Wikipedia talk:WikiProject Women in Red/Missing articles by dictionary/Dictionary of Women Worldwide
Notes
[edit]- I have been marking “Stub” where the entry is extremely thin and I hope leaving it on the list might encourage attention. This may be totally quixotic and I won’t be offended if consensus is we shouldn’t do that (I don’t expect anyone else to, necessarily).
- Antiqueight was kind enough to explain to this Luddite the method they’ve been using to sort and link (!!) the batches of names from the Dictionary index. I thought it could be helpful to save here for anyone who wants to take up this task.
- From Wikipedia_talk:WikiProject_Women_in_Red#Dictionary_of_Women_Worldwide:_25,000_Women_Through_the_Ages:
- ...basically, I dump the names into an excel file. Then you need to get it so that first name, surname and the date are in separate columns. That takes a little work. I use Data - Text to Columns. You'll see the details if you go to excel and follow along. I select the column with everything, click on Data, then click on Text to columns. The way will depend on how the data is laid out. Here I started with Fixed width to get rid of the asterisk (I would use search and replace and I could here in the wiki page but in windows asterisk is the default wildcard and so search and replace for asterisk doesn't tend to work). Then I use Delimited. I start with the comma - that separates surname from the rest. Then use Other -"(" that separates out the date. Tidy any weirdness up (I search and replace to remove the other bracket). Then I create a column with the equation =CONCATENATE("*","[[",B1," ",A1,"]]"," ",C1) (In my sheet A is the surname, B is the first name and C is now the date. This gives an output of the links. Dump them into the wiki, clean it up a bit and Ta-Da!. Huge long lists of names can be reordered and made into links as fast as 1 name. Now, that may not make sense so yell with any questions. There are, of course, various other ways to achieve this. This is how I did it this time. ☕ Antiqueight chatter 17:30, 9 March 2021 (UTC)
- Antiqueight, wow extremely valuable skills!! This is beyond my immediate abilities— saving these instructions so I can noodle around and try to learn! Another 101 question in the meantime—at the moment we have a 240ish-page pdf of names (which I can send you if simply looking at it is easiest), and when I copy-paste names into WP it doesn’t retain the list structure—example. Is that something your method could sort out or is it helpful if others (as I have been) manually split them into individual lines? (When I dump it into a spreadsheet I still have to manually split up the lines but clearly my Excel use is child’s play!) Innisfree987 (talk) 18:04, 9 March 2021 (UTC)
- So, Innisfree987 - with that output I'd use Word (I flick between preferred abilities on both - I am certain I could do it in one but ... well... I don't) and replace the final bracket ")" with a carriage return "^l". That would break it into new lines every time there was a ). Then I could copy and paste it into excel and use the instructions above but there is no need to remove the asterisk then as it hasn't been put in. Also I could replace the "(" with a comma and then when I do the Text to Columns and select the comma, it would do both the name and dates into new columns in one go. I have reasonably good excel skills but I am lazy - when I know how to do something I just grind it. There are probably much more elegant solutions but this works for me for now. ☕ Antiqueight chatter 18:33, 9 March 2021 (UTC)
- Antiqueight, wow extremely valuable skills!! This is beyond my immediate abilities— saving these instructions so I can noodle around and try to learn! Another 101 question in the meantime—at the moment we have a 240ish-page pdf of names (which I can send you if simply looking at it is easiest), and when I copy-paste names into WP it doesn’t retain the list structure—example. Is that something your method could sort out or is it helpful if others (as I have been) manually split them into individual lines? (When I dump it into a spreadsheet I still have to manually split up the lines but clearly my Excel use is child’s play!) Innisfree987 (talk) 18:04, 9 March 2021 (UTC)
- ...basically, I dump the names into an excel file. Then you need to get it so that first name, surname and the date are in separate columns. That takes a little work. I use Data - Text to Columns. You'll see the details if you go to excel and follow along. I select the column with everything, click on Data, then click on Text to columns. The way will depend on how the data is laid out. Here I started with Fixed width to get rid of the asterisk (I would use search and replace and I could here in the wiki page but in windows asterisk is the default wildcard and so search and replace for asterisk doesn't tend to work). Then I use Delimited. I start with the comma - that separates surname from the rest. Then use Other -"(" that separates out the date. Tidy any weirdness up (I search and replace to remove the other bracket). Then I create a column with the equation =CONCATENATE("*","[[",B1," ",A1,"]]"," ",C1) (In my sheet A is the surname, B is the first name and C is now the date. This gives an output of the links. Dump them into the wiki, clean it up a bit and Ta-Da!. Huge long lists of names can be reordered and made into links as fast as 1 name. Now, that may not make sense so yell with any questions. There are, of course, various other ways to achieve this. This is how I did it this time. ☕ Antiqueight chatter 17:30, 9 March 2021 (UTC)
- FWIW, the “Actress” section runs 13 pages, so, halfway done, and it will be a big bite out of the whole (236).
Other programming notes welcome! Innisfree987 (talk) 06:18, 11 March 2021 (UTC)
- I've written a Python script to do the Excelish transformation steps above in one go. Happy to send it if it's useful, or simply to run it when needed on this page. Dsp13 (talk) 20:54, 15 April 2021 (UTC)
- Dsp13, I am 95% Luddite: is it something I could install and run on WP or would I need to run it from a console on a computer? (I edit on mobile so could be a challenge.) Apologies if this is itself a very silly question—as I think about it, special:MyPage/common.js probably indicates that scripts for Wikipedia are written in JavaScript...anyway you see the level of non-technical-ness on my end! So the answer from me at least might be that I’d be grateful for you to run it. I could set up bigger batches so it’s needed less frequently? Whatever makes the least hassle for you! So many thanks. Innisfree987 (talk) 16:26, 17 April 2021 (UTC)
- It sounds as though the easiest thing is simply for me to run it whenever needed. Very happy to do that! Dsp13 (talk) 18:03, 17 April 2021 (UTC)
- I very gratefully accept this kind offer! Innisfree987 (talk) 20:58, 17 April 2021 (UTC)
A suggestion
[edit]Perhaps, additionally, for the if you want to help out rules, maybe we should put that if you find an article does exist for a red-linked name, you should make that red linked name variation into a redirect? SilverserenC 23:08, 11 March 2021 (UTC)
- Silver seren, good idea! I’ll add that now. Also everyone should feel free to revise... everything, we’re all just making this up together! Innisfree987 (talk) 23:12, 11 March 2021 (UTC)
- Eg if anyone can phrase it more concisely than I did, please do! Innisfree987 (talk) 23:17, 11 March 2021 (UTC)
Opinions sought — stubs?
[edit]Thus far when I find a really thin entry (like, I’m concerned it would not survive AfD—often just a few sentences and a list of performances), I’ve been leaving it in the list and marking as a Stub. My thought is that an entry at risk of deletion is kind of like not having an entry at all. But, I’m increasingly thinking: they all do seem to be marked as stubs in the entry itself, so if a person wanted to go improve stubs in a given category, they could already do that—they don’t need this list. So maybe for efficiency and to manage the length of the list (it figures to be maybe 5000 names long...) it’s better to keep it strictly to names that have no entry at all. (There are a few edge cases, eg where a woman is covered in a page on her family, but the stubs are mainly not that.) I welcome thoughts. Innisfree987 (talk) 16:53, 12 March 2021 (UTC)
- The above reasoning makes sense to me. I know I go through stub categories when I'm in an "expand" mood and through redlists when I'm in a "create" mood (strangely, not the same mood!
