes called proxy editing or proxying) unless they are able to show that the changes are either verifiable or productive and they have independent reasons for making such edits.
In a nutshell, you are not allowed to make edits solely behalf of banned/blocked editors, so you must have an independent reason for making such an edit. The only time when it would work is when a non-blocked/banned editor has consensus with a blocked/banned editor, so then making such a change is allowed because you do have an independent reason. But by the time there is consensus, the non-blocked/banned editor will have made the change soon.
So note that many people know about this; I'll just note it here for anyone to discuss the {{Edit partially-blocked}} and the WP:PROXY policy.
I believe you meant to link to the WP:PROXYING banning policy, not the policy on open proxies. In any case, I created the template at the off-wiki request of another editor, who can comment if they feel inclined to. In my opinion, partial blocks are used quite differently than full site bans/blocks. There are many cases where partial blocks are used instead of page protection, and in those cases it makes sense to have a clear method to allow non-COI edit requests. I think it's a fairly clear case of WP:IAR for a policy that hasn't been updated to match current practice. --AntiCompositeNumber (talk) 02:59, 19 June 2020 (UTC)[reply]
Oh boy, Wikipedia:Requests for comment/Partial blocks, and now WP:PROXYING. Another partial block RfC is coming. A follow-up RfC should be held to discuss any additional partial blocking policy elements not discussed in one of the survey subsections.— JJMC89 (T·C) 09:37, 11 January 2020 (UTC.
We don’t have RfCs on theoretical issues that no one has actually raised a non-theoretical concern about. Policy is based on practice, not vice versa. Let the practice develop and then if it needs to be documented, we can do it when the need arises. Likely without needing an RfC. TonyBallioni (talk) 05:53, 19 June 2020 (UTC)[reply]
Past images in articles
I would like to discuss what the policy is, or should be, on images that were once used in articles but no longer are. To my mind, such images should be retained, either here or on Commons, because they form part of the history of the page. Deleting old images breaks old versions of the page. It also removes some of the attribution, which becomes a licensing and legal problem for any reusers of old versions of the page relying on hyperlinks for attribution. In my view, we should also retain images used as part of talk page discussions. Sometimes, the discussion makes no sense without the image.
Some background; I deprodded around one thousand images back in May that had been mass proposed for deletion on the grounds that they were not used in any article (although the vast majority of them had once been so used) and the "poor quality" made them unsuitable for Commons. These deprods went largely unchallenged, but there is still a steady stream of simmilar cases turning up in CAT:PROD and the deprod sometimes gets challenged at Files for discussion, the latest case being Wikipedia:Files for discussion/2020 June 21#File:Null-balance voltmeter.png. SpinningSpark10:37, 21 June 2020 (UTC)[reply]
I dunno, how important is it for images to display in old versions of the page? In practice it may also be difficult to know about a file's past usage. Jo-Jo Eumerus (talk) 11:42, 21 June 2020 (UTC)[reply]
The importance of old images is exactly the same as importance of old text for precisely the same reasons (which are in my opening statement). If you think this is unimportant you need a better rationale than "dunno". As for "difficult to know", on the contrary, in the vast majority of cases it is very easy to find out. One has only to check the edit history of the uploader; the image is often used in an article in the very next edit. Not checking this is just pure laziness on the part of the nominator. SpinningSpark13:08, 21 June 2020 (UTC)[reply]
Not sure I understand why old images need to be deleted. If the image was good enough for an article at some point, then it should be moved across to Commons. If the quality is seen as poor, then tag it somehow as "needing a better version". If the image is re-used and somebody is motivated by the poor quality, then a new version will happen. — GhostInTheMachinetalk to me12:39, 21 June 2020 (UTC)[reply]
I'm assuming this is about fair-use images? One of the fair use conditions is for images to be used on pages. We can't be an image gallery for copyrighted materials. If items are free, they should be moved across to commons anyway, so I'm not sure what the policy change here is being proposed? Best Wishes, Lee Vilenski(talk • contribs)13:18, 21 June 2020 (UTC)[reply]
