🇮🇷 Iran Proxy | https://www.wikipedia.org/wiki/Wikipedia_talk:Reliable_sources/Perennial_sources/Archive_12
Jump to content

Wikipedia talk:Reliable sources/Perennial sources/Archive 12

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 5←Archive 10Archive 11Archive 12Archive 13Archive 14

YouTube is not a source

From its first mention in RSP (i.e., before there was a formal process for categorizing items), YouTube has been marked as a "questionable source". Template:Questionable source displays as "Generally unreliable" (red). However, this seems inappropriate to me for any large platform. I think WP:RSPYT should be marked as "additional considerations apply". Specifically, the "additional consideration" that applies is who the publisher is.

See previous discussions:

There has never been strong support for saying that a copy of a reliable source on YouTube, uploaded in the official channel of its publisher, is worse than a copy of that same source on a different platform. {{cite YouTube}} redirects to Template:Cite AV media, where YouTube is given as the example of how to use the |via= parameter. There are more than 150,000 articles citing YouTube (and that's counting only those using a citation template).

I therefore suggest that we make these changes, to set it to 'yellow' for additional considerations and to reorganize the description.

−
{{WP:RSPSTATUS|gu}} Most videos on YouTube are anonymous, [[WP:SPS|self-published]], and unverifiable, and should not be used as a reference. Content uploaded from a verified official account, such as that of a news organization, may be treated as originating from the uploader and therefore inheriting their level of reliability. However, many YouTube videos from unofficial accounts are copyright violations and should not be linked from Wikipedia, according to [[WP:COPYLINK]]...
+
{{WP:RSPSTATUS|nc}} Whether a video on YouTube can be cited depends on whether the publisher is reliable. Content uploaded from a verified official account, such as that of a news organization, inherits their level of reliability. However, most videos on YouTube are [[WP:SPS|self-published]], unreliable sources, and many YouTube videos from unofficial accounts are copyright violations and [[WP:COPYLINK|must never be linked from Wikipedia]]...

Does anyone object? WhatamIdoing (talk) 03:22, 25 July 2025 (UTC)

YouTube is a platform primarily consisting of user-generated content, and this list has historically classified this type of platform as generally unreliable. Every other listed social media website is classified similarly, e.g. Twitter (RSP entry). My understanding is that the classification is due to the proportion (vast majority) of the content on these platforms that would not be usable in any Wikipedia article due to their unreliability. — Newslinger talk 03:41, 25 July 2025 (UTC)
@Newslinger, how sure are you of your first assertion?
I found a website that loads random YouTube videos. Here's what I got (in order): two high-budget music videos, a professionally produced video of a woman wearing brightly colored clothes and being entirely too enthusiastic (estimated target audience: toddlers 12 to 24 months), two sports things from official sports channels (one was MotoGP; I don't remember the other), a clip from The Tonight Show Starring Jimmy Fallon, a video of someone's kid playing baseball, free clips from a Twitch streamer whose YouTube account has 19 million subscribers, a movie trailer from the studio's official channel, and a news report from the Tamil language news channel, Polimer News.
That's 10 randomly selected videos, and only one (10%) is just some random person on the internet. The rest may not be particularly useful to Wikipedia editors, but it would be difficult to justify calling them WP:UGC. It would even be difficult to justify calling them unreliable. They're mostly primary sources that are technically reliable for content we wouldn't want to include ("Sal Singer sang this song on this TV show on this date") or that we'd frequently choose a more convenient/text-based source for ("Rae Rockstar made a music video of this song"). WhatamIdoing (talk) 04:15, 25 July 2025 (UTC)
I doubt that those 10 videos are representative of the distribution of all YouTube videos, as we have no information on how the site selected them, and the video selection has a strong focus on high-budget content while excluding many common categories of YouTube videos such as reaction videos, influencer marketing videos, and YouTube Shorts. Research into YouTube video view counts has consistently found their distribution to be long-tail, just as with other social media sites. While the more polished content is suggested by YouTube's recommendation system more frequently, there are many more videos uploaded by individual creators who are not subject-matter experts, which makes their videos self-published and user-generated by Wikipedia standards. — Newslinger talk 05:42, 25 July 2025 (UTC)
All of these are still user-generated content; the users just happen to be companies. The point is that YouTube as a publisher gives very little assurance of reliability. Aaron Liu (talk) 13:26, 25 July 2025 (UTC)
I support this change, many professional news organizations have been moving away from hosting their own videos and towards the major commercial platforms like Youtube. Horse Eye's Back (talk) 03:45, 25 July 2025 (UTC)
I would rather leave this as unreliable. The quantity of possibly reliable content on YouTube, in comparison to what is unreliable, is tiny. Most editors adding YouTube links aren't adding news organisation or major commercial platforms. Additional scrutiny of YouTube links is both useful and necessary, given potential copy infringement. Discouraging it's use is also helpful in preventing link rot, as the Wayback Machine doesn't archive YouTube videos. -- LCU ActivelyDisinterested «@» °∆t° 10:37, 25 July 2025 (UTC)
@ActivelyDisinterested, do we have any way to investigate the assertion that Most editors adding YouTube links aren't adding news organisation or major commercial platforms?
