Jump to content

Wikipedia:Good article reassessment/Death of Elisa Lam/1

From Wikipedia, the free encyclopedia
Article (edit | visual edit | history) · Article talk (edit | history) · WatchWatch article reassessment pageMost recent review
Result: Other editors have been quick in responding to concerns raised, and there is no consensus that the article includes original research in its current form. The article otherwise fits the six good article criteria and will be kept.StoryKai (talk) 16:36, 8 July 2019 (UTC)[reply]

GAs cannot contain OR. Both the reviewed version and the present version include the text One page of the report has a form with boxes to check as to whether the death was accidental, natural, homicide, suicide or undetermined, in large type and a sufficient distance from each other. The "accident" box is dated June 15; however three days later the "undetermined" box was checked instead. This was at some point in the three days before the report's release noted as an error and crossed out and initialed., attributed directly to a scan of the autopsy report itself. The review did not address this, and seems to have completely missed that there was such textbook OR in the article. Hijiri 88 (やや) 14:28, 27 March 2019 (UTC)[reply]

@Hijiri88:Tell me when in the article the OR is and I will remove it.MagicatthemovieS (talk) 15:28, 27 March 2019 (UTC)MagicatthemovieS[reply]
@Hijiri88:OK I fixed the problem you brought up.MagicatthemovieS (talk) 15:28, 27 March 2019 (UTC)MagicatthemovieS[reply]
It doesn't matter if you "fixed the problem I brought up"; the article had a really blatant problem that was apparently missed in the review, so we should be assuming that there were a lot of more subtle problems that were missed in the review. A thorough source check is needed. Hijiri 88 (やや) 23:21, 27 March 2019 (UTC)[reply]

@Hijiri88: I'm not reverting the edit, but can you do more to explain your complaint than say "textbook OR"? All the now-deleted sentences said was merely descriptive of what the primary source said. I'll grant that the wording needed a little work, but I don't find the verifiable description of what the autopsy report showed to be problematic. I think the cited source may even have discussed that detail. Daniel Case (talk) 22:43, 27 March 2019 (UTC)[reply]

It is clearly inappropriate to speculate on the reason why "accident" or "undetermined" was filled in unless reliable sources have done so previously, let alone to talk about how far apart they are and how clear the lettering is. Hijiri 88 (やや) 23:19, 27 March 2019 (UTC)[reply]
@Hijiri88: The source cited does indeed say exactly that: "On 15 June the cause of death was ticked as "accidental", but on 18 June the cause of death was ticked as “could not be determined” but was crossed out, and the word "error" was written next to it." That's why I included a link to the autopsy report, so readers could verify that detail with their own eyes.

Second, if we took out "in large type and a sufficient distance from each other" and left the rest in, would you still call it "OR"?

Third, just how does the text speculate on why? The mere juxtaposition of these details might lead readers to speculate, but they do not manipulate them into doing so, properly worded. Daniel Case (talk) 02:55, 28 March 2019 (UTC)[reply]

Inclusion of details gleaned from a primary source, even if those details are completely accurate (and that's a big "if"), implies some reason for doing so. The text as it was at review clearly implied that there was some confusion over the cause of death, rather than the much simpler reading that it was simple clumsiness. I have not checked to see if you or someone else has inserted a new citation of a secondary source that actually supports the content, and I don't see why I should have to. The initial review was inadequate, and I decided to leave it to the community to discuss how to address that. This is not an individual reassessment (I have been very careful about that ever since an incident in 2016). Please do not ping me again. Hijiri 88 (やや) 05:01, 28 March 2019 (UTC)[reply]
  • Keep: After reading the article and all of the sources used to reference any potentially OR text, I'm satisfied that this article doesn't contain original research. To be clear, I don't believe simple WP:TRANSCRIPTION constitutes OR. Maybe the last sentence in the text quoted by Hiriji88 could be perceived as OR, but the entire quoted text has been removed in any case. I couldn't personally see any other issues/any reason to demote while reading the article. Which all leaves me with the impression that this could've easily been dealt with at the article's talk page. Homeostasis07 (talk) 23:44, 30 March 2019 (UTC)[reply]
@Homeostasis07: The policy you cite does not address the issue of selecting certain material from primary sources and creatively interpreting it; in fact it doesn't apply to the text I quoted at all, since it is not transcription of quoted material (or faithful translation of foreign-language material) but rather original prose interpretation of a primary source -- yes, maybe a lot of it is accurate description of some of what is in that primary source, but that's different from faithful transcription of quoted content. 99% of experienced Wikipedians would demand a reliable secondary source for such content. Hijiri 88 (やや) 00:29, 31 March 2019 (UTC)[reply]
But there isn't much difference between the now-removed prose you quoted above and the text Daniel Case quoted from the source. So the intent of transcription could be applied to this. And I don't see any further issues arising in the article. Homeostasis07 (talk) 15:58, 31 March 2019 (UTC)[reply]
  • Comment I have some sympathy for Hijiri88s view. These death of ..., murder of ..., killing of ... etc articles are a haven for fan theories and speculation. However, I don't think this is that bad compared to some I have run across. Saying that I have actually raised similar concerns at this very article over the use of primary sources. This was over an overly descriptive account of the video using the video itself as a source (see Talk:Death of Elisa Lam#Video section - My original post was split up by Daniels responses so it looks a little messy). This was a couple of years before the article was passed as good. The same issue is still there, a description of the video based on the video itself. It needs secondary sources if for nothing else than to allow us to assign the correct weight in what we describe. I don't know if I would classify it as textbook OR, but it is definitely not best practise.
"These death of ..., murder of ..., killing of ... etc articles are a haven for fan theories and speculation. However, I don't think this is that bad compared to some I have run across." Why thank you ... what articles like this need is someone keeping a regular eye on them, and I've been doing that for the last five years.