). I'm hoping that since these are reasonably connected to the corresponding Encyclopedia.com entries, it's not as hard to do a WP:BEFORE search for anyone thinking of putting something up for AfD anyway. DanCherek (talk) 17:02, 12 March 2021 (UTC)
- Two more good reasons. I think Ser was going to have a look through this weekend so maybe I’ll give him and others a few more days to weigh in, and if not we can just make the change. Am already wondering how we’ll manage a list with thousands of names to scroll through... may ultimately need to split across a few pages? I’m not sure. Innisfree987 (talk) 22:32, 12 March 2021 (UTC)
- I’m ready to call it. The redlinks alone will be such a vast list, and the stubs are searchable by other means. I’ll go back and remove the ones I left in, and won’t add anymore. Innisfree987 (talk) 01:27, 15 March 2021 (UTC)
- It's cool that we're successfully finding so many new names from this book that aren't on here yet (not cool they aren't on here yet, but progress is progress). Imagine how many we would have missed without looking into this dictionary. SilverserenC 01:33, 15 March 2021 (UTC)
- Silver seren, absolutely! Kudos to Ser Amantio di Nicolao for proposing and encouraging the undertaking. The sourcing policies are an obstacle to gender parity but plenty of women meet them who nevertheless don’t have entries yet. Thousands, it turns out! Great to have an index of so many! Innisfree987 (talk) 01:52, 15 March 2021 (UTC)
- It's cool that we're successfully finding so many new names from this book that aren't on here yet (not cool they aren't on here yet, but progress is progress). Imagine how many we would have missed without looking into this dictionary. SilverserenC 01:33, 15 March 2021 (UTC)
- I’m ready to call it. The redlinks alone will be such a vast list, and the stubs are searchable by other means. I’ll go back and remove the ones I left in, and won’t add anymore. Innisfree987 (talk) 01:27, 15 March 2021 (UTC)
- Two more good reasons. I think Ser was going to have a look through this weekend so maybe I’ll give him and others a few more days to weigh in, and if not we can just make the change. Am already wondering how we’ll manage a list with thousands of names to scroll through... may ultimately need to split across a few pages? I’m not sure. Innisfree987 (talk) 22:32, 12 March 2021 (UTC)
@Innisfree987: Thanks very much. (We haven't even gotten to the one whose name I found that started me on this thought process in the first place. :-) ) Apologies...I have not had the time, nor the wherewithal, to work on this this weekend. Perhaps in the next few days, though I expect to be fairly severely busy off-wiki for a week or two more. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 03:52, 15 March 2021 (UTC)
- As to the stub issue...I'm not sure but that it wouldn't be a bad idea to make note of the worst of them, only because this seems like a solid-enough source that merely citing it could be a boost to notability discussions. I'm often of the opinion that it's better to have some kind of a stub rather than nothing at all, and anything that can be used to bolster a "keep" argument at AFD could prove to be quite useful. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 03:54, 15 March 2021 (UTC)
- Ser Amantio di Nicolao, I’m definitely amenable to keeping a list of the worst stubs (and it would be very easy to collect the ones I just took out). What’s your thought on where we could most productively keep that list? I’m not sure AfDers will think/know to look on here... The best probably would be to have a list of the stubs and go ahead and add this source to each of them as its own project (some have it, actually, but don’t seem to have made much use). I suspect I’m lacking imagination here though! Welcome ideas. Innisfree987 (talk) 04:39, 15 March 2021 (UTC)
- @Innisfree987: I think your idea is best - it would be a good subpage of this one, I think. I'm less concerned about AFDers finding it; I think it would be just fine if they were pointed to it, instead. So what matters is that people contesting the deletion have access to it. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 05:22, 15 March 2021 (UTC)
- Ser Amantio di Nicolao, ah yes I see about AfD! Ok, solid—I started amassing them in my sandbox for starters but if you have a good idea for a subpage name, feel free to make one! I’ll chew on it, otherwise. I’m also wondering if we’re going to end up needing to split this page to manage the size—another naming Q to consider. Innisfree987 (talk) 05:30, 15 March 2021 (UTC)
- @Innisfree987: For another day, I'm afraid - I'm off to bed shortly. I'll think on the subpage question and see what I can come up with. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 05:34, 15 March 2021 (UTC)
- Ser Amantio di Nicolao, buona notte! This will certainly all still be here tomorrow, next week... ! Innisfree987 (talk) Innisfree987 (talk) 05:38, 15 March 2021 (UTC)
- @Innisfree987: For another day, I'm afraid - I'm off to bed shortly. I'll think on the subpage question and see what I can come up with. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 05:34, 15 March 2021 (UTC)
- Ser Amantio di Nicolao, ah yes I see about AfD! Ok, solid—I started amassing them in my sandbox for starters but if you have a good idea for a subpage name, feel free to make one! I’ll chew on it, otherwise. I’m also wondering if we’re going to end up needing to split this page to manage the size—another naming Q to consider. Innisfree987 (talk) 05:30, 15 March 2021 (UTC)
- @Innisfree987: I think your idea is best - it would be a good subpage of this one, I think. I'm less concerned about AFDers finding it; I think it would be just fine if they were pointed to it, instead. So what matters is that people contesting the deletion have access to it. --Ser Amantio di NicolaoChe dicono a Signa?Lo dicono a Signa. 05:22, 15 March 2021 (UTC)
- Ser Amantio di Nicolao, I’m definitely amenable to keeping a list of the worst stubs (and it would be very easy to collect the ones I just took out). What’s your thought on where we could most productively keep that list? I’m not sure AfDers will think/know to look on here... The best probably would be to have a list of the stubs and go ahead and add this source to each of them as its own project (some have it, actually, but don’t seem to have made much use). I suspect I’m lacking imagination here though! Welcome ideas. Innisfree987 (talk) 04:39, 15 March 2021 (UTC)
Copying to redlists
[edit]Now that we are progressing on this and filtering sections down to those without articles, it might be useful to copy names from completed sections to existing redlists. If there are no obvious crowd-sourced lists, they could always be added to Wikipedia:WikiProject Women in Red/Missing articles by nationality. If we just keep them in "Dictionary of Women Worldwide", they are unlikely to be picked up in relevant meetups. It might be a good idea to create new "crowd-sourced" redlists, for example for actress, re-ordering them by country. Any other ideas?--Ipigott (talk) 16:21, 19 March 2021 (UTC)
- Next steps has been on my mind too. I think it probably depends on how folks want to spend energy. For my money, so many blue links still need checking that I have a hard time imagining undertaking reordering by country—but if someone else wants to, don’t let me stand in the way! A few thoughts I did have:
- To save the labor, we could copy to (or create) occupational CS lists as a set mentioned in Dictionary of Women Worldwide rather than by nationality (I think I have seen similar at the bottom of some lists?)
- MarioGom had the good idea that perhaps once there were red links, someone might be interested in adding them to Wikidata, which would help ensure they show up on relevant events. We could advertise for interest back at main Talk.
- For now at least I think it would be preferable to keep this list intact (rather than deleting redlinks copies elsewhere) in case it makes sense to copy them to more than one place, among other reasons.
- But I am open to revising my opinions on all—the best way to manage a list this large is not obvious! Innisfree987 (talk) 16:51, 19 March 2021 (UTC)
- Innisfree987 I fully agree we should keep this list of redlinks intact, even if we add some of them to other lists. The fact that they are listed by occupation is indeed very useful. The reason I am interested in nationality is that I know quite a few of us concentrate on people from certain countries when creating new biographies. And don't forget we are also soon to embark on our European continental contest. As perhaps 90% of the women listed are from the English-speaking world, even if quite a few were born elsewhere, it is difficult to pick out those from other countries as so many American names originated in Europe. As you say, the ideal solution would indeed be to include them in Wikidata if they are not already there. I wonder if MarioGom has any ideas on how this could be handled more efficiently, perhaps even semi-automatically on the basis of the info we already have. Once we have clearer views on how to proceed, I agree we should advise on progress and future developments on the main WiR talk page. (cc Oronsay, Tagishsimon, Ser Amantio di Nicolao and Rosiestep)--Ipigott (talk) 11:08, 20 March 2021 (UTC)
- Min'n'match is a Wikidata tool designed for this kind of cases. The idea is that a bulk list of items is loaded in the tool, and then various editors can collaborate on the list by checking items and linking them to existing Wikidata entities or creating new ones when needed. I haven't loaded a Mix'n'match dataset before, but it's probably the right tool for the job. I wonder if other Wikidata experts here know more about this. We can probably generate the Mix'n'match dataset automatically based on the current list we have in this page. MarioGom (talk) 12:24, 20 March 2021 (UTC)
- Innisfree987 I fully agree we should keep this list of redlinks intact, even if we add some of them to other lists. The fact that they are listed by occupation is indeed very useful. The reason I am interested in nationality is that I know quite a few of us concentrate on people from certain countries when creating new biographies. And don't forget we are also soon to embark on our European continental contest. As perhaps 90% of the women listed are from the English-speaking world, even if quite a few were born elsewhere, it is difficult to pick out those from other countries as so many American names originated in Europe. As you say, the ideal solution would indeed be to include them in Wikidata if they are not already there. I wonder if MarioGom has any ideas on how this could be handled more efficiently, perhaps even semi-automatically on the basis of the info we already have. Once we have clearer views on how to proceed, I agree we should advise on progress and future developments on the main WiR talk page. (cc Oronsay, Tagishsimon, Ser Amantio di Nicolao and Rosiestep)--Ipigott (talk) 11:08, 20 March 2021 (UTC)
- As there isn't an endless supply of volunteer time and inclination to do this work, first and forement, I'd like to see all of the women named in this (and other dictionaries in the Women in Red redlist index) to have an item in Wikidata based on their inclusion in a specific dictionary. This way, the names are available to the 'world' and not just to someone who knows about and can find Women in Red redlists. Pinging Gamaliel who is a Wikidata yoda and might be able to better explain how this would be set up -- "article in [Foo book]" -- as per this example.
