Wikipedia talk:Reliable sources/Noticeboard

Active discussions
WikiProject Reliability
This page is part of WikiProject Reliability, a collaborative effort to improve the reliability of Wikipedia articles. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.

Automatic archiving not fully workingEdit

In the past month, Lowercase sigmabot III has not archived a significant number of discussions on this noticeboard that have been inactive for at least 5 days (the duration set in the User:MiszaBot/config template at the top of the noticeboard page). These discussions do not contain the DoNotArchiveUntil tag. Since the bot operator did not respond to my inquiry at User talk:Σ § Automated archiving on the reliable sources noticeboard, I am now going to use OneClickArchiver to manually archive all discussions that:

  • Have not received any new comments within the last 5 days, and
  • Do not contain a DoNotArchiveUntil tag

Hopefully, Lowercase sigmabot III will work smoothly after this one-time operation. — Newslinger talk 17:31, 18 October 2019 (UTC)

This problem appears to be recurring. "Is wikispooks.com a reliable source or should it be deprecated?" is an example of a discussion that was not archived after almost 8 days of inactivity. If anyone else sees discussions on the noticeboard that are not archived despite meeting the above criteria, please help reduce the clutter by archiving it manually. Scripts like OneClickArchiver can help. — Newslinger talk 08:13, 15 November 2019 (UTC)

List of deprecated sources somewhere? Which bot identifies them when editing an article?Edit

Ashley Alexiss (model) (edit | talk | history | protect | delete | links | watch | logs | views) was giving me a message about a deprecated source in it somewhere after I went over the article. I can't get the message to appear again, so I'm guessing I removed the ref and the bot was misbehaving. How do I check? I assume there's a list somewhere that a bot is using. --Ronz (talk) 19:57, 8 November 2019 (UTC)

We have around 1300 cites to Reddit, excluding IAmAs.Edit

I just noticed, in a quick search, that we have roughly 1300 cites to Reddit, excluding R/IAmA. At a quick glance, the vast majority of these seem totally unusable - often just cites to random reddit posts saying something. Even the IAmAs, though they often pass WP:SPS / WP:ABOUTSELF, should probably get a look, but the ~1300 cites in that link are almost all unusable. Unsure if this is worth raising anywhere, but I thought I'd mention it here. --Aquillion (talk) 19:50, 10 November 2019 (UTC)

  • Can they be converted to the original source if its reliable ? Atlantic306 (talk) 20:31, 10 November 2019 (UTC)
    • In a few cases, but in the majority that I've glanced at, either there's no original source (ie. it's just some anonymous user saying something), or the original source is unreliable (eg. a screenshot of someone's twitter feed). Right now the ones at the very top of the search lean more towards borderline because I've been removing the obviously unusable ones as I come across them, but even they don't generally seem easy to convert. --Aquillion (talk) 07:16, 11 November 2019 (UTC)
    • Thanks for explaining, that is a pain to resolve Atlantic306 (talk) 19:01, 12 November 2019 (UTC)

What (exactly) does "Deprecation" mean?Edit

(Title was: "What (exactly) does "Depreciation" mean?) --Guy Macon (talk) 14:21, 12 November 2019 (UTC)

We have had a lot of discussions recently about whether some source should be “Depreciated”... and as I read them, one thing stands out: I am not at all sure that we are in total agreement about what “depreciation” actually MEANS. At one extreme, we have editors who seem to think it means a blanket ban on using the source, and that we should nuke it when found. At the other extreme, we have editors who seem to take it more as a “caution”... and that depreciated sources can be used in limited situations. I think we need to reach a consensus on what “depreciation“ means, before we have any further discussions on specific sources. Blueboar (talk) 22:25, 10 November 2019 (UTC)

Please read Wikipedia:Deprecated sources. Cullen328 Let's discuss it 22:29, 10 November 2019 (UTC)
That doesn't exactly represent a consensus though. GMGtalk 23:12, 10 November 2019 (UTC)
But that essay is probably the place to start if someone wants to formalize that consensus, right? I actually don't think there's much disagreement on what deprecated sources are: they're sources that are so consistently unreliable that they should generally not be cited in Wikipedia except under exceptional circumstances. There are disagreements on whether they should be nuked-on-sight, but I don't actually know if we need to have a consensus on that question in order to keep having deprecation discussions. Nblund talk 23:23, 10 November 2019 (UTC)
I guess that depends on whether the difference causes a mess or not. GMGtalk 23:25, 10 November 2019 (UTC)
Doesn't it? I think it gets linked in discussions that suggest depreciation nowadays, so I'd argue that any discussion that reaches a consensus to depreciate a source is finding a consensus to, well, do that. Giving it its own page just provides an easy way to summarize what such RFCs are suggesting without a wall-of-text. --Aquillion (talk) 07:19, 11 November 2019 (UTC)
Blueboar, a deprecated source is presumptively unreliable. Put simply: if a better source exists, please use it instead; if no better source exists the content should probably be excluded altogether. Guy (help!) 11:29, 12 November 2019 (UTC)