. As for the description of that part of the autopsy report, I've been thinking about that that maybe we could just take a screenshot of that part of the page and put it in there; that would speak for itself and end any need to describe it. Such an image wouldn't create any copyright issues, either, as just words and incomplete phrases. Daniel Case (talk) 21:15, 1 April 2019 (UTC) (signed belatedly)[reply]

The autopsy report is used 16 times as a reference. This is far too much use of this type of primary source in an article like this. Again if it is descriptive and not interpreting the results it is not strictly speaking OR, but it does again provide a weight issue, which is essentially a NPOV concern.
I am suspicious of long in popular culture sections. I looked at one at random - the "How to Get Away with Murder" paragraph - and from the review the only reference to Lam I could find was Is Lila Stangard inspired by Elisa Lam?. I don't know how a stray seven word thought from a reviewer deserves a paragraph and it overplays the source in suggesting that it is actually inspired by Lam.
Was this one recent? I have taken so much out of that section over the years that I can't remember all the additions. Daniel Case (talk) 03:43, 1 April 2019 (UTC)[reply]
Overall there is enough to suggest this article relies too much on primary sources for information that would better sourced from secondary ones. Primary sources are allowed to be used so they don't all need to be replaced. I am not a fan of Hijiri88's hit and run approach to these reassessments, but there is merit to looking closer at this one. AIRcorn (talk) 23:39, 31 March 2019 (UTC)[reply]
@Aircorn: I am not a fan of Hijiri88's hit and run approach to these reassessments Umm ... what? That's a ... pretty bizarre accusation, given my history at GAR. Can you back it up with something? I've almost always been told that I should be less involved with these kinda things, and leave them for the community to decide, the one exception being a disastrous occasion on which I accidentally opened an individual assessment when I meant to open a community one. Hijiri 88 (やや) 23:32, 5 April 2019 (UTC)[reply]
You tend to drop an article here after identifying one or two problems and then expect other editors to identify the rest. Now in theory that is perfectly reasonable and maybe even one of the purposes of this part of the project. I even do something similar myself to see if any editors are interested in addressing the concerns before I go too far in depth. However in practice you will be lucky to get many other opinions here and in the end it just creates work for someone else or ends up with the other issues not being addressed. So, yeah I am not a fan. I am not a fan of many things so don't take it personally. I don't know who told you to be less involved with reviews, I would say that unless things get heated or unproductive you should stay involved as much as you can. I feel that with your experience you are more than capable of conducting individual reassessments. BTW I see nothing disastrous about the above review. In fact you left some good points, others agreed with you and it was delisted. The beauty of this reassessment is that there are editors interested in it keeping its Good status. That is relatively rare and worth working with. AIRcorn (talk) 06:19, 10 April 2019 (UTC)[reply]
Good point (about the description of the video). The prose there struck me as a bit repetitive, but nothing insurmountable. Perhaps it could be tightened up a bit, @MagicatthemovieS: @Daniel Case:? And I'm sure it wouldn't take much effort to find a secondary source describing what she's doing in the video—considering the coverage this... "incident"(?)... attracted? As far as your point about the autopsy report being used 16 times, several of those occasions see that source being used as a secondary reference to information also sourced by other references—so that isn't necessarily a deal breaker. Homeostasis07 (talk) 01:19, 1 April 2019 (UTC)[reply]
@Homeostasis07: Part of the reason for the long description is that it has always been my goal to get the video itself in the article ... there are sure enough copies out there, if someone knows where we can get one and upload it I'm all ears (so to, uh, speak). It's entirely justifiable under fair use (although we may have to go with a 30-second clip; perhaps we can edit it down to the most interesting parts). If someone else can upload a good copy, let me know beforehand so I can write the fair-use justification. Daniel Case (talk) 03:43, 1 April 2019 (UTC)[reply]
@Daniel Case: The only videos I've ever seen uploaded to Wikipedia are 100+ year old silent films whose copyright expired long ago. I'm sure hotel security footage is public domain, but I wouldn't even know where to start with any of that. Maybe someone at the Wikipedia:Teahouse might have some advice on how to proceed? Homeostasis07 (talk) 20:42, 1 April 2019 (UTC)[reply]
@Homeostasis07: Take a look at American Beauty, which has one clip from the film nestled next to a section discussing critical interpretation of the film relevant to it, per the fair-use criteria. We also have user-created video, like the one I produced (my son was the cameraman; I edited the separate takes) showing how a French press is used to make coffee. I particularly like the one I found on Commons that has been the lede media on rain ... if you were one of the possible handful on the planet who had grown to adulthood without ever experiencing liquid natural precipitation, that video would explain it.