- In addition, adding the redlinked names to crowd-sourced Women in Red redlists is a bonus for the reasons Ipigott mentioned. Agree with you, Innisfree987, regarding keeping all the names intact in the current list. Adding Megalibrarygirl, our Librarian in Residence, who might have additional comments on this topic.
- Using the A Woman of the Century example which supports a CS and a WD redlist for all the named women, see: Wikipedia:WikiProject Women in Red/Missing articles by dictionary/Woman of the Century (CS) and Wikipedia:WikiProject Women in Red/Missing articles by dictionary/A Woman of the Century (WD). --Rosiestep (talk) 15:21, 20 March 2021 (UTC)
- Thank you all so much—this all sounds wonderful! Am a bit under the weather (post-jab, I’m counting it as a good thing!), so I will wait to look at the tool until I can think straighter (also doubtless many of those above are far more capable than I am on a good day!) I do know the Dictionary of Women Worldwide is already recognizable in WD; I tried it out here (in the “described by source” field—unless there’s a better way?) Super! Very excited these could be made broadly available. And maybe this could make country-sorting more manageable, a need that makes much sense Ipigott—some will surely already be in WD, so those needing to country-checking would be a smaller set. Innisfree987 (talk) 16:00, 20 March 2021 (UTC)
- Innisfree987, Rosiestep I am all for anyone working to include more women on redlists. I think it's good to have as many different kinds of redlists as people need--multiple points of discovery can be very helpful to people who are browsing for information. So it's OK if there are many different ways to discover women and if they are listed on more than one list. Megalibrarygirl (talk) 16:40, 20 March 2021 (UTC)
- I use the described by source field exactly the way Innisfree987 described. I've used that field to generate a number of WIR listeria lists for particular reference works. What I do is take the list of names in the reference work and dump it in OpenRefine, which matches it (with a lot of manual tweaking and verification) to existing Wikidata items, then lets me add the described by source field to all of them at once. I can also mass create items for the missing people. Be glad to do this for anyone who has a book they are interested by doesn't want to go through all that work. Gamaliel (talk) 02:46, 21 March 2021 (UTC)
- Gamaliel, wow that’s incredible. So, what we have here is roughly the first 20 pages of a 236-page index, each with a couple hundred names (the total is actually more than 25,000 names because some people are listed under multiple occupations, altho we could get a different version of the index, without occupations). Still that seems like a lot to process given checking you mention! Would you rather we carry on as we have been and just give you what we come up with for redlinks? I’m guessing that will be about 5000 names. Immensely grateful for your offer, whatever works for you would be terrific. Can also forward you the index so you can size it up! Innisfree987 (talk) 07:16, 21 March 2021 (UTC)
- 25000! That's quite a bit larger than any of the projects I've worked on before but I'm willing to give it a try and at least get started. Gamaliel (talk) 13:03, 21 March 2021 (UTC)
- I did a quick test with the first section of the redlist, the seven abolitionists. Of them, six were already in Wikidata and I created a new item for the seventh. Three of those six already have Wikipedia articles so I created redirects. OpenRefine matched them to variants of their names. (This only works when those variants are already in the alias field, hence the necessity for manual checking.) So this seems more doable now. The format of the redlist with the dob/dod is very helpful for this; is the index this way as well or was getting it in this format your hard work? Gamaliel (talk) 13:19, 21 March 2021 (UTC)
- Gamaliel, so much appreciated! I think the indices all have the dates included (there are three versions to choose from—era, nationality or occupation and I’ve got occupation, but one of the others might be better for what you’re describing, because I assume it won’t have repeats like occupations does.)
What I’m wondering is how much bigger the task is for you working from the full index versus what we’ve winnowed as red links—would you have to manually check 25,000 names?! I worry maybe yes just because in what we’ve done so far, the auto-generated blue links point to the wrong person maybe 2-5% of the time? which is why we’re still checking every single one (to catch false-negatives, so to speak) although I admit I have considered whether it’s worth it for this manual list. But for WD, if batch-adding the “described by source” field, even that small % could introduce a fairly large number of errors in a set this big, so would you have to check them all on your own?? Or is there a technical solution to this problem? If not, we could carry on as we have been winnowing to a redlinks list and just ask you to put those into WD? That way the amount of manual checking is spread around a bit rather than all on you!
I’ll send you the index to consider though—I am guessing I may not be fully groking the process and you’ll see better than I what’s required. Additionally you can see the full list of Abolitionists, for comparison, in this diff. Tell me if I can answer any Qs/spell anything out about what we’ve done so far more clearly! Innisfree987 (talk) 20:10, 21 March 2021 (UTC)- Innisfree987 If you and others were willing to remove the blue linked names, that would significantly reduce my workload. Since our goal is to spur the creation of en.wp articles we can dispense with matching the existing articles. We need not edit all 25k items on wikidata. As far as false positives, having both occupation and dates of birth & death is the optimum scenario for matching. so I don't see many mismatches happening. Far more likely is a duplicate wikidata item, but those are easily fixed with a merge later on. Gamaliel (talk) 04:17, 22 March 2021 (UTC)
- Gamaliel, agree completely we need not edit all 25k WD items. Thanks to Antiqueight’s work creating the batches of links, we should be able to give you just the red links one way or another (or at least a good chunk of them, if not the whole index.) So I have a sense, are the ones we’ve checked so far (about the first five pages) a good place for you to start or would you prefer a larger batch? Innisfree987 (talk) 13:22, 22 March 2021 (UTC)
- Innisfree987 Whatever is easiest, I'm fine regardless of the size of the batch. I've been working on them by occupation so there's a natural limit to the size of the batch anyway. Gamaliel (talk) 13:44, 22 March 2021 (UTC)
- Gamaliel, ah! Are you already ahead of our pages-checked then? (I’ll go mark now for clarity, per Ipigott’s suggestion.) Have been weighing how much sense it makes to keep combing through all of the blue links manually when we yield relatively few redlinks for our efforts, and if your process is quicker than we can keep up with, that would really be a vote for just axing them en masse to keep the train moving! Innisfree987 (talk) 15:16, 22 March 2021 (UTC)
- Innisfree987 Oh, no! I've only just started. I've been skipping around, I think I've only done abolitionists and aviators. Gamaliel (talk) 17:28, 22 March 2021 (UTC)
- Gamaliel Oh ok! Me too :) Just wave if/when you run out of checked sections and we can reevaluate. Innisfree987 (talk) 17:37, 22 March 2021 (UTC)
- Innisfree987 Oh, no! I've only just started. I've been skipping around, I think I've only done abolitionists and aviators. Gamaliel (talk) 17:28, 22 March 2021 (UTC)
- Gamaliel, ah! Are you already ahead of our pages-checked then? (I’ll go mark now for clarity, per Ipigott’s suggestion.) Have been weighing how much sense it makes to keep combing through all of the blue links manually when we yield relatively few redlinks for our efforts, and if your process is quicker than we can keep up with, that would really be a vote for just axing them en masse to keep the train moving! Innisfree987 (talk) 15:16, 22 March 2021 (UTC)
- Innisfree987 Whatever is easiest, I'm fine regardless of the size of the batch. I've been working on them by occupation so there's a natural limit to the size of the batch anyway. Gamaliel (talk) 13:44, 22 March 2021 (UTC)
- Gamaliel, agree completely we need not edit all 25k WD items. Thanks to Antiqueight’s work creating the batches of links, we should be able to give you just the red links one way or another (or at least a good chunk of them, if not the whole index.) So I have a sense, are the ones we’ve checked so far (about the first five pages) a good place for you to start or would you prefer a larger batch? Innisfree987 (talk) 13:22, 22 March 2021 (UTC)
- Innisfree987 If you and others were willing to remove the blue linked names, that would significantly reduce my workload. Since our goal is to spur the creation of en.wp articles we can dispense with matching the existing articles. We need not edit all 25k items on wikidata. As far as false positives, having both occupation and dates of birth & death is the optimum scenario for matching. so I don't see many mismatches happening. Far more likely is a duplicate wikidata item, but those are easily fixed with a merge later on. Gamaliel (talk) 04:17, 22 March 2021 (UTC)
- Gamaliel, so much appreciated! I think the indices all have the dates included (there are three versions to choose from—era, nationality or occupation and I’ve got occupation, but one of the others might be better for what you’re describing, because I assume it won’t have repeats like occupations does.)