BTW, the term is deprecation, not depreciation. Headbomb {t · c · p · b} 23:30, 10 November 2019 (UTC)

The first two sentences of Wikipedia:Deprecated sources (WP:DEPS) provide a brief summary:

Deprecated sources are highly questionable sources that editors are discouraged from citing in articles, because they fail the reliable sources guideline in nearly all circumstances. These sources may be subject to restrictions, including an edit filter.

Historically, deprecation has been proposed through a formal RfC on this noticeboard to ensure a well-publicized discussion. Deprecated sources are a subset of questionable (and generally unreliable) sources. The main difference between a deprecated source and a non-deprecated questionable source is that technical measures (including auto-reverts by XLinkBot and edit-filters set to warn) are typically used to actively discourage the use of deprecated sources (see WP:DEPSOURCES for a full list). These measures are rarely used for sources that are considered questionable, but have not been deprecated. — Newslinger talk 01:18, 11 November 2019 (UTC)

As a reminder, deprecated sources are also subject to WP:ABOUTSELF and WP:CONTEXTMATTERS. — Newslinger talk 01:41, 11 November 2019 (UTC)
I feel that definition is missing at least the software meaning of the terms, and which I've seen it used here (eg like with Daily Mail), in that it was at one point an acceptable source, but due to consensus changes here and/or external situation changes, no longer is such a source. What's important with this is that this may mean that earlier versions of that source are fine, while later versions should be avoided (eg the question being asked about Newsweek right now on RS/N). Deprecation does not mean the source must be removed (it definitely should not be added unless essential) but understanding why deprecation happened would explain what is the right approach to deal with existing entries. --Masem (t) 15:15, 11 November 2019 (UTC)
Masem, Daily Mail considered harmful... Guy (help!) 15:26, 11 November 2019 (UTC)
From WP:DAILYMAIL decision : The Daily Mail may have been more reliable historically, and it could make sense to cite it as a primary source if it is the subject of discussion. These seem to be good points, but should come up very rarely. Editors are encouraged to discuss with each other and apply common sense in these cases. I 100% agree that if we're talking an article from DM dates post 2010, avoid it at all costs, but the decision would allow for older users when the DM wasn't a flat out tabloid. This is how deprecation works in software frequently: older, complex code is based on deprecated features that are no longer supported but still function, so its fair to use it there, but new code should avoid these deprecated features. --Masem (t) 15:29, 11 November 2019 (UTC)
That's a good point, and the phrase "reliable historically" is noted in both WP:RSP § Daily Mail and WP:DEPS § Daily Mail. However, from a glance at the list of currently deprecated sources, I think the Daily Mail's historical reliability is more of the exception than the rule. Setting aside the three print tabloids (Daily Mail, National Enquirer, and The Sun), the remainder of the deprecated sources were established recently enough to lack a substantial history that could be evaluated differently. — Newslinger talk 21:48, 11 November 2019 (UTC)
My point is that "deprecated" never has really meant "remove with all vigor", even in the case of the Daily Mail. If DM was to be forbidden in all cases, there is other language that could be used. Deprecated gives the intent that we should be going through all links that we use from that deprecated source, determine if they are allowed to stay by the reasoning of the deprecated decision and if not, either carefully take the time to replace with more appropriate sourcing or otherwise remove the information that was strictly dependent on that source. Deprecation from a software side has never meant to be a full immediate wipe of all versions of the deprecated without thinking since this can have other unintended consequences. If it is truly the case that enough users like JzG feel that DM is harmful to keep around, they can organize an effort to review each DM entry for replacement as to rid ourselves of bad DM uses in the most expedient but human-reviewed way possible. --Masem (t) 22:34, 11 November 2019 (UTC)
Masem, that's exactly what I do. I tag as deprecated unless it's redundant to a better source already in place, in which case I remove; some time later (usually months) I review the tagged sources and either replace with a better source or, if it looks like trivia, just remove it. This is similar to the approach I have taken for years with predatory journals, which has reduced the levels of citation to such journals from tens of thousands to a few hundred at most, and with Breitbart.
The one big difference between the Mail and, say, a SciRP journal, is that the latter cites are very often added by IPS that coincidentally are associated with one of the authors' institutions. Weird how that goes. So the Mail is not spam and I treat it much less aggressively as a result.
There are, at least in theory, valid uses for the Mail, I just don't find many of them. It's more likely to be stories about footballers illustrated with pictures of their girlfriends in bikinis, which fits delightfully with the "sidebar of shame". And when the sidebar goes I will be happy to reconsider whether the new editor is putting the paper on a more secure journalistic footing. Guy (help!) 11:38, 12 November 2019 (UTC)
Then you're fine, I just get the tone from how heavy you come on DM being harmful (which is fair enough) that it might imply more hasty methods that may be harmful themselves. Your method described above is safe and appropriate and how I would expect we handle something "deprecated". --Masem (t) 15:35, 12 November 2019 (UTC)