I looked at a disabled Special:AbuseFilter. In addition to a couple of WP:ELOFFICIAL websites and a couple of false positives (e.g., rearranging existing links, someone typed in just youtube.com in plain text), I found a music video (by a news organization), a news report from Armenia (one of several from this editor), spam about a tech news website (in a section that attracts such additions), a clip from a music awards show (proving the singer won the award), and basically nothing that supports the claim that "most editors adding YouTube links" (back in 2019) were doing anything wrong at all. I didn't find any obviously unreliable sources being added. WhatamIdoing (talk) 16:15, 25 July 2025 (UTC)
Personal experience from dealing with thousands of articles, that's not a perfect source to base things on but neither is sample a few examples from a disabled filter. -- LCU ActivelyDisinterested «@» °∆t° 17:22, 25 July 2025 (UTC)
Grabbing the first an article that uses YouTube and should be some well watched there is Shaquille O'Neal, which uses YouTube 5 times.
  1. A video that is now private / permanently and unrecoverable dead, based on it's title I'm guessing it was copyvio.
  2. Copyvio
  3. Copyvio
  4. Reliable if promo source that is backed up to ghost archive (which backs up the video unlike Wayback machine)
  5. Another reliable source, but no backup exists so if the publiaher ever pulls the video it's will be useless.
That's the results for a high traffic BLP. -- LCU ActivelyDisinterested «@» °∆t° 17:43, 25 July 2025 (UTC)
I found seven in that article:
  1. https://www.youtube.com/watch?v=fI8bDq3x7Fg which is archived and from @TheGametimeHighlights (business with 762K subscribers, so probably not a serial copyvio source)
  2. https://www.youtube.com/watch?v=vjqL7bFTsIw which is archived and from @TheNBAHighlighter, which displays the NBA's official logo, which is not a behavior associated with copyvios. (Sure, it's possible that not just their copyright but also their trademark lawyers have been asleep on the job for over a decade, but I personally wouldn't rate that as a very likely possibility.)
  3. https://www.youtube.com/watch?v=mvzMou9qN1M (archive.org) also strikes me as a probable copyvio (low traffic account, no claim to affiliation, varied content, etc.)
  4. https://www.youtube.com/watch?v=b6Mv-30qYoQ, I agree
  5. https://www.youtube.com/watch?v=OYL11q4hb3Y, account closed, looks completely unreliable (I've removed this).
  6. https://www.youtube.com/watch?v=SXC5QKLHVUo, official sports channel
  7. https://www.youtube.com/watch?app=desktop&v=yFQmosVrLO0, official sports channel
WhatamIdoing (talk) 00:22, 26 July 2025 (UTC)
You marked those as archived as they are on the wayback machine, but as I said the wayback machine does not archive YouTube videos. It just creates a broken page without saving the video. Unless someone had the forethought to manually archived it at ghost archive it's useless. -- LCU ActivelyDisinterested «@» °∆t° 10:56, 26 July 2025 (UTC)
The archive links WAID links do have the videos archived, though. IIRC the Wayback Machine started archiving YouTube videos in the 2010s. Aaron Liu (talk) 13:21, 26 July 2025 (UTC)
I didn't attempt to play the videos. But it's archived "enough" to determine whether the videos are WP:NOTGOODSOURCES, i.e.:
  • It has a reputation for fact-checking and accuracy.
  • It is published by a reputable publishing house, rather than by the author(s).
  • It is "appropriate for the material in question".
  • It is a third-party or independent source, with no significant financial or other conflict of interest.
  • It has a professional structure in place for deciding whether to publish something, such as editorial oversight or peer review processes.