- Gamaliel, wow that’s incredible. So, what we have here is roughly the first 20 pages of a 236-page index, each with a couple hundred names (the total is actually more than 25,000 names because some people are listed under multiple occupations, altho we could get a different version of the index, without occupations). Still that seems like a lot to process given checking you mention! Would you rather we carry on as we have been and just give you what we come up with for redlinks? I’m guessing that will be about 5000 names. Immensely grateful for your offer, whatever works for you would be terrific. Can also forward you the index so you can size it up! Innisfree987 (talk) 07:16, 21 March 2021 (UTC)
- Thank you all so much—this all sounds wonderful! Am a bit under the weather (post-jab, I’m counting it as a good thing!), so I will wait to look at the tool until I can think straighter (also doubtless many of those above are far more capable than I am on a good day!) I do know the Dictionary of Women Worldwide is already recognizable in WD; I tried it out here (in the “described by source” field—unless there’s a better way?) Super! Very excited these could be made broadly available. And maybe this could make country-sorting more manageable, a need that makes much sense Ipigott—some will surely already be in WD, so those needing to country-checking would be a smaller set. Innisfree987 (talk) 16:00, 20 March 2021 (UTC)
- Innisfree987: It seems to me that it would indeed be a good idea to delete all the blue-linked items which have been checked. To be on the safe side, it might be useful to add "Checked" or "All above checked" where appropriate. Then we can go ahead and delete them. If anyone were to be interested in retrieving the deleted blue links, they would simply need to retrieve earlier versions from the edit history.--Ipigott (talk) 12:49, 22 March 2021 (UTC)
- Ipigott, I might be misunderstanding but I think that’s what’s happening? I’ve been using the edit summary “verified” for deleting blue links; and then marking very thin stubs and removing to my sandbox in batches pending a conversation with Ser Amantio di Nicolao who was interested to see them (but busy offline just now I think). I’m afraid there aren’t many checked blue links still on the page, unless you have checked some without removing?
However, slow progress makes me wonder if we need to consider removing them without checking. In my checking, I find that perhaps 5% (or fewer) are directed to the wrong person—BUT, when corrected, many are still blue links. So simply deleting without checking would only miss a small number of redlinks (plus those stubs). It may be the only realistic way. I’ve been doing this for about 10 days and it’s not 10% done which is worrying. Innisfree987 (talk) 13:33, 22 March 2021 (UTC)
- Ipigott, I might be misunderstanding but I think that’s what’s happening? I’ve been using the edit summary “verified” for deleting blue links; and then marking very thin stubs and removing to my sandbox in batches pending a conversation with Ser Amantio di Nicolao who was interested to see them (but busy offline just now I think). I’m afraid there aren’t many checked blue links still on the page, unless you have checked some without removing?
- Innisfree987: Yes, I know from your edit summaries you've been checking them but I thought it might help if you could add "Checked" (or "Verified") to the list itself, perhaps one section or page at a time, so that we don't start deleting items which have not been checked. I haven't been deleting any blue links myself as I'm never sure whether they've been checked or not. I've also been going through quite a few items which looked as if they could already be included, making redirects or minor typographical changes to have them correctly blue linked where possible. As for those linked to the wrong person, if you find them, then I suggest you add dates or occupation in brackets to make them redlisted but if this is going to take too much time, then I would suggest we simply delete them for the time being. I think it's important we try to advance fairly quickly or we'll be spending all year on the job.--Ipigott (talk) 13:54, 22 March 2021 (UTC)
- Oh! Ok if I’m understanding properly: I know for myself and I believe for DanCherek and Silver Seren, we’re deleting all the blue links we’ve checked. So any blue links left in the list still need to be checked, sorry to say (if we’re going to check them), except a few marked “Stub” that I haven’t moved out yet and occasionally a different note of uncertainty. And yes for now if we find one that’s misdirected we always disambiguate, although like I say that only sometimes leads to a redlink. Much of the time the person had an entry under a different name. Correcting the few misdirected blue links isn’t really burdensome; the bottleneck is in checking the other 95% of blue links looking for the relatively rare mistakes. Innisfree987 (talk) 15:05, 22 March 2021 (UTC)
- All that said I can definitely mark the first five pages checked, that’s no problem and might help Gamaliel too even if it doesn’t indicate blue links that need removal. Innisfree987 (talk) 15:14, 22 March 2021 (UTC)
- Oh! Ok if I’m understanding properly: I know for myself and I believe for DanCherek and Silver Seren, we’re deleting all the blue links we’ve checked. So any blue links left in the list still need to be checked, sorry to say (if we’re going to check them), except a few marked “Stub” that I haven’t moved out yet and occasionally a different note of uncertainty. And yes for now if we find one that’s misdirected we always disambiguate, although like I say that only sometimes leads to a redlink. Much of the time the person had an entry under a different name. Correcting the few misdirected blue links isn’t really burdensome; the bottleneck is in checking the other 95% of blue links looking for the relatively rare mistakes. Innisfree987 (talk) 15:05, 22 March 2021 (UTC)
- I'd ask that we keep all of the red links if we can. Gamaliel (talk) 17:28, 22 March 2021 (UTC)
- Yes I think we have a unanimous vote for not deleting any of this page’s redlinks—they may be copied elsewhere but not removed. Innisfree987 (talk) 17:36, 22 March 2021 (UTC)
- Couldn't we just move all unchecked blue links to a subpage that we can work on later, leaving just the red links (and the few blue links we've already confirmed go to the wrong person) here on this page to use for the Wikidata tool stuff you all have been discussing? SilverserenC 18:53, 22 March 2021 (UTC)
- I think that is a genius idea. Innisfree987 (talk) 20:11, 22 March 2021 (UTC)
- Couldn't we just move all unchecked blue links to a subpage that we can work on later, leaving just the red links (and the few blue links we've already confirmed go to the wrong person) here on this page to use for the Wikidata tool stuff you all have been discussing? SilverserenC 18:53, 22 March 2021 (UTC)
- Yes I think we have a unanimous vote for not deleting any of this page’s redlinks—they may be copied elsewhere but not removed. Innisfree987 (talk) 17:36, 22 March 2021 (UTC)
- Innisfree987: Yes, I know from your edit summaries you've been checking them but I thought it might help if you could add "Checked" (or "Verified") to the list itself, perhaps one section or page at a time, so that we don't start deleting items which have not been checked. I haven't been deleting any blue links myself as I'm never sure whether they've been checked or not. I've also been going through quite a few items which looked as if they could already be included, making redirects or minor typographical changes to have them correctly blue linked where possible. As for those linked to the wrong person, if you find them, then I suggest you add dates or occupation in brackets to make them redlisted but if this is going to take too much time, then I would suggest we simply delete them for the time being. I think it's important we try to advance fairly quickly or we'll be spending all year on the job.--Ipigott (talk) 13:54, 22 March 2021 (UTC)
Poll: Move blue links to a subpage?
[edit]Silver seren had the (IMO) very good idea to move blue links as a group to a separate subpage. This would give Gamaliel expedited access to straight redlists to input to Wikidata, while also allowing the interested to comb the blue links for any misdirects or very poor stubs, at their leisure. Before we take the plunge: anyone watching the page see any reason not to? Innisfree987 (talk) 22:05, 22 March 2021 (UTC)
- Ok, so absent objection and not being sure that this is the sort of thing WiR usually hosts, I made User:Innisfree987/Dictionary of Women Worldwide blue links and just copied over everything we have. So now we can simply remove the blue links here and any energy left over can be directed to combing the other list for misdirects that might not have an entry yet. Innisfree987 (talk) 17:47, 24 March 2021 (UTC)
- Ok I’ve adjusted the instructions and will start taking the plunge of removing blue links here rn masse. Does seem like the only realistic path forward. Innisfree987 (talk) 22:01, 24 March 2021 (UTC)
- Just to note I am provisionally saving all new blue links to User:Innisfree987/Dictionary of Women Worldwide blue links as Ser plans to look through them shortly and then we may further reassess. Innisfree987 (talk) 18:31, 28 March 2021 (UTC)
- Just a note to say it makes sense to me. ☕ Antiqueight chatter 08:17, 29 March 2021 (UTC)
- Question for you Antiqueight. When you convert names to wikilinks, does that method allow for automatic sorting the red links from the blue? I.e. to potentially generate a list of just red links?