Masem, did you read the linked article? It's old school geek-speak. As in "GOTO considered harmful". Always inviting the riposte: "sez who?" Guy (help!) 15:54, 12 November 2019 (UTC)
I admit I presently did not recall hearing that term and mistook the statement for something else. Fully agree with your point with that and do apologize for suggesting you might not be taking the proper care with removing DM links. I can see nothing wrong with how you are handling it, in assuming good faith that you are checking, and that we agree on the ultimate goal of minimizing the use of DM articles save where DM is the focal point of the situation. --Masem (t) 01:04, 13 November 2019 (UTC)
Masem, no offence taken, my friend. As one who grew up on CompuServe and Usenet, I am used to people not getting my references :-) Guy (help!) 13:44, 13 November 2019 (UTC)
  • The crux of this is whether deprecation means to avoid creating more of a problem, or else to immediately remove all the instances of it. The off-wiki sources are clear on this. It's a term widely used in technology industries for just this situation. WP:DEPS agrees. There are cases when we might wish that a source wasn't ever cited, but it has been, and immediately just removing all uses of it (and maybe removing significant blocks of content too) would make things worse. It's a common problem in technology, it's particularly a problem with formal specifications.
But on this page, a couple of editors are advocating immediate bulk removal, on the basis that a source is "deprecated". That's not what the word means. It's not what policy supports. It's not what specific RfCs have decided. Andy Dingley (talk) 22:06, 11 November 2019 (UTC)
So, let's speak in specifics, not meta-level abstractions. Please explain e.g. why the Daily Mail is a good enough source to keep in any circumstances; and your own recent defence of News of the World - David Gerard (talk) 23:09, 11 November 2019 (UTC)
I've never defended the NotW. You know that. So for you to make such a statement indicates that you have as flexible a relation with truth as the Mail itself does. I'm very disappointed in you. Andy Dingley (talk) 23:56, 11 November 2019 (UTC)
And yet, your actual editing was to do precisely that thing - you didn't even want it even tagged as a bad source. Your reasoning - such as anyone could extract it from you - turned out to be personal animosity and extended personal attacks on another editor, and never mind a single thing about the quality of the source itself - despite many people asking you repeatedly, in a massive one-against-many.
The trouble is that when you do a thing, you do the thing, even if the reason is that you hate another editor so much you feel compelled to oppose even their good ideas. You can't come here and then deny that you did the thing you did. The edit history exists.
You still haven't given a reason why deprecated sources shouldn't be removed, and indeed why claims based on the deprecated source shouldn't be removed. They're deprecated because they're sewage-quality sources.
If you insist that - despite your repeated attempts in discussion to hold back work on removing these sources - you don't support continued use of these sources, please list your work to reduce their use in practice - David Gerard (talk) 08:11, 12 November 2019 (UTC)
  • David, you are so insistent on misrepresenting anything that I've written here that I cannot explain that without being forced to assume the worst bad faith on your part. I have never defended the NotW or the Sun. I'm not against their removal. But I am against their removal in a way which is either badly done, or which leaves the overall content in a worse state. In particular, I have seen JzG do this over and over on a number of targets, and to do so in a way which is a gross failure of WP:CIR. Read above for the second-to-latest one where he takes WoodElf to ANI with all sorts of accusations for edits that simply didn't happen. Anyone who takes on the task of making bulk, gross changes to our content in any way has it incumbent upon them to at least check that their results match what their intentions had been, and Guy shows no sign of ever doing that: therefore he's not a fit person to be doing it.
We need to discuss sensibly what "deprecate" means in the WP context, agree what we're actually deprecating, then what sort of actions to take after that. I note that you've been removing NotW sources yourself, but you seem to be doing them at a slower rate which indicates some sort of editorial judgement going into each one, and I'm perfectly happy with that. To do them at Guy's typical rate of less than 5 seconds each (which seems to be our agreed rate for determining "'bot-like") and to do them simply as cite removals leaving the content dangling would not be the same thing at all – and that's what I don't support. Although I see that today you've started on the DM at one-a-minute rates citing an RfC which doesn't say "remove on sight" to remove them on sight – and that's much more of a problem.
As to " please list your work to reduce their use in practice " then you can stop that sort of tactic right now. That's sheer clique-building: "Only those engaged in the action are allowed to comment on the action" and we do not work that way here. Partly because it produces WP's wort feature of all, the wagon-circled self-sustaining and unchallengable cliques. But mostly because it's an incompetent and uncontrolled way to work: decide what's going to be done before starting to do vast quantitites of it, otherwise we get trapped into WP:FAIT. And if your implication is instead, "You don't have enough edit count to be challenging Me", then you can shove that one too. Andy Dingley (talk) 11:00, 12 November 2019 (UTC)
It appears that David Gerard has accurately described your behavior, and has backed up his description of your behavior with diffs showing the behavior. --Guy Macon (talk) 14:21, 12 November 2019 (UTC)