Note the absence of any criteria that sounds like "It is demonstrably non-self-published, but I don't like their choice of web service for distributing the source to the public. All true editors agree that reliable sources run their own websites instead of using popular commercial alternatives that are also available to the unwashed masses". WhatamIdoing (talk) 15:35, 26 July 2025 (UTC)
Really? "I like it" is also a made up quote, and not like an argument that anyone has used. -- LCU ActivelyDisinterested «@» °∆t° 17:03, 26 July 2025 (UTC)
How else would you describe the POV that any specific video should be assumed to be a copyvio solely because it's possible to watch it on YouTube? WhatamIdoing (talk) 22:58, 14 August 2025 (UTC)
What?! I have no idea where you've got that from. -- LCU ActivelyDisinterested «@» °∆t° 11:35, 15 August 2025 (UTC)
With over 20 billion videos, I'd estimate that >99.9% are user-generated and not reliable. There are maybe upto 20 million videos (~0.1%) from reliable news orgs that are reliable. This clearly puts it in the category of generally unreliable. This isn't a case of YT is reliable so long as A, B and C is considered, as it does not concern the overwhelming majority of user-generated content. The fact that approximately half of the most popular YT news channels comes from established news orgs misses the entire point here regarding overall content. Nor is it the case that there lacks consensus (NC) regarding user-generated content as being generally unreliable. CNC (talk) 11:08, 25 July 2025 (UTC)
"NC" isn't just "lacks consensus". It's also the code for "Additional considerations apply". It's this latter that I think is relevant: You have to evaluate the individual video, rather than YouTube as a whole. WhatamIdoing (talk) 15:59, 25 July 2025 (UTC)
insert shameless plug of my idea to separate NoCon from AdCon here, which would probably need an RfC at this point Aaron Liu (talk) 17:10, 25 July 2025 (UTC)
  1. Most UGC sources with identifiable authorship are marked as generally unreliable and have the same expert-published exception.
  2. There is a RfC behind marking YouTube as GUnRel. If you want to change this status, you probably need to ask RSN instead.
Aaron Liu (talk) 13:25, 25 July 2025 (UTC)
Do you have a link? The 2020 RfC is about implementing an edit filter not reliability per say, or at least that's the closing statement. According to @Hemiauchenia from that RfC, YT was added to RevertReferencesList after the Facebook RfC, which again I'm not seeing the direct reference of either. Hoping @ProcrastinatingReader can shed some light on this as the closer. Genuinely didn't get anywhere near finding any form of consensus on YT, or edit filter use to be honest. I assume it's in one of the 35 discussions that has occurred over the years, or came via technical request post FB RfC? CNC (talk) 14:50, 25 July 2025 (UTC)
Yes, that RFC link is a little bit misleading, but I don't see any way in RSP's tools to say "This was an RFC, and it's relevant, but it wasn't really about what to say here at RSP". WhatamIdoing (talk) 15:46, 25 July 2025 (UTC)
Ah, that's very misleading. The intention of the RfC labeling is clearly for highlighting the an uninterrupted request for comment on the source's reliability that took place on the reliable sources noticeboard as mentioned in the inclusion criteria. I'll change the styling of that into just another regular discussion. Aaron Liu (talk) 17:09, 25 July 2025 (UTC)
Thanks, have updated the discussion count to 36. CNC (talk) 17:15, 25 July 2025 (UTC)
  • Youtube is a medium or a platform and not a source. We don't generally comment on radio or television as a source and much of the arguments here would apply equally. GMGtalk 14:52, 25 July 2025 (UTC)
This argument I still don't get (after skimming through the previous RfC). The basic definition of a source is "[a] thing from which something originates or can be obtained" (Google); no-ones describing it as publisher here, it's just the origin as the medium/platform as you describe. It's been used as a source 244,000+ times, so it's not immune from criticism, in fact the opposite is true. It's merely included in RSP because it has been discussed repeatedly at RSN, and while TV and Radio stations are in fact discussed often enough at RSN, it's usually not enough to warrant an RSP entry. Being included at RSP otherwise has nothing to do with the validity or the source, only the frequency of discussion, thus the "it's not a source" arguments sounds like semantics to me. The sniff test of a source is whether you can wrap a <ref> tag around it in order to use as a citation. Unless there is some alternative WP meaning of source here, it's a pretty low bar, any url will do. CNC (talk) 15:14, 25 July 2025 (UTC)
If NBC News puts a video on their YT channel, then NBC news is the source. It is immaterial whether that content appeared via television on their nightly broadcast or via YT. GMGtalk 15:19, 25 July 2025 (UTC)
@CommunityNotesContributor, from WP:SOURCE:
"A cited source on Wikipedia is often a specific portion of text (such as a short article or a page in a book). But when editors discuss sources (for example, to debate their appropriateness or reliability) they are usually talking about one or more related characteristics:
  • The work itself (the article, book) and works like it ("An obituary can be a useful biographical source", "A recent source is better than an old one")
  • The creator of the work (the writer, journalist: "What do we know about that source's reputation?") and people like them ("A medical researcher is a better source than a journalist for medical claims").