I ask because it looks like Ser is too busy to get into the question of processing the blue links and ironically enough, it takes me enough time to sort them manually—even without checking—that I don’t have time left to do anything with them, so I’m thinking we’re not gaining a lot from all that time spent doing that. (Also, when I was checking, the red links identified were quite rare, maybe 1% of all names. So not a great deal wishes getting missed by not looking at them.)
So, if it were possible from your method just to get the red inks, I’m thinking that’s the more sensible thing to do. But maybe manually separating is just required! Thanks for any info, and ALL your work! Innisfree987 (talk) 21:55, 1 April 2021 (UTC)- PS, CC Silver seren who has good ideas about these things! Innisfree987 (talk) 21:55, 1 April 2021 (UTC)
- No, the creation of the links is a purely mechanical process which is unable to check if they are red or blue. I haven't found a simplistic way to even separate them out so that you could have a list of blue links to be checked without doing it by hand. It IS intensely tedious and you'll notice I have cruelly and selfishly been leaving that up to you and others. Perhaps someone else knows of a way to do it? ☕ Antiqueight chatter 08:06, 2 April 2021 (UTC)
- Oh hardly! Your efforts have been heroic. And I can only speak for myself but I don’t find the separating too terribly tedious, just very time-consuming. Probably a personal problem, I need to be more disciplined about putting it down to go work on something else sometimes, so I don’t stop writing articles completely.
Meanwhile though, good point—if there were an easy way to separate, we wouldn’t have to lose the blue links; we’d just make two lists. Well maybe Ser will have a free moment—I have the impression he is a genius for AWB, which I’ve never used, but maybe there’s a novel way to apply it here? All ideas from anyone else reading very welcome! Innisfree987 (talk) 15:28, 2 April 2021 (UTC)
- Oh hardly! Your efforts have been heroic. And I can only speak for myself but I don’t find the separating too terribly tedious, just very time-consuming. Probably a personal problem, I need to be more disciplined about putting it down to go work on something else sometimes, so I don’t stop writing articles completely.
- No, the creation of the links is a purely mechanical process which is unable to check if they are red or blue. I haven't found a simplistic way to even separate them out so that you could have a list of blue links to be checked without doing it by hand. It IS intensely tedious and you'll notice I have cruelly and selfishly been leaving that up to you and others. Perhaps someone else knows of a way to do it? ☕ Antiqueight chatter 08:06, 2 April 2021 (UTC)
- PS, CC Silver seren who has good ideas about these things! Innisfree987 (talk) 21:55, 1 April 2021 (UTC)
- Question for you Antiqueight. When you convert names to wikilinks, does that method allow for automatic sorting the red links from the blue? I.e. to potentially generate a list of just red links?
- Just a note to say it makes sense to me. ☕ Antiqueight chatter 08:17, 29 March 2021 (UTC)
Sections marked done
[edit]Sections marked "done" are sections where I've added all the redlinks to Wikidata. If any new names are added to any of these sections (such as blue links that were found to be actually red when checked), please mark them somehow and I will be glad to add them to Wikidata as well. Thanks! Gamaliel (talk) 19:05, 1 April 2021 (UTC)
- Maybe just put (not checked) next to those name cases if they come up, I guess? SilverserenC 19:25, 1 April 2021 (UTC)
Wikidata and redirects
[edit]Question for you Gamaliel. If I spy an obvious misspelling like a professor named Catherine A. Mackinnon (surely Catharine A. MacKinnon), should I just make the redirect and remove the blue link? Or is that something you would add to Wikidata? Thank you—I only know the broad strokes of how WD works so I’m unsure of their policies on whether this is an “alias”, something else, or just not something they include at all. Innisfree987 (talk) 19:26, 6 April 2021 (UTC)
- Adding aliases is very useful IMHO in adding to discoverability and I've been adding alternate spellings as aliases to Wikidata when I think it's a transliteration issue (like all the Russian names) or a legit alternate spelling, but misspellings like this I've been ignoring. Gamaliel (talk) 22:30, 6 April 2021 (UTC)
- Thank you—that all makes sense! Will correct accordingly. Innisfree987 (talk) 00:27, 7 April 2021 (UTC)
Gunnhild fl. 1150s
[edit]I am baffled by this one, as I keep finding contradictory parentages when I try to figure out who this is and connect her to other Norwegian royalty. Help! Can anyone untangle this? Gamaliel (talk) 04:14, 8 April 2021 (UTC)
- Having the same issue with Ida Plantagenet. Gamaliel (talk) 21:24, 11 April 2021 (UTC)
- Thank you for pressing forward with these—just making sure I’m not deleting the wrong blue links for the pre-moderns is dizzying to me! Innisfree987 (talk) 21:49, 11 April 2021 (UTC)
- For most of the pre-modern royalty, it's easy to confirm who is who due to family relationships. Most of the British peerage and much of the European peerage has been imported into Wikidata, so I can just look up which Isabel married the third Earl of Whatever. A few have confounded me though. Gamaliel (talk) 22:55, 11 April 2021 (UTC)
- Thank you for pressing forward with these—just making sure I’m not deleting the wrong blue links for the pre-moderns is dizzying to me! Innisfree987 (talk) 21:49, 11 April 2021 (UTC)
"(Checked)"?
[edit]Looking at this splendid list for the first time I was mystified by the "(Checked)" annotation to various sections. Ploughing through the talk page (hard work on a phone), as well as being much impressed by the amount of work a lot of people have put in, I gather it means something like "Any blue links in this section have been checked to see that they are the right link" and/or possibly "These red links have been checked to see that the person isn't listed under a variant name or typo"? Could there please be a note at the top of the list to explain and define, for editors coming new to it? Thanks. PamD 07:44, 11 April 2021 (UTC)
- PamD, thanks so much for reading all that! I have updated the instructions (tell me what you think), but basically yes: when we started, we were checking every single blue link to verify it directed to the person listed in the DoWW, and ... it moved so slowly that we had to change course. Now each time we get a batch of new links, I save the whole to user:Innisfree987/Dictionary_of_Women_Worldwide_blue_links for possible later examination, but for now the priority here is to clear out blue links so the red links can go into Wikidata. I think we could even just remove the “(checked)” marks here (they’d still be on the other page, where they are more relevant), unless anyone disagrees?
In any case thank you for this question and please do ask any others to make it more accessible for anyone interested to help! Innisfree987 (talk) 12:04, 11 April 2021 (UTC)- I went ahead and removed the (checked) indications. I don’t think we’re losing anything really. Innisfree987 (talk) 22:06, 11 April 2021 (UTC)
Fork?
[edit]Hi all,
My sense is that we’re probably reaching the outer limit of how long a list can be before it becomes too unwieldy to be useful. It will (already is! Hooray!) shrinking over time, but for the time being, I’m thinking at the end of the “F” occupations, it’s probably time for a fork. First of all, does that seem correct? And secondly, any strong feelings on how we set it up? I’m thinking to keep this as a main directory page and then make eventually three or four subpages (A to F; G to P; Q to Z, roughly, but we could do four instead?) Also I don’t know if it’s preferable to do structure as subpages versus separate pages that just refer to one another? (Does that distinction make sense?)