So, a specific allegation at last: you're claiming that JzG is removing sources by bot, and you can back that claim up?
Even then, that's a claim about a specific editor's behaviour - and it belong at WP:ANI. Where you have, again, been told repeatedly is the place to take your claims of that sort, and which you keep refusing to do, then continuing with the personal attacks. It is not a discussion point for deprecation in general, and if you aren't trying to divert discussion of deprecation, you should probably stop behaving like you are.
Your last paragraph seems to answer that you're not doing anything yourself then - ok - David Gerard (talk) 11:08, 12 November 2019 (UTC)
  • " you're claiming that JzG is removing sources by bot,"
<sigh> So this continual misrepresentation is just an attempt to goad me into an uncivil reply, which you can then use at ANI? That's rather pathetic. Andy Dingley (talk) 11:27, 12 November 2019 (UTC)
Your literal words, which I am responding to are: I note that you've been removing NotW sources yourself, but you seem to be doing them at a slower rate which indicates some sort of editorial judgement going into each one, and I'm perfectly happy with that. To do them at Guy's typical rate of less than 5 seconds each (which seems to be our agreed rate for determining "'bot-like") and to do them simply as cite removals leaving the content dangling would not be the same thing at all – and that's what I don't support. If you're not accusing JzG of cite-removal by bot, then you need to write more clearly - David Gerard (talk) 13:28, 12 November 2019 (UTC)
  • Are you having trouble with the phrase "'bot-like"? It's widely used on WP and generally well understood. For someone who has been here some time, I'd have expected you to be familiar with the concept. See WP:MEATBOT or even Commons:COM:BOT#Bot speed, where it's stated in more detail (and although Commons is not WP, their interpretations are helpful and non-contradicting here). Users may still be editing manually, but if they do this at such a rate that their basic accuracy starts to fail, then we have long considered that to fall under the same rules of conduct as for doing the same thing with 'bot automation (and that line has usually been seen at being around 5 seconds per edit).
The more and faster edits you are making, however you do them, the more pressure is on the controller of those to ensure that they're still accurate and appropriate. WP:FAIT also applies, and Guy has been criticised for using that in the past (the castles case). Andy Dingley (talk) 15:37, 12 November 2019 (UTC)
So you're accusing him of behaving functionally indistinguishably from a bot for Wikipedia purposes? And - and you haven't answered this one yet - are you accusing him of doing this bot-like editing to remove sources indiscriminately?
Because if you are accusing him of doing that, then that's an actionably bad thing for him to have done - assuming that you can show it (with diffs and so forth).
And if you aren't accusing him of doing that - what are you accusing him of doing?
The hard part here is getting you to actually state your claim that is the root cause of you going on and on and on about JzG, and not about the nature of deprecated sources.
I'm having to play twenty questions with you here because you're repeatedly using discussions of deprecation as a launchpad for endless not-quite-specific personal attacks on JzG, but without actually clearly stating an actionable claim against him. We could save a lot of back and forth if you could state your problem with him, that violates a Wikipedia rule or guideline, in a sentence, with what should be done about it - and not keep, every damn time, derailing discussions of deprecation of sources with something that's functionally indistinguishable from personal animosity - David Gerard (talk) 15:45, 12 November 2019 (UTC)
David Gerard, I am in fact using AWB, and removing (rather than tagging) ONLY where the source is redundant - i.e. adjacent to other refs supporting the same text. That's only a small minority of cases. Mostly (rough count something over 90%) I am just tagging. And the last run removed 100% manually - I opened the article and edited on Wikipedia. There have been previous cases where I have discussed a removal of a source, nobody has ventured an opinion, and there has been feedback later. In response, I have changed the way I work. But this is not really analogous, as those were half a dozen self-published sources, of which one was claimed (albeit still with no actual proof, I note) to be reliable. I have removed that one form my list and not looked at those sources since anyway.
The problem here is different. The source is deprecated, and there is a genuine difference of opinion on what deprecation means and whether it should happen at all, which is why (and I am amazed that I keep having to point this out) I am NOT removing sources in AWB UNLESS they are redundant to other sources. Some of it is relitigating past consensus. Some of it is complaining that the consensus is fine as long as we don't actually do anything about it. There are tens of thousands of links to these tabloids, thousands of them on biographies or supporting biographical information. I think that's a problem worht fixing. Guy (help!) 15:55, 12 November 2019 (UTC)
  • Can you please explain Reinhard Marx (Church Militant) in the context of "I am NOT removing sources in AWB UNLESS they are redundant to other sources.". You remove a paragraph which appears to represent a substantial statement of Cardinal Marx' position on refugees, somewhat contradicted by the following paragraph.