  • The publication (for example, the newspaper, journal, magazine: "That source covers the arts.") and publications like them ("A newspaper is not a reliable source for medical claims").
  • The publisher of the work (for example, Cambridge University Press: "That source publishes reference works.") and publishers like them ("An academic publisher is a good source of reference works")."
Note the absence of anything about "the method used to distribute the work" or "the platform used for hosting the work" or anything like that.
GMG gives an example from NBC News. I expect that all experienced editors agree that last night's NBC Nightly News broadcast, which is at both of the following two links (and probably others, for that matter):
is a reliable news source no matter which of the links, leading to identical content, you use.
But the way WP:RSPYT is color-coded, I can easily imagine someone glancing at it and saying "Oh, no. YouTube is Generally Unreliable. It's colored red to warn you away. You shouldn't use anything from the https://www.youtube.com/@NBCNews official channel because YouTube is bad." WhatamIdoing (talk) 15:56, 25 July 2025 (UTC)
I get these examples are "usually" (note: not explicitly) examples of reliable sources, I'm not arguing against these characteristics and I've already clarified I don't believe YT is generally reliable. It's interesting to see what editors are referencing to support their arguments over what a source is, so I appreciate the referencing. That said, examples aren't definitions nor is this a universally accepted position (per the RfC). Hence I'll stick to WP:UGC which is crystal clear here: Websites whose content is largely user-generated are generally unacceptable as sources (emphasis included). This if further clarified in the following paragraph "Examples of unacceptable user-generated sources [...]" (emphasis included). This completely eliminates any ambiguity that YT, as UGC, is in fact categorized as a source per wording of guidelines. This is why it has been discussed 35 times at a noticeboard specifically to discuss source reliability, and thus categorized at RSP. Granted there's obviously a lack of consensus over what is a source, despite this UGC inclusion, so that's a big problem that needs resolving. CNC (talk) 17:05, 25 July 2025 (UTC)
I don't see anyone arguing that Youtube is generally reliable... The request is for additional considerations apply which seems appropriate in this context. Horse Eye's Back (talk) 19:04, 25 July 2025 (UTC)
@CommunityNotesContributor: I suggest looking at the wikipedia definition of a source rather than asking google, there is often a wide gulf between the most common understanding of a term and what it means in our specific context. Horse Eye's Back (talk) 16:02, 25 July 2025 (UTC)
It might be better to start that platforms like YouTube, Vimeo, TikTok, etc. are not publishers, but simply the medium which users can upload and share videos, so that in terms of evaluating sources, we consider who the uploader is, not the platform. These platforms are neither reliable or unreliable. Masem (t) 19:34, 25 July 2025 (UTC)
I agree with this change and would generally agree with changing this for all social media sites. IMO saying that Youtube is unreliable because most Youtube videos are unreliable is like saying the internet is unreliable because most websites are unreliable.
I also think the current status of YouTube tends to make it harder to use new-media-primary WP:NEWSORGs. It was surprisingly annoying to get People Make Games declared reliable, for instance, and linking to their channel with certain source marking userscripts (e.g. this one that I use) still marks links to PMG's YouTube channel as unreliable even though they're not. Loki (talk) 19:55, 25 July 2025 (UTC)
I agree with the view that YouTube is a platform, not a source, and treating it as a source is inappropriate. I think we should generally try to avoid confusing the two. Of course, for sites that are exclusively generated by anonymous users, like Wikipedia itself, it’s useful to warn editors away. But when a site has significant content that is actually reliable, as YouTube does, it is a disservice to imply that the entire platform is unreliable. John M Baker (talk) 00:10, 26 July 2025 (UTC)
Not long ago I was reading an article in the Washington Post which included a video. That video was hosted on Youtube, but the mere fact that it formed part of the WP article makes it a reliable source in principle. It should be possible to write the rules to allow such cases without allowing too much. Zerotalk 02:42, 26 July 2025 (UTC)
  • There are effectively two different proposals here, one that YouTube should not be considered a source at all, and one that its classification as a source should be changed. However, this page is supposed to reflect external consensus, and in either case there isn't an external consensus to support the change. Using an "additional considerations" classification would involve overruling not only RSN but also WP:UGC, which says that such sources are "generally unacceptable". I suppose classifying them as non-sources could work, thus creating a fourth category for UGC sources, but that would still implicitly change the meaning of the term "UGC" in a way that would be idiosyncratic to this page. In addition, it wouldn't reflect any practical difference, as UGC sources would still be treated in the same way as other generally unreliable sources - usually unacceptable with some exceptions.