Anyway these are my preliminary thoughts: what say you? Innisfree987 (talk) 23:32, 14 April 2021 (UTC)
- I agree the page is getting big (and adding name variants only makes it bigger) so that it would be a good idea to split into several. The options you suggest all sound good - I don't have any strong preference between them Dsp13 (talk) 20:51, 15 April 2021 (UTC)
- Ipigott, may I run this by you before I make the move? I am thinking to make this a directory page and then have subpages called “Occupations A to F” and “Occupations G to P” (tentatively on the second—there are an awful lot of “P” professions so maybe G to O would be better?) All to say, would be grateful for the benefit of your experience handling red lists, if you have a moment! Innisfree987 (talk) 21:29, 15 April 2021 (UTC)
- I agree with that plan. Pratically speaking, I think the best thing to do would be to move this project page to an "Occupations A-F" page. Then the history will move with it. You can then overwrite the redirect page to make a new directory page. Dsp13 (talk) 21:45, 15 April 2021 (UTC)
- Also, as far as breaking into roughly equal sections, can we use page numbers to help decide the divisions? If occupations start on page 2337, what page do they end on? What letters do we get if we divide that page span into four? (Phone books used to be A to D, E to K , L to R and S to Z, if I remember right.) Dsp13 (talk) 21:50, 15 April 2021 (UTC)
- So a four-way split would look roughly like:
- A to E (63 pages)
- F to N (56 pp)
- P to R (59 pp)
- S to Z (57 pp)
- I have a mild preference for fewer subpages especially since they will shrink over time but I’m not wedded to it, especially if those with fresher eyes find the length just too much. Innisfree987 (talk) 01:31, 16 April 2021 (UTC)
- So a four-way split would look roughly like:
- Totally agree and happy with what ever split you come up with.... working on some of the entries that I feel comfortable writing as fast as I can ;-) ☕ Antiqueight chatter 22:06, 15 April 2021 (UTC)
- Can’t tell you how relieved I feel every time I see one—the occupations are so useful IMO but they also repeat, so every name turned blue saves us time down the line! Superb. Innisfree987 (talk) 01:31, 16 April 2021 (UTC)
- Innisfree987: You're doing a wonderful job on this and I am happy to go along with any suggestions you make. I also think it's a good idea to split it up into three separate pages, especially as these will be easier to manage during the preparatory work. I just wonder though, when we've completed all our checks, whether it might not be a good idea re-establish one main page covering all the red links for all the occupations. This would make it easier for people to make just one search to see if a given name is covered anywhere in the dictionary. (I've also been wondering whether an alphabetic index on all the redlinked names might be even more useful but we can come back to that when we've worked through the entire index.) In parallel, thanks to the work being undertaken by Gamaliel, it should also become increasingly easy to see from Wikidata whether a given name is covered in the dictionary. While most names should appear under our more general redlink headings, for some of the occupations, it may be useful to create new redlink pages (or extend the scope of some of the existing ones). I'll continue to work on the unrecognized character problems and on redirects but please let me know if you need assistance with anything else.--Ipigott (talk) 06:54, 16 April 2021 (UTC)
- Thank you so much Ipigott, I had similar thoughts about the ideal end results. Let’s do three forks for now which will be easier to put back together at the end if it seems sensible by then, or at least, not so many to navigate among. (I thought perhaps we could end up listing the occupations at least on the directory page...) Thanks for all your work, immensely valuable! Innisfree987 (talk) 07:45, 16 April 2021 (UTC)
- Innisfree987: You're doing a wonderful job on this and I am happy to go along with any suggestions you make. I also think it's a good idea to split it up into three separate pages, especially as these will be easier to manage during the preparatory work. I just wonder though, when we've completed all our checks, whether it might not be a good idea re-establish one main page covering all the red links for all the occupations. This would make it easier for people to make just one search to see if a given name is covered anywhere in the dictionary. (I've also been wondering whether an alphabetic index on all the redlinked names might be even more useful but we can come back to that when we've worked through the entire index.) In parallel, thanks to the work being undertaken by Gamaliel, it should also become increasingly easy to see from Wikidata whether a given name is covered in the dictionary. While most names should appear under our more general redlink headings, for some of the occupations, it may be useful to create new redlink pages (or extend the scope of some of the existing ones). I'll continue to work on the unrecognized character problems and on redirects but please let me know if you need assistance with anything else.--Ipigott (talk) 06:54, 16 April 2021 (UTC)
- Can’t tell you how relieved I feel every time I see one—the occupations are so useful IMO but they also repeat, so every name turned blue saves us time down the line! Superb. Innisfree987 (talk) 01:31, 16 April 2021 (UTC)
Name variants
[edit]While we’re all here, maybe a quick query on how to handle name variants. I find them particularly helpful on the aristocrats that I can’t keep straight, really helps make clear which have pages and which not (my republican biases thank you for the hand!) Do others find it generally helpful (for those or in other cases)? As far as end users of the redlink list, is it helpful for all sections to list name variants? As Dsp13 mentions there’s a tradeoff with the length of the list (and also the extra work!) Gamaliel, I don’t know if it affects your method for entering the list into Wikidata but thought I’d check! Anyway, thoughts welcome! Innisfree987 (talk) 08:01, 16 April 2021 (UTC)
- The name variants would help a bit, but with aristocrats I find myself having to look at the entry almost every time and match it with family relationships anyway, so I wouldn't want you to do the extra work just for me. However, I could use that work to add all the variants to Wikidata as aliases. Gamaliel (talk) 00:35, 17 April 2021 (UTC)
- Name variants seem helpful both for Wikidata and just with general redirects and confirmation that there isn't an article already somewhere out there under one of the many possible names. SilverserenC 01:18, 17 April 2021 (UTC)
Ran my first batch with the name variants. Advantages: Open refine can search the names and the aliases simultaneously, increasing the possibility of a positive match. I can also add the alternate names as aliases to Wikidata easily, enriching those items. Disadvantages: It's a lot harder to set up the data in Openrefine in this format. I hate to be a bear, but is it possible we could add some kind of unique separator between the names and the dates? Anything that isn't already used like a slash or a dash. That would make setting the data up in Openrefine significantly easier. Gamaliel (talk) 16:04, 18 April 2021 (UTC)
- Gamaliel, I was worried that could happen! I am happy to go add separators, no problem. So really any character? Maybe we could go back to having dates inside parentheses, would that do the trick? If you have a suggestion I’m all ears! Innisfree987 (talk) 16:18, 18 April 2021 (UTC)
- Thank you! Parenthesis will be fine, but any unique character will do. Sometimes Openrefine can automatically parse the data into separate columns, and sometimes it can't. When it can't I have to either do it manually (which is horrifically difficult in OpenRefine) or tell OpenRefine to split the data into columns at a particular character. Gamaliel (talk) 17:54, 18 April 2021 (UTC)
- Yup makes sense! Ok I will start working on those. Innisfree987 (talk) 18:01, 18 April 2021 (UTC)
- Ah a question: in a section like this wikipedia:WikiProject_Women_in_Red/Missing_articles_by_dictionary/Dictionary_of_Women_Worldwide/Occupations_A_to_F#Entrepreneur, where some red links have variants but not all—should all dates still be enclosed in parentheses so as not to trip up OpenRefine? (It would be nicer for aesthetics but right now I figure we focus on essentials!) Innisfree987 (talk) 18:10, 18 April 2021 (UTC)
- I'm afraid they'd all have to be in parentheses for it to work properly. Though if there is a section with no variant names, you could just skip adding the parenthesis to that section. Just as long as each individual section has the same style throughout the section. Gamaliel (talk) 22:46, 18 April 2021 (UTC)
- Gotcha. In that case let me tag in Dsp13, to ask whether going forward, there’s any way to automate including either parentheses or some other character that satisfies this need? I am happy to work on what’s already up, but the entirety of what remains will probably outmatch me! Innisfree987 (talk) 23:33, 18 April 2021 (UTC)
- No problem to add parentheses. I'll try to look at it this evening. Dsp13 (talk) 08:29, 19 April 2021 (UTC)
- Gamaliel Actually, it was easier than I expected. I think I've added brackets - does that meet your need? Am happy to change anything as needed Dsp13 (talk) 09:06, 19 April 2021 (UTC)