Under current US law copyright attaches to the video even though it was produced automatically by the hotel's security cameras. No, I don't think the Cecil would sue, but the fair use policy doesn't take the likelihood of a rightsholder suing into account. Daniel Case (talk) 21:14, 1 April 2019 (UTC)[reply]

@Daniel Case: What exactly do you mean by "upload a good copy" of the video? Do you mean quality-wise? Because the video linked to in the article as an 'External video' is the best quality one I've seen. I seriously doubt anyone is going to remaster this, and I doubt we'll ever get a full, unedited version of it. Is the problem that you can't download it off YouTube? Homeostasis07 (talk) 21:28, 1 April 2019 (UTC)[reply]
@Homeostasis07: If you find a downloadable version, I'll take it. Seriously, I thought you couldn't just download anything you wanted ... the uploader has to give permission? Daniel Case (talk) 22:17, 1 April 2019 (UTC)[reply]
@Daniel Case: Technically, you can't download anything off YouTube, either with or without uploader permission. Technically. ;) Uploading to Wikipedia without permission of the copyright holder – if one exists, which in this case is extremely doubtful – is a separate thing entirely. I'll leave that up to you to decide. Homeostasis07 (talk) 22:50, 1 April 2019 (UTC)[reply]

Aircorns comments

[edit]

I guess I should make a more formal review section.

  • I looked at the autopsy reference and am a little concerned. It appears to be scanned and uploaded to PDF archive. There are no indication of who the uploader is, copyright status or even any way to be sure it is legitimate. This is the information page and is not very helpful. I have a similar concern about this one too. I can't access the original site [1] so have no idea on its reliability. It again looks like a scan of some court document without any indication of legitimacy.
  • The use of primary sources has been raised as a concern so I looked closer at them
    • Lams blog is used to source the date it started. I would be a bit more comfortable if it actually used Lams name instead of Ether Fields, but another source seems to confirm it is her blog.
    • Her Tumblr is also used as a source to say she was starting a tumblr blog and then describe what was on it. I am not sure how that detail is important. That is why we need secondary sources to mention this stuff as otherwise we are assigning our own weight to information based on our interpretation of primary sources.
    • The video is at least linked to the LA police so is a reliable primary source,even though it is hosted on youtue. The cites below the main one are not up to GA standard though. It should actually link to the video, Detectives means nothing as far as cites go. I brought this up years ago.
      • It drew worldwide interest in the case is not supported
      • The whole description is too convoluted. Take the She walks to it again and stands in the doorway, leaning on the side. Suddenly she steps out into the hall, then to her side, back in, looking to the side, then back out. She then steps sideways again, and for a few seconds she is mostly invisible behind the wall she has her back to just outside. The door remains open. paragraph. You would be much better off using a secondary soure that says she acted strangely than trying to convey that by describing stage prompts. Descriptive enough not to be original research, but very dry. There must be secondary sources describing the video, otherwise why would we include a full section on it
As I've said before, my goal has been to have the video in the article, so there wouldn't be a need to describe it.