The category of "generally unreliable" inherently allows for exceptions. If an exception applies, the source is perfectly acceptable to use, and this is not a challenge to the classification itself. The fact that UGC sources inherit the reliability of their publishers is a well-established exception, and in this case the summary explicitly points it out. Theoretically, yes, this is a type of additional consideration, but in that case so is ABOUTSELF; they both indicate cases where an otherwise unreliable source can be used, so the argument can be expanded indefinitely until it applies to every source.
The idea that citing YouTube videos from a reliable source is worse than a copy of that same source on a different platform is based on reasons such as copyvio, linkrot, or accessibility. However, RSP is a reliability classification; there is no reason related to reliability to discourage this practice, and the RSP summary does not suggest that there is. Certainly the classification can be misused if applied without context, but this does not change the general rule that YouTube is usually unreliable (or that an "additional considerations" classification would also be misused, and likely at a much higher rate, given e.g. the Shaquille O'Neal example above). However, if the RSP classification and/or the UPSD script are too simplistic for an individual editor, perhaps WP:CiteUnseen would be a useful alternative. Sunrise (talk) 08:42, 26 July 2025 (UTC)
Thanks for summarizing the discussion so far, there are indeed multiple issues documented here:
1. Lack of consensus on whether YT and other platforms are sources or not (this goes well beyond YT classification).
2. Lack of discussion in a centralized noticeboard, such as RSN or UGC, notifying the other in the process (a basic requirement).
3. Lack of RfC reference regarding the current consensus of YT (consensus appears deferred from UGC, could be wrong).
4. Request to re-categorize YT from GUNREL to MREL at RSP, rather than at RSN where it belongs (not an uncontroversial request).
It should be obvious to most by now that this needs a well crafted RfC in an appropriate location to resolve the abundance of issues, as well as lack of consensus, as this serves as nothing more than RFCBEFORE at present. CNC (talk) 09:05, 26 July 2025 (UTC)
This makes sense to me. YouTube - and social media sites in general - blur the line between medium and publisher. In those cases where there is some sort of verifiable official publisher (eg. BBC or Sky publishing to their own channel) the material should inherit the reliability of that publisher, and everything else is SPS (or copyright violations). Void if removed (talk) 10:55, 26 July 2025 (UTC)
Just to note that this situation doesn't just apply to YouTube, all UGC sources are considered unreliable. Whether that twitter, Facebook, wordpress, or some vlog site. That some of the content on such sites might be useable as an exception, because the publisher is reliable or the author is an EXPERTSPS, doesn't change any of the others. So I don't see why we should allow an exception from the rule for YouTube. -- LCU ActivelyDisinterested «@» °∆t° 11:05, 26 July 2025 (UTC)
The reason for an "exception" is that YouTube isn't a source in the first place. YouTube is more analogous to "a bookstore" than to "a book". When we wrote WP:UGC, we were primarily thinking about "the Internet Movie Database (IMDB), Cracked.com, CBDB.com, collaboratively created websites such as wikis, and so forth, with the exception of material on such sites that is labeled as originating from credentialed members of the sites' editorial staff, rather than users.
The list of named websites has gotten much longer since then, but it tends to have the same feel: "User-generated content" is part of the self-published sources, and UGC in particular is primarily about self-published collaborative authorship.