- I'm working with Duchesses right now in OpenRefine and the new format is so much easier to work with, thank you everyone. Also I've noticed an uptick in automatic matching with the addition of the name variants. Gamaliel (talk) 14:00, 19 April 2021 (UTC)
- Gamaliel oh this is great! Thank you and Dsp13 so much for this solve. Things are really humming along! Innisfree987 (talk) 17:14, 19 April 2021 (UTC)
- I'm working with Duchesses right now in OpenRefine and the new format is so much easier to work with, thank you everyone. Also I've noticed an uptick in automatic matching with the addition of the name variants. Gamaliel (talk) 14:00, 19 April 2021 (UTC)
- I'm afraid they'd all have to be in parentheses for it to work properly. Though if there is a section with no variant names, you could just skip adding the parenthesis to that section. Just as long as each individual section has the same style throughout the section. Gamaliel (talk) 22:46, 18 April 2021 (UTC)
- Ah a question: in a section like this wikipedia:WikiProject_Women_in_Red/Missing_articles_by_dictionary/Dictionary_of_Women_Worldwide/Occupations_A_to_F#Entrepreneur, where some red links have variants but not all—should all dates still be enclosed in parentheses so as not to trip up OpenRefine? (It would be nicer for aesthetics but right now I figure we focus on essentials!) Innisfree987 (talk) 18:10, 18 April 2021 (UTC)
- Yup makes sense! Ok I will start working on those. Innisfree987 (talk) 18:01, 18 April 2021 (UTC)
- Thank you! Parenthesis will be fine, but any unique character will do. Sometimes Openrefine can automatically parse the data into separate columns, and sometimes it can't. When it can't I have to either do it manually (which is horrifically difficult in OpenRefine) or tell OpenRefine to split the data into columns at a particular character. Gamaliel (talk) 17:54, 18 April 2021 (UTC)
Completed categories
[edit]Any objection to removing the occupation categories with no more links at all, i.e. the ones currently marked (Category completed)? Originally I was enjoying the victories (wahoo!) but with the lists being so long now, I think the greater interest may be in minimizing clutter. Anyone opposed? We would definitely keep all the categories that still have links (red or blue). Innisfree987 (talk) 16:17, 24 April 2021 (UTC)
- Innisfree987: I think everything should be removed which isn't pertinent to a listing of redlinks. If people want to see the original headings, they can go either to the dictionary itself or look at the editing history of the various sections. On the other hand, I think it may be useful to take account of some of the headings (including those deleted) to extend coverage of some of our Wikidata redlists or possibly create new ones. I must say I'm really impressed with the progress you are making on this, together with the inputs from Dsp13 and Gamaliel. Please let me know if there's anything more I can do to help you along.--Ipigott (talk) 14:25, 28 April 2021 (UTC)
- Excellent, I agree and will start by removing all the “Category completed” sections. For the short-term we need to keep the “done” markings so that Gamaliel can see which are already entered in WD and which are still to-do but once ultimately complete (completely complete!), we can take those off too—I’m sure they might confuse someone looking for red links to use. Meanwhile thank you very much for the kind words which mean so much from considering the source—and for all your work, Ipigott, which has been another key element of steadily moving forward! It’s been a long haul but for processing 25,000 entries, not too shabby! Innisfree987 (talk) 16:21, 28 April 2021 (UTC)
- It strikes me that once finished, we will also be able to remove the page numbers and solely organize by category, which will help too, to streamline things for ultimate users of the list. Innisfree987 (talk) 17:19, 28 April 2021 (UTC)
- In fact I’ll go ahead and start that for A-F; it’s just in the way and I think it was only ever helpful to me anyhow (for knowing what pages to add, but that’s done here). Should def make it more readable. Innisfree987 (talk) 00:29, 1 May 2021 (UTC)
- It strikes me that once finished, we will also be able to remove the page numbers and solely organize by category, which will help too, to streamline things for ultimate users of the list. Innisfree987 (talk) 17:19, 28 April 2021 (UTC)
- Excellent, I agree and will start by removing all the “Category completed” sections. For the short-term we need to keep the “done” markings so that Gamaliel can see which are already entered in WD and which are still to-do but once ultimately complete (completely complete!), we can take those off too—I’m sure they might confuse someone looking for red links to use. Meanwhile thank you very much for the kind words which mean so much from considering the source—and for all your work, Ipigott, which has been another key element of steadily moving forward! It’s been a long haul but for processing 25,000 entries, not too shabby! Innisfree987 (talk) 16:21, 28 April 2021 (UTC)
Small victories!
[edit]I’m excited that our redlinks yielded some dozen new articles in the April “plants and gardens” editathon! (Visible under the red link section now turned blue!) Innisfree987 (talk) 09:02, 2 May 2021 (UTC)
And a milestone!
[edit]Wahoo!! Innisfree987 (talk) 18:41, 5 May 2021 (UTC)
Wikidata questions that stumped me
[edit]I thought I’d roll up my sleeves and do a few of the shorter WD lists—ha ha! Easier said than done. I was stymied by the “motorcyclist” WD entries for which pairs of women have just one WD entry—do we want to make them separate at least in WD? I don’t know the policy! And then “literacy activist” and “mental-health reformer” were not available occupations; so I wasn’t sure how to proceed there (tho for literacy the same person is list under at least four more occupations so I got the entry started...) Or, what of just plain “reformer” which is evidently a WD dab? What then? ... really, all hail Gamaliel who has been sorting out these and many more questions for months now! Innisfree987 (talk) 00:38, 30 May 2021 (UTC)
- PS. I am curious about the above but this is not in any way an urgent request (especially with the DC contest coming down to the wire!), more to note what’s been...halfway done, I’m afraid! Innisfree987 (talk) 00:38, 30 May 2021 (UTC)
- On the third hand, I’m finally seeing that OpenRef must allow for batch adding of FAR more info than I was, so I’ll cease mucking around above my pay grade!! A few entries started at least, but none remotely completed in the same way. Innisfree987 (talk) 17:42, 30 May 2021 (UTC)
Hip, hip..
[edit]...hooray! Incredible Gamaliel, immense thanks for immense work! Wahoo!! Innisfree987 (talk) 16:24, 17 June 2021 (UTC)
En-wiki entry for DoWW
[edit]Partly in response to concerns raised at Wikidata, I have (finally) begun an entry for DoWW: Dictionary of Women Worldwide. It’s a very tentative start so please don’t hesitate to pop in and improve on my initial sketch if you feel so moved! Innisfree987 (talk) 01:33, 4 July 2021 (UTC)
- @Innisfree987:, I'm all done with it. By the way, it's now over 1600 characters, so it meets the requirements for DYK. SilverserenC 05:24, 4 July 2021 (UTC)
- Silver seren, thanks so much for your work and great idea about DYK. I feel like the 5% stat would be interesting. Innisfree987 (talk) 20:04, 4 July 2021 (UTC)
Wikidata done?
[edit]With the exception of the list of oddities we've compiled, I think all the women who lack en.wp articles are now in Wikidata. Okay, now what.... Gamaliel (talk) 14:34, 12 July 2021 (UTC)
- !!!!! I cannot believe it. Back in the spring Ser Amantio suggested this dictionary of 25,000 women would be interesting to look at—and now they’re fully accounted for in WiR and Wikidata. Absolutely incredible. To start I think I will take a beat to sift through the list in awe!! But am all ears about next steps (beyond my minor organizing, removing now unneeded index page numbers)—I know there were some good ideas. Innisfree987 (talk) 16:45, 12 July 2021 (UTC)
What comes next??
[edit]Alright folks, it’s really happening: we’re wrapping up this project! All of the red links are in Wikidata as well as in a single CS list (by occupation) on the main page for the red list. Three cheers and more to everyone who has been chipping away for months, in ways big and small, to make this happen: @Gamaliel, Dsp13, Ipigott, Antiqueight, Silver seren, DanCherek, and Ser Amantio di Nicolao:...it’s been going on long enough I fear I’m leaving out key early contributors who helped get this off the ground? Immense thanks to all for your hard work and the pleasure of your collaboration.
Now the big question: what folks would like to do next. Ipigott had the good idea to draw out some sublists and I would make the suggestion that we use Wikidata for those, if anyone capable is willing or could be enlisted—here on the CS list, a few sections (countess, duchess, etc.) remain where name duplicates still need either weeding or clarification to indicate a blue link refers to a different person, but as Gamaliel has done the hard work of sorting through those to add them Wikidata, any that really do have an entry will be automatically removed from a WD list. We could start with lists for upcoming editathons, like a big DoWW list inclusive of all the different categories of writers for the September event; STEM for October; and Film+Stage for November.