A lot of secondary sources simply embedded the video, making it unnecessary for them to describe it, so there's a paucity of that. Daniel Case (talk) 04:26, 5 April 2019 (UTC)[reply]

    • This [2] is not a reliable source
I didn't add that but it is used as a source for the tanks being open and the roof being accessible after her death. Presumably we can trust the datestamp. If we just use specific time references in the video (near the end) to what it is said to support, which is visually self-evident, I don't see how it's a problem. Daniel Case (talk) 04:33, 5 April 2019 (UTC)[reply]
    • The autopsy report is used as a reference for
      • It took the Los Angeles County Coroner's office four months, after repeated delays, to release the autopsy report, This is in the lead and the four month can be supported by the dates in the body. There is no mention of repeated delays however in the body.
      • I don't see why the autopsy report needs to be used alongside reliable secondary sources (i.e the LA times saying a video has been released) In fact the autopsy report doesn't say anything about a video being released so I am not sure why it is needed here. Also the report makes specific mention of her not appearing distressed. The description and comments, which are from random viewers much less qualified to make calls in this regard, in the article tend to suggest otherwise. Some argued that she was attempting to hide from a pursuer, perhaps someone ultimately responsible for her death, while others said she was merely frustrated with the elevator's apparent malfunction. I would be tempted to tag the "somes" with a {{who}} as it is important to identify armchair detetives from real ones. At the least it needs to make clear at the start of the paragraph that these are armature theories. It reads a bit like some reddit thread on conspiracy theories, which is understandable since a part of the notability of this case is the releasing of the video, something the police only really have themselves to blame. I think we need to be clearer though on this and not give too much weight to the conspiracies over the actual police conclusions.
      • The tank was drained and cut open since its maintenance hatch was too small to accommodate equipment needed to remove Lam's body Not seeing this in the autopsy report cited
      • Toxicology tests – incomplete because not enough of her blood was preserved – showed traces consistent with prescription medication found among her belongings, plus nonprescription drugs such as Sinutab and ibuprofen This is cited to five pages of the report. It is doing my head in because they are scanned sideways and everytime I click on next a pop-up appears. I am really not liking how this source is presented (I am getting banner ads flashing at me). I will take your word on the drugs, but can you confirm where it says that it was "incomplete because not enough of her blood was preserved".
  • It also records subcutaneous pooling of blood in Lam's anal area,[34] which some observers suggested was a sign of sexual abuse; however one pathologist has noted it could also have resulted from bloating in the course of the body's decomposition,[3] and her rectum was also prolapsed. Some observers really needs more info as there is a big difference between a doctor, pathologist or coroner compared to a layperson.
  • Apart from the exageration in the popular culture source mentioned above the others at least make Lam a major part of ther story. I still feel some of the descriptions are a bit detailed, but probably not a GA thing.
    • There is quite a paragraph devoted to a filmaker saying they would make a film of the incident, but it doesn't say whether this happened or not. This was four years ago so might need an update (and removal of all the speculation). In fact a few of these need either updating if they have been done or removing or trimming if they never made it past the concept phase

Overall I personally don't like the use of primary sources in this article, and many of them should be pretty easily replaced with secondary sources. Some of the points above are important for a GA, while others are not. I really would like to know how you came across the autopsy report as it seems a strange site to be hosting it. AIRcorn (talk) 10:32, 3 April 2019 (UTC)[reply]

Keep I don’t view this as being original research at all. It's just a straightforward description of what the document says, and it clearly wasn't meant to imply anything. Maybe there is some misunderstanding here, but a careful inspection of the article and the sources definitely shows no original research. Granted, it's already been removed, so this observation doesn't really matter anymore. The only issues worth discussing are those that haven't already been addressed. Looking over the original good article review, I can see that it was very thorough, so if the misunderstanding over original research is the only issue here, then of course I have to vote to keep. My recommendation is that further misunderstandings be discussed on the talk page, rather than through good article reassessment.ErinRC (talk) 16:42, 3 April 2019 (UTC)[reply]

@Rp2006: MOS:LEAD says the lead section should summarize the article and be no more than four grafs long. I think this intro fits within those parameters.

Whether other people have persuaded you to shorten intros to other articles during the GA process really isn't relevant to that, IMO. Daniel Case (talk) 18:12, 25 April 2019 (UTC)[reply]