Saying "Oh, YouTube is an unreliable source because it's a video host" is also logically inconsistent. Video-hosting platforms, like all websites, are likely to contain self-published content. But Vimeo is a video-hosting platform, and we never say "Vimeo is unreliable per UGC because it's a video-hosting platform". Microsoft Stream is a video-hosting platform marketed to businesses, and we never say "MS Stream is unreliable per UGC because it's a video-hosting platform". Amazon Web Services is a video-hosting platform, and we never say "AWS is unreliable per UGC because it's a video-hosting platform". The fact is that the hosting platform doesn't actually determine whether the source is reliable. What matters is the author and the publisher, not the location of the files. WhatamIdoing (talk) 15:30, 26 July 2025 (UTC)
The comparison to AWS and YouTube isn't valid, YouTube is not a background service. The much better comparison is WordPress or BlogSpot, sources that let you publish content that they then host. -- LCU ActivelyDisinterested «@» °∆t° 17:07, 26 July 2025 (UTC)
I would really like us to try to not classify YouTube or the like as a publisher. If we were filling a cite template, they would be listed in the "via=" part of the template, rather than "publisher=", outside of videos specifically created by YouTube (for example, the old YT Rewinds). A YouTube video, for example, should be considered self-published via YouTube because outside of moderation activities, YouTube does nothing to review the video for content, where as what we call a publisher traditionally (like a book publisher) is going to at least review the book to make sure that's something they want to publish. Which is why YouTube et al should be neither reliable nor unreliable. Masem (t) 17:39, 26 July 2025 (UTC)
Yes they host but do not publish the content, channels on YouTube are similar to blogs on Blogger. -- LCU ActivelyDisinterested «@» °∆t° 17:51, 26 July 2025 (UTC)
And the way to determine whether a blog post on Blogger is reliable is to figure out who actually wrote and published it, and not to assume that (e.g.,) a small town newspaper deciding to use Blogger's website instead of setting up their own is 'generally unreliable' because everything on Blogger is 'generally unreliable'", right? WhatamIdoing (talk) 01:48, 27 July 2025 (UTC)
Maybe, but consensus is that YouTube and Blogger and Medium and Substack and et cetera (quick challenge: find the RAS syndrome in this enumeration!) are all GUnRel. At this point your best path forward would probably be to start an RfC at RSN on what all such "proxying mediums" (please don't actually call them "proxying mediums" in the actual RfC I beg you) as a category should be classified as. Aaron Liu (talk) 02:03, 27 July 2025 (UTC)
I think that a "platforms" category is probably the right approach. The explanation would be something along the lines of "Check the actual publisher, not just the hosting service".
But skimming back through this conversation, it feels like a subject that some editors will misunderstand. I mean, we all know, once our brains are engaged, that there's a difference between a video on YouTube from a notable news outlet and a video on YouTube from an ordinary person. But getting this adopted requires getting the brain engaged, plus overcoming the fear that if it's not a big scary red entry, then somebody will abuse it (the old 'written rules must overstate, so we can defend the wiki against bad actors' model of policy writing). WhatamIdoing (talk) 23:16, 14 August 2025 (UTC)
Coming back to this because I just saw a RSN discussion where TLDR News was dismissed out of hand due to this. Now, by all accounts TLDR News is a fairly ordinary WP:NEWSORG except for publishing almost exclusively on YouTube. They have ordinary journalistic processes and a published editorial policy, albeit in video+Q&A format. This would ordinarily make them pretty standard reliability except that because YouTube is unreliable it creates this strange bias where they're now unreliable as well. Loki (talk) 04:31, 15 August 2025 (UTC)
again why I think we should be clear "Youtube is neither reliable nor unreliable as they are only a platform for individuals or groups to publish videos with no editorial oversight, so to determine reliability, one must look to the individual or group that published the video." Masem (t) 05:01, 15 August 2025 (UTC)
As a reminder, RSN is this way to discuss such matters. CNC (talk) 09:35, 15 August 2025 (UTC)
RSN is the usual place for discussing a specific source. I think this page is a better choice for discussing whether the categories we put on this page are sensible.
@Masem, in addition to WP:RSPYT, what else in this page is more 'platform' than 'source'? WhatamIdoing (talk) 16:14, 15 August 2025 (UTC)
This logic by default assumes YT is not a source and there is no consensus for this ideology, not in this discussion, nor in the previous 36 documented AFAIK (it's probably close but no cigar last I checked and not a lot has changed since then). If consensus had been gained here or another central location (with relevant talk pages notified) that it's not a source then sure, it'd make sense to the update list, and a discussion would barely be relevant by that point. Otherwise to overturn the de facto status quo that it is a source, then UGC would also need notifying of the discussion, in order to avoid contradicting current guidelines (as I've already said). So given I've already said this, and nobody has bothered (AFAIK) to notify RSN or UGC, I'll leave it alone now, as this opinion must be considered irrelevant, misunderstood, and/or being ignored. I'm just trying to avoid the inevitable blow up from RSN/UGC if major changes are made to sources without appropriate and necessary consensus . I can only sound the alarm as a preventative measure and I've done that now. Regards, CNC (talk) 16:45, 15 August 2025 (UTC)
@CommunityNotesContributor, I'm not sure we're working from the same set of facts. For example:
  • WP:UGC doesn't actually mention YouTube by name.