That’s my main thought anyway. I’m eager to hear how else you all think this list could be made maximally productive. And please tag in anyone you think would have useful input! And again my gratitude and kudos: 25,000 women processed!! Innisfree987 (talk) 02:29, 25 July 2021 (UTC)
- It's incredible that it's all complete. It seemed like an insurmountable task when we started, honestly. I think i'll have more to say after this upcoming week. I'm deep in work on getting everything ready for my thesis defense on Wednesday and then i'll be officially done with graduate school and will be able to help out with WiR stuff a lot more. But, yeah, i'm up for any of the suggestions you listed. SilverserenC 05:36, 25 July 2021 (UTC)
- Congratulations, an amazing achievement! Well done for coming up with it in the first place. ☕ Antiqueight chatter 10:48, 25 July 2021 (UTC)
- Thank you both and good luck Silver seren! Innisfree987 (talk) 17:01, 25 July 2021 (UTC)
- Congratulations, an amazing achievement! Well done for coming up with it in the first place. ☕ Antiqueight chatter 10:48, 25 July 2021 (UTC)
Accessibiliy
[edit]You've done a really fantastic job on this over the past few months, Innisfree987. Nevertheless, without trying to be critical, I think we could do more to make your work more accessible. A Google search on "Dictionary of Women Worldwide: 25,000 Women Through The Ages Women in Red" revealed Dictionary of Women Worldwide but it was only when I included "Redlist" that I found Wikipedia:WikiProject Women in Red/Missing articles by dictionary/Dictionary of Women Worldwide/Occupations Q to Z. I had difficulty in finding it on Wikipedia:WikiProject Women in Red/Redlist index as it is not listed under Dictionaries but for some reason under Encyclopædias. There you can find Wikipedia:WikiProject Women in Red/Missing articles by dictionary/Dictionary of Women Worldwide and Wikipedia:WikiProject Women in Red/Missing articles by dictionary/DoWW which has not been updated for a month and appears to have lots of unnecessary blue links. I was also surprised to see that our Redlink index is no longer included in the Women in Red template. It does however contain a "List of resources" which points to Wikipedia:WikiProject Women in Red/Resources which in turn contains a section on "Biographical dictionaries" with a link to Wikipedia:WikiProject Women in Red/Resources/Fully accessible biographical dictionaries. This includes the following item: "Dictionary of Women Worldwide: 25,000 Women Throughout the Ages, ed. Anne Commire, Deborah Klezmer, Gale and Yorkin, 2006. Fully accessible via the Wikipedia Library bundle." But there are no clear guidelines on how to access the Wikipedia Library. (I must say I always have considerable difficulty myself in accessing the Wikipedia library as when I do so I am always linked to pages some years back when I tried to access specific resources.)
I must say I find this all rather confusing and difficult to follow. After the enormous amount of effort spent on this enterprise, I would have thought it would be much easier for inexperienced editors to find pertinent redlists and more specific instructions on how to find articles in the dictionary itself. (One step might be to develop an easily accessible list of the professions and occupations included in the dictionary with appropriate links.) Does anyone have any other suggestions on how access could be improved?--Ipigott (talk) 09:09, 25 July 2021 (UTC)
- Hi Ipigott, thanks so much, these were just the kinds of items I was hoping to surface now that we have a relatively complete list to prepare for dissemination. Let me try to take them one at a time.
- Redlink index. I assume it was put under Encyclopedias because biographical dictionaries is mostly organized by country. If it would be better dropped down to “by field/topic”, let’s do it.
- WiR template and inclusion of redlink index. Perhaps an item for main page talk? I know others think the template could be improved as well.
- Google-ability. I don’t know much about SEO to make the page more discoverable that way, but it does strike me we should probably NOINDEX the subpages, or even redirect them to the main list if folks are happy having one main redlink page (if it’s not too long? Do people like it this way?)
- WD list not updating: at the moment it seems SPARQL quits when trying to refresh it. I hope that gets fixed; alternately some shorter lists that get used in editathons might help trim the quantity of data it’s handling!
- Blue links in the WD list. Some could be from lack of updating. Others are what I alluded to above—occurring when a subject, for instance British poet Anne King, does not have a Wikipedia entry but others who share her name do. Once the poet’s page is created and linked to the WD entry, it will be automatically removed from the list. We should definitely include that explanation with the list. Are there many you noticed that weren’t either of this type, or newly created?
- Wikipedia Library. Getting WiR folks signed up and comfortable using it is probably a bigger project to (continue to) pursue maybe on the main talk page.
What I will say is once they are, just clicking the link included here at the redlink list takes the user directly to the search tool inside DoWW. Is that part clear? If yes we can just add it to the page where you note it’s missing on the Wikipedia:WikiProject_Women_in_Red/Resources/Fully_accessible_biographical_dictionaries page.ETA: I misunderstood this, thinking it linked back to the redlist. It already links to the Library page, so just needs info on signing up. I’ll add the Bundle link we have, hoping that’s of some assistance (but I continue to think it would be beneficial to have a more general effort to introduce WiR folks to the Library.) Innisfree987 (talk) 22:01, 25 July 2021 (UTC) - I don’t quite follow what you mean when you suggest
”an easily accessible list of the professions and occupations included in the dictionary with appropriate links”
. Something different from the table of contents here?
- Thank you again for all this feedback; now that the data is processed, I’m eager to make it as widely available as we can. Innisfree987 (talk) 16:58, 25 July 2021 (UTC)
- I should probably note that the Wikidata list is so large it is timing out when Listeria tries to update it. Perhaps Tagishsimon has some advice on that matter, or perhaps we could break it up into smaller Wikidata lists by occupation or time period. Thoughts are welcome. Gamaliel (talk) 19:46, 25 July 2021 (UTC)
- Gamaliel, sorry for the delayed response—I wanted to give the more technical folks a chance to reply and then time got away from me! Era seems like a good way to go—I know some entries didn’t get their dates into WD because of a “circa” listing that WD doesn’t seem to allow but it’s still probably the easiest way to cover the most people?
Alternatively as I mention above, there are upcoming editathons that could use large chunks of the list (e.g. I made a list of all the occupations related to next month’s Women writers, and there are many!); I’d be grateful for assistance or pointers on creating that WD list. Ditto the subsequent months on Stage/Film and then STEM. Maybe those editathons would even knock out enough redlinks to make the original list start loading.
What makes most sense to you? Innisfree987 (talk) 19:47, 5 August 2021 (UTC)- Hey turns out reading the directions helps just a bit! Baby’s first SPARQL query, redlinked writers described by DoWW.
Still need to sort out how to write “or” syntax (to include all who are writers or calligraphers or poets, rather than only the few if any who are writers and calligraphers and poets). But evenETA: thanks to Tagishsimon’s guidance, this now has multiple types of writers and related professions and would do for the Women writers editathon—I think? Notes very welcome! Innisfree987 (talk) 03:37, 6 August 2021 (UTC)- I have also started a Film+Stage list for November’s editathon in the same vein. Next STEM for October. Innisfree987 (talk) 22:15, 7 August 2021 (UTC)
- Hey turns out reading the directions helps just a bit! Baby’s first SPARQL query, redlinked writers described by DoWW.
- Gamaliel, sorry for the delayed response—I wanted to give the more technical folks a chance to reply and then time got away from me! Era seems like a good way to go—I know some entries didn’t get their dates into WD because of a “circa” listing that WD doesn’t seem to allow but it’s still probably the easiest way to cover the most people?
- I should probably note that the Wikidata list is so large it is timing out when Listeria tries to update it. Perhaps Tagishsimon has some advice on that matter, or perhaps we could break it up into smaller Wikidata lists by occupation or time period. Thoughts are welcome. Gamaliel (talk) 19:46, 25 July 2021 (UTC)
Citation template?
[edit]This is a restricted issue, but I would find an easy and standard way to cite DWW useful in writing using articles using it - maybe a citation template akin to Template:cite ODNB? Dsp13 (talk) 11:11, 26 July 2021 (UTC)
- Oh I think that would be great, Dsp13—just the other day an editor suggested we go back to all the entries where it’s cited and link to the new mainspace article Dictionary of Women Worldwide, but that’s quite cumbersome—way better to set up something that would automatically link to it for the future, if it’s not too much trouble. Innisfree987 (talk) 15:47, 26 July 2021 (UTC)
One main list?
[edit]What do folks think of the centralized CS list? Too long or does it work? If folks like it ok, I’ll redirect the three subpages because a few folks are kindly making edits on the subpages and I want to make sure the main list doesn’t wind up out of date. On the plus side, people seem to be finding the list! Innisfree987 (talk) 00:38, 27 July 2021 (UTC)
- So it’s not too drastic, I’ll start by just commenting out the subpage links from the main page. Innisfree987 (talk) 17:57, 27 July 2021 (UTC)
Main DOWW Wikidata list updated!
[edit]Hooray! So now all of the entered red links are there—less all the new articles! Innisfree987 (talk) 05:50, 30 August 2021 (UTC)
- and there's been some fantastic content creation by @Antiqueight: turning redlinks blue! Dsp13 (talk) 09:14, 1 September 2021 (UTC)