I would claim three things here: that statement is important - our coverage of Cardinal Marx is incomplete and unrepresentative without it. That it's a large claim, not a small one. The sort of thing which is likely to have been covered by multiple sources, although German Catholicism is not an area of my expertise. Also it appears to be the sort of claim which is so clear as to be hard to misrepresent, i.e. even the Church Militant was probably accurate in its statement here.
I don't care how you make these edits, I care what their result is. Now yes, I agree that Church Militant appears to have so much POV bias that it is useless as a source. If you claim "I did not use AWB to make that edit", then your claim above is literally true. No doubt you can claim that BLP permits any content to be removed. But yet – is your edit here really improving the article?
BLP does not merely mean "All statements must be reliably sourced", it also requires us to observe NPOV and to be careful in our selection of statements, so as to accurately reflect the overall description of our subjects. Selective removal goes against that.
You stated initially, " I propose to tag and then remove the couple of hundred links we have to this site." (WP:RSN#News of the World). You have since claimed, "I only tag them" and "I only remove them if they're redundant". Yet that's not what you stated you were intending to do. Nor is it what you did with Cardinal Marx. So which should I believe?
And just for clarity, I have no problem with anyone bulk-tagging deprecated cites as "better source needed" and leaving the cite, or removing that cite if it really is redundant. But either removing the content without some human review, or removing the last cite to leave content unsourced – that's what I've been against all along. Andy Dingley (talk) 18:56, 12 November 2019 (UTC)
That's really obviously an WP:UNDUE extended quotation cited to at best a primary source. It's aggrandisement for the subject, and adds nothing encyclopedic to the article. It's a perfectly reasonable editorial decision to make. You might disagree editorially, and that's fine - but in terms of claiming malfeasance, then if this is the best you have for a smoking gun, it's not very good, and you've hung four paragraphs of ranting which isn't really very well supported by the best example you could find - David Gerard (talk) 20:20, 12 November 2019 (UTC)
lol, Andy is now adding back the Daily Mail as a source on articles. Apparently the DM is just the source on hijabs. WP:POINT. This is probably not a good way to demonstrate your previous claim that you weren't posting here in support of the use of deprecated sources - David Gerard (talk) 00:24, 13 November 2019 (UTC)
I eagerly await Andy Dingley's explanation for the above edit. I will make popcorn. If it was anyone else I would have issued a warning and, if the behavior continued, opened up a case at WP:ANI. I refrained because I am biased against Andy Dingley. I think my bias is legitimate and based upon past behavior, but then again, I would say that, wouldn't I? It is not often that someone disagrees with their own opinions... --Guy Macon (talk) 06:13, 13 November 2019 (UTC)
Arguably, Andy's approach - keeping the DM source in there until a better source can be found, rather than removing the source and leaving a CN in place - is the better way to handle deprecation if we are talking from a software concept. Even if the DM source is sloppy or incorrect, the text within it and context in the WP article can help locate a much better source. Losing the source reduces how easy a replacement can be found. --Masem (t) 06:18, 13 November 2019 (UTC)
"Arguably", but not actually. I think you're going a bit far off into the meta-level theory there, and working way too hard to keep the deprecated source in - David Gerard (talk) 07:41, 13 November 2019 (UTC)
Any mass action - like mass removal of DM links - should be done in a way to minimize disruption. Stripping out DM links to content that is not immediately problematic and leaving behind CN is more disruptive than leaving the DM links in, with a "deprecated" notice, to be fixed by others later. --Masem (t) 13:55, 13 November 2019 (UTC)
If it's from the DM, it's sufficiently problematic to be removed. Misleading people with bad sources is bad. I've quoted the relevant parts of the WP:DAILYMAIL decision below; I ask you to please re-familiarise yourself with it, rather than keep coming up with reasons to resist it - David Gerard (talk) 17:15, 13 November 2019 (UTC)
If there was a time-expediency expected in either DM RFC closure, it would have been made clear. There isn't any, so we should be reasonably slow and careful with how DM links are removed, only after assuring either that there is a replacement source for it (including existing sources) or that the content sourced to it is removed with care as to avoid disruption. As RGloucester points out below, most DM cites were likely added in good faith when DM wasn't questioned. Those *still* must be removed or replaced, but there's no deadline for when that is to be completed by. --Masem (t) 17:45, 13 November 2019 (UTC)
David Gerard, if that isn't a [[WP:POINT] violation I don't know what is. The one area where the DM is likely to have strongest consensus against any use at all, is in respect of Islam. And Andy also restored material based on Bored Panda. That website is funny, the edit is not. Guy (help!) 13:49, 13 November 2019 (UTC)