  • Wikipedia:External links/Perennial websites#YouTube says that YouTube videos from official channels are acceptable ("If the source would normally be considered reliable (e.g., a segment from a well-known television news show, or an official video channel from a major publisher), then a copy of the source on YouTube is still considered reliable").
Those "36 documented" discussions (several of which are about whether getting X subscribers or Y views is evidence of Wikipedia:Notability) include statements like:
As the saying goes, you're entitled to your own opinion, but not to your own facts. This is a lot of editors saying that YouTube is a platform and not a WP:SOURCE for anyone to be saying something like YT is not a source and there is no consensus for this ideology, not in this discussion, nor in the previous 36 documented. WhatamIdoing (talk) 18:40, 15 August 2025 (UTC)
All I think we should be doing on RSP here is to explain that YouTube (along with a host of other user-content-sharing platforms like Twitter/X, Vimeo, Instagram, etc, where there is no/minimal editorial control of material before its posted) should not be judged as the source when it comes to evaluating whether the material is reliable, and this question has come up sooooo many times that it should be an entry on RSP even if we are actually saying "YouTube is not a source". It makes sense to answer that question on RSP and not hide it elsewhere. Masem (t) 20:11, 15 August 2025 (UTC)
It's good to discuss categories, but to make this category yellow instead of red you'd need an RSN discussion. Aaron Liu (talk) 04:24, 16 August 2025 (UTC)
If we go the route of introducing a separate category, I think we need an RFC, but it doesn't necessarily need to be at RSN itself. (Though one might reflect on why we believe we need an RSN discussion to change the color, when there was never an RSN discussion to choose the original color. WP:NOTBURO?) WhatamIdoing (talk) 17:39, 16 August 2025 (UTC)
Indeed, with WP:STATUSQUO being the main reason here rather than BURO applying (for YT at least). RSN or here is fine for an RfC, with the other being notified, along with RS. FYI this is why I didn't respond to your previous lengthy reply as I wasn't arguing YT is a source in my previous comment, only trying to highlight the current consensus (whether you, I, or others agree with it or not). Otherwise the more I hear of arguments for YT (along with others like FB, Instagram and X) being categorised as platforms the more I agree in principle (as a compromise to avoid being categorised as MREL predominantly), but I'm unable to support this change in this discussion based on the purpose or RSP: to summarise established consensus (or lack of). It's not intended for establishing a change in consensus, only the interpretation of it. We also have the RfC's of Facebook and TikTok/Instagram to overturn as well. For an RfC, I'd also strongly recommend avoiding the selective pinging of editors who have described YT as a platform to support your argument, per WP:CANVASSING. Either ping all relevant contributors of such discussions, or don't bother, otherwise it's got a good chance of derailing it's legitimacy. CNC (talk) 18:34, 16 August 2025 (UTC)
The entries for Twitter, LinkedIn, Blogger, LiveJournal, Flickr, Medium, WordPress, Stack exchange, etc would also need to be looked at. Also maybe HuffPost and Forbes contributors, as the reason they are marked as unreliable is that they are only hosted rather than being under editorial control of Huff/Forbes. -- LCU ActivelyDisinterested «@» °∆t° 19:29, 16 August 2025 (UTC)
+1. We need an extensive list of UGC documented at RSP that such a broad change would cover, thus probably worth having a sub-section for RfC crafting/workshop at this point. Using "etc" in an RfC is not going to cut the mustard for a sweeping change (as I have also been doing for reference). Also agree this would cover contributors, but others might disagree with that. CNC (talk) 19:43, 16 August 2025 (UTC)
At least with Forbes contributors, that is not the same as YouTube; contributor articles there do require some type of approval before they are published, but there's no strong editorial control for that; you also have to be sign up and be approved to become a contributor rather than with YouTube where you just need an account. I'd assume the same is true for HuffPost.