No, this is not about fair use. The images being discussed here were all nominated for quality reasons. They were all uploaded with a free license, usually self-created. SpinningSpark13:38, 21 June 2020 (UTC)[reply]
Depends on WHY the old image was removed. If it was simply because editors thought that a new image was better, then we have no problem with maintaining the old image somewhere (such as commons). However, if the old image was removed for cause (such as violating copyright), then we MUST purge it completely. Case by case situation. Blueboar (talk) 13:30, 21 June 2020 (UTC)[reply]
Agreed, of course, we should not keep copyrighted images. You say "we" have no problem keeping but the NFF page is absolutely causing such a problem. Its reasons for nominating include orphaned, obsolete, low quality, and unencyclopedic, all without any reference to policy or guidelines. All of these can include files that appear in article histories. And on the last criterion, I have created files in the past to clarify a talk page discussion. These are clearly unusable in an article but are open to deletion (some have been) under these criteria. SpinningSpark13:47, 21 June 2020 (UTC)[reply]
I believe that how we handle files is outside the rest of all parts of the other contributions to the history of a article page, and thus retention of older, unused files is not necessary, though we retain the file: space history aspects related to that file that relate to who had uploaded the file and the source/other details. All of our disclaimers to users, and our upload notifications all suggests that the hosting of images (and other AV file types) is wholly separate contributions from mainspace contributions, and while you are still agreeing to make your contribution to the project, it is not being tracked as part of the contribution of the article it will belong to. When looking at contributions to an article, it is not the image that we look at as the contribution, but the text/code that puts the image in place as the contribution, since that image can change independently after that text is added to the article (by a file update). A reuser of an article that is taking images would be required (if they are following the letter) to point to the histories of both the article and all used images to track contributions. So no, I don't think we're required to keep old images. That said, obviously we delete unused NFC, but I see no reason why we need to delete unused free images unless its clear they are completely unusable or otherwise clearly violate other policies (eg: graphic nudity beyond what's necessary for articles on human anatomy). Poor but not unusable is not a good reason to delete a free image --Masem (t) 13:57, 21 June 2020 (UTC)[reply]
Images not currently in use that are of lower quality than others that do not have any other (offsetting) added educational value are not in scope for commons. A recent dicussion on commons concluded that "image was formerly in use in a wikipedia article" was not a reasonable basis for keeping an image tagged for deletion. One of the reasons mentioned there is an important one unrelated to images directly...we routinely make widescale changes that make previous revisions of a page no longer "as they were at the time" when viewed in article-history. A simple example is a change to a template that makes some formerly-used field no longer visible (same effect as if an image in use at the time were deleted). Images solely for discussion of wikipedia issues and brainstorming (no encyclopediac or educational content) obviously aren't in scope for commmons...they should be tagged {{Keep local}} with some rationale so others will be less likely to mis-handle them. DMacks (talk) 14:13, 21 June 2020 (UTC)[reply]
Suggestion: a Codification of all policies of the that govern Wikipedia
I'm a new user and it was kind of duanting to see all the policies that Wikipedia has. What if we codified all of Wikipedia's policies and popular essays so that it would be easy for new users to know all of Wikipedia's policies. If you're a new user it could take days or weeks to read all of the guidelines Wikipedia has which each are ~30 min reads. This codification could over-simplify Wikipedia's policies. Yes, I see there are essays: Wikipedia:Summary of important policies and guidelines and Wikipedia:Simplified ruleset -- they don't address everything. I also suggest to tie this into the Template:Welcome. There maybe some organizational inertia.
Hi, while your idea is forward looking in concept it has several problems:
Any codification is a fork of the Rules/Policies/Guidelines/Generally accepted practices/essays and would quickly become out of date. We're contstantly revising and updating based on experiences we have.
very few new editors are going to spend ~10 hours reading through codifications to understand all the nuances of the corpus of behaviors before making their first edit.