Whether or not we agree with Andy's approach, it directly contradicts the decision of the community in the RfC. Nowhere was there any consensus for retaining DM links for the reason above, and certainly not for putting them back after they were removed.

In those rare cases where we want to retain the ability to use the DM source to help us find a better source without violating the RfC. We can do it like this:

 {{Citation needed|reason= https://www.dailymail.co.uk/Turn-out-that-everything-in-The-Onion-is-true removed per WP:DAILYMAIL |date=November 2019}}

...which gives us: [citation needed]

In cases where the claim is removed as well, we can replace it with

 <!-- Claim "Everything in The Onion is true" and reference https://www.dailymail.co.uk/Turn-out-that-everything-in-The-Onion-is-true removed per WP:DAILYMAIL --> 

--Guy Macon (talk) 07:50, 13 November 2019 (UTC)

or put it on talk, or something. But I've been doing a lot of DM removal lately, all by hand, applying a small amount of thought to each individual edit - if the fact is likely true, just remove DM and add a {{cn}} if needed; if not, remove the cite and the claim, and don't trust the DM, it's actively misleading. Over the course of several hundred? removals, I've really seen zero cases where there's reason to jump through hoops to keep the DM link lying around somewhere - David Gerard (talk) 08:16, 13 November 2019 (UTC)

Let's take the case where no existing use of DM is valid per the DM decision (used on BLP, etc.) The point when using the word "deprecation" is that we recognize all those uses need to go, but outright, blind removal of all uses could create more problems than leaving them in. Namely, we lose the citation on a statement that might be true and useful to retain in the BLP. That's where we need a human to check each use, see if there's a better source to support the same fact or otherwise remove the questionable fact and work the prose so that it doesn't leave the article hanging. I would strongly suggest getting a bot to run through and add {{Deprecated inline}} after each DM citation (for example), which will alert readers and point them to WP:RSP to see if they maybe can help fix. This again is similar to software space solutions, where one can add in code comments to call out deprecated function calls so that future developers can find alternate solutions. --Masem (t) 00:13, 12 November 2019 (UTC)

Namely, we lose the citation on a statement that might be true and useful to retain in the BLP. Yet again, you seem to greatly misunderstand WP:BLP - we don't keep statements in a BLP because they might be useful, and we especially don't if the statement was cited only to a source that we deprecated because it frequently lies and just makes stuff up, especially about living people! We remove the statement from the tainted source - David Gerard (talk) 14:58, 12 November 2019 (UTC)
I would assume that if we have a case where we have a highly contentious or dubious statement within a BLP which only relies on DM, and there's no other context about that statement, then it would be reasonable to remove that statement without having to go source hunting for a replacement or to leave behind a CN. But if we're talking a more pedestrian fact - one that still may be challenged but not something of significant doubt, then that's the point where leaving the DM link with some "deprecated" tag is reasonable (for example, outlining a footballer's career via team history). DM was not deprecated because 100% of their material was false; I would argue that somewhere between 80-90% of their material is still factually true and corroborates with other more reliable sources. But it is the fact that that 10-20% misleading/untrue information is far too high for us (we'd want something that's 1% or less) so we've depreciated it. DM links are not like copyvio links in that they present immediate harm to WP, but they need to be ultimately scrubbed to their essential uses to "complete" the RFC's deprecation approach, there's just no time factor here to remove. --Masem (t) 13:54, 13 November 2019 (UTC)
And yet, the consensus is that the links don't belong here and need to go. On BLPs - or any material concerning a living person, which as far as I can see way too many DM cites do - any cite to such a poor source does in fact need to go immediately. On any other subject, the generally prohibited in WP:DAILYMAIL means the onus is on the editor who wants to keep it. DM links themselves do, per WP:DAILYMAIL, constitute harm to Wikipedia - and the only reason they haven't been scoured already is that there's so damn many of them. Stop working so incredibly hard to throw up spurious obstacles to removing cites to a known liar of a source where the clear consensus is that it literally cannot be trusted - David Gerard (talk) 20:31, 13 November 2019 (UTC)
Masem, what I do is this:
  1. If the DM is one of a number of sources, remove the DM (I use DM as a proxy, it is the majority case).
  2. If the DM is the sole source and it's a biography and potentially contentious (e.g legal) I either replace with a better source or remove the content and the source.
  3. If the DM is the sole source and it's not a biography and probably not contentious (e.g. team sports results) I tag it and move on for now.
And all the above suggest to me that this is exactly in line with what everyone - including my critics - wants! Guy (help!) 15:58, 12 November 2019 (UTC)

To me it means do not use, avoid. That also means if it has been used we should not use it.Slatersteven (talk) 14:14, 12 November 2019 (UTC)

We know that's what you think it means. However that's at odds with both the dictionary definition, the common definition outside WP and also WP:DEPS. Andy Dingley (talk) 15:30, 12 November 2019 (UTC)
Which is why I said "to me".Slatersteven (talk) 15:34, 12 November 2019 (UTC)
I think that's a pretty reasonable concise summary of the first paragraph of WP:DEPS - David Gerard (talk) 15:50, 12 November 2019 (UTC)
  • Please, please don't remove the Daily Mail without first evaluating whether the content it was supporting can reasonably exist without such a citation. I think JzG's approach, as written above, is sensible. What is not sensible is an edit like this, which leaves behind text that was produced from the Daily Mail without actually attributing that text to its source. RGloucester 01:21, 13 November 2019 (UTC)
    • You'll be pleased to know I did, and thought at the time that was fine. If you disagree - which is also fine - why did you put back the source, rather than remove the text - David Gerard (talk) 07:40, 13 November 2019 (UTC)
Great, so we now we have a section called 'Airport battle' without any indication of when or how the battle started. I despise the Daily Mail as much as the next person, but please be sensible. If you're going to remove the Daily Mail, you need to work to compensate for its removal rather than just wreck articles. And yes, I do but the burden on the person making the change. RGloucester 09:18, 13 November 2019 (UTC)
I've sorted it out myself, as other sources were easily available. I really don't appreciate your flippancy, however. RGloucester 09:30, 13 November 2019 (UTC)
"flippancy" - no, it's bafflement at you putting back the deprecated, known-untrustworthy source which I'd left as a {{cn}}, then complaining I hadn't removed the text in question from the deprecated, known-untrustworthy source, which you were then making look like it was cited when it really wasn't trustably cited. We literally can't trust the DM, so the text was functionally uncited already - David Gerard (talk) 11:07, 13 November 2019 (UTC)
It was 'cited'. At the time the text was added, the Daily Mail was not deprecated. When whoever wrote that text inserted it, they did so in good faith. The relevant text was directly tied to the Daily Mail citation, so removing the citation whilst leaving the text creates a vacuum of attribution, which is in no way desirable. I am not sure why, if you think the Daily Mail is untrustworthy, that you'd want to leave material produced from the Daily Mail in an article WITHOUT indicating that it was originally from the Daily Mail. That makes no sense whatsoever. That's exactly how fake news starts to spread. Instead, the correct thing to do is to check whether other sources, which are reliable, support the text. If they do, insert them, and remove the Daily Mail. If they don't, remove the text. You did not do this. Your second edit was even worse than your first edit. Instead of bothering to check whether the relavent material was supported by other sources (it was), you simply removed it, destroying the flow of the whole paragraph without making any attempt to repair it. Please, please, be more careful. Replacing the Daily Mail with reliable sources is a noble mission, but simply removing such citations is not adequate. You must consider the nature of the material supported, check reliable sources, and make efforts to repair any prose damaged by removal of the Daily Mail. RGloucester 15:47, 13 November 2019 (UTC)
That it was added in good faith at the time does not mean it was a good source that should have been kept. You can disagree with the editorial decision - which should not be conflated with whether the DM should be kept, because it almost never should - but this claim is bizarre - David Gerard (talk) 17:31, 13 November 2019 (UTC)
  • I would agree with RGloucester here. WP:DAILYMAIL does not support the bulk removal of DM sources, without replacement, as is being done at present. Same with hijab.
Also there have been a lot of actions lately where someone says here "I will only tag cites, or remove redundant ones." (which seems to have general suport), but then goes off for a run of blanket removals, contradicting all of WP:DEPS, WP:DAILYMAIL and what they've personally posted here. Andy Dingley (talk) 11:31, 13 November 2019 (UTC)
Literally the text: Volunteers are encouraged to review them, and remove/replace them as appropriate. That's what I'm doing. Your replacement of it, and RGloucester's, directly contradicts also literally the text: its use as a reference is to be generally prohibited. Using it as a reference at all on a general topic is prohibited - David Gerard (talk) 12:13, 13 November 2019 (UTC)
  • as appropriate is clearly the stumbling block.
Also "generally prohibited" means "permitted", rather than "never permitted". This is not a blanket ban. WP:DAILYMAIL has never gone that far. Maybe WP:NotW would do (and even then, some self-referential primary uses might still be viable).
Neither RGloucester nor myself are adding the DM in these cases, we're simply opposing what we see as inappropriate removal in favour of a vacuum.
For Guy's edits, there's also the problem that he does them in rapid blanket runs without time for study of each, and also that he denies he does this. Andy Dingley (talk) 12:56, 13 November 2019 (UTC)
I would strongly suggest you drop the "bot-like" claims here if you are talking indirectly about JzG. One can load up multiple pages in a time in a browser, check each one, then rapid fire "publish" each change. The process JzG and their contributions show they are sticking to that in good faith in dealing with DM, the changes too complex to be just be non-human-oversighted, and JzG is clearly on board with treating deprecated as "remove with care" rather than "remove immediately". --Masem (t) 14:04, 13 November 2019 (UTC)
"generally prohibited" means "permitted", rather than "never permitted" - looking for loopholes is unhelpful behaviour. If you're not trying to support the continued use of deprecated sources, you probably need to stop working so hard to support the continued use of deprecated sources - David Gerard (talk) 17:15, 13 November 2019 (UTC)
"Generally prohibited" means that there is an increased burden on the person intending to use a source in a new cite, or retain the source that is proposed for removal. It basically means that every specific instance needs to be expressly justified to the satisfaction of others, and that the presumption is that any use of the source is invalid except in cases where it has been discussed and agreed to be valid. --Jayron32 19:23, 13 November 2019 (UTC)
Dictionary definitions of deprecate: "to not approve of something or say that you do not approve of something | to say that you think something is of little value or importance" (Cambridge), or "1. to express disapproval of 2. play down | belittle, disparage 3. to withdraw official support for or discourage the use of (something, such as a software product) in favor of a newer or better alternative (Merriam-Webster). Against that some editors are saying we should use an essay or care what the word means to them -- but Wikipedia editors are not dictionaries. When the real effect is not merely deprecation but prohibition, questioners should say. Peter Gulutzan (talk) 15:24, 2 December 2019 (UTC)


RfC on mainstream newspapers as RSEdit

See Wikipedia talk:Verifiability#Mainstream newspapers: please discuss there, not here. --Francis Schonken (talk) 08:51, 13 November 2019 (UTC)

Return to the project page "Reliable sources/Noticeboard".