But a key thing here is that this WP:RSP, and the question about YouTube is absolutely a perennial question, compared with things like Twitter and those others which are nowhere close to being sources that are asked about frequently. It would be best to address the YouTube here now, and perhaps then develop a separate guideline about such sites. Masem (t) 20:08, 16 August 2025 (UTC)
If we are going to create a new category then it is appropriate to consider all the similar sources that may become part of that category at the same time. -- LCU ActivelyDisinterested «@» °∆t° 09:14, 17 August 2025 (UTC)
As CNC mentions I feel like RSN's appropriate and necessary notification since its participants would probably be interested. Aaron Liu (talk) 16:26, 19 August 2025 (UTC)
I expect this to be a very widely advertised RFC, no matter what page it's on. It might be best to put it on a separate page, e.g., WP:Requests for comment/RSP platform category.
The more urgent question is how to make this idea legible to someone who just glances at the RFC question. We can hardly write "It's stupid to say 'on YouTube, therefore it's reliable/unreliable' – you've got to look at the uploader/publisher", but that's basically what we're trying to communicate. WhatamIdoing (talk) 16:40, 19 August 2025 (UTC)
I wouldn't use TLDR not because they publish on YouTube but because they are relatively new and not 'well-established' per NEWSORG. -- LCU ActivelyDisinterested «@» °∆t° 11:38, 15 August 2025 (UTC)
TLDR News says that it was started in 2017. What's your standard for "relatively new"? For one point of comparison, they've been publishing news for twice as long as you've been editing Wikipedia. WhatamIdoing (talk) 16:13, 15 August 2025 (UTC)
In 2017 there was a YouTube channel and one person, something far different from what it is now. If it had started in 2017 with as it is now that would be different, but using a date without any context isn't a good comparison. For instance I edited for some time before making an account, so just using the date my account was created can't tell you how long I've edited. -- LCU ActivelyDisinterested «@» °∆t° 17:09, 15 August 2025 (UTC)
It's still been around for eight years. When does a news outlet stop being "relatively new"? WhatamIdoing (talk) 17:24, 15 August 2025 (UTC)
That will depend on the organisation itself, but it won't be some arbitrary start date unless it started as a fully established organisation. -- LCU ActivelyDisinterested «@» °∆t° 17:35, 15 August 2025 (UTC)

YouTube is not a source. Its a publisher of other sources. I think its generally fine and preferable to other video platforms (eg facebook, tiktok, instagram), but the actual source of the video should be reliable (eg. a CNN video vs a Daily Mail video). Metallurgist (talk) 18:13, 15 August 2025 (UTC)

  • Generally agree with others saying that YouTube is not itself a source, it is a medium, but it is one which hosts an absolute trash dump of unreliable and unusable information. What that means to me is that any YouTube video used as a source should be presumed unreliable, unless it can be demonstrated that the specific source meets our criteria for reliability. The onus should be on editors wanting to add information to demonstrate that their source is reliable; it should not be on editors reviewing additions to demonstrate that it is not, and I see the proposed change as moving the goalposts in the wrong direction. Ivanvector (Talk/Edits) 17:39, 16 August 2025 (UTC)
    Isn't that true for anything on the internet? www.RandomWebsite.com is, and IMO should be, presumbed unreliable (and also self-published) unless and until it can be demonstrated that the specific website is reliable. WhatamIdoing (talk) 17:41, 16 August 2025 (UTC)
    I think I made the first part of my point poorly. The reliability of a Youtube video can't be evaluated on the basis of it being a Youtube video, it needs to be evaluated based on the reliability of the actual publisher of the video, which is not Youtube. It's roughly the same as evaluating the reliability of a newspaper article on the basis of the press it was printed on. I think it's correct that Youtube probably doesn't belong in a citation at all except in a |via= parameter, but I'm not sure that either the original or proposed RSP entries really handles the situation. But I think it's more appropriate to describe it as "questionable", because if it's described as "no consensus" then it will spawn arguments about there being no consensus against using Youtube as a source. Ivanvector (Talk/Edits) 18:14, 16 August 2025 (UTC)
    We might need two entries:
    • YouTube as a platform (it's really the TV news show, not YouTube), in which case there's a consensus that it's a platform instead of a source, and
    • YouTube as a primary source of for information generated directly by YouTube (e.g., how many views or subscribers), in which case there's a consensus that it's a questionable source at best.
    WhatamIdoing (talk) 18:18, 16 August 2025 (UTC)
    I think you're right, but I would argue for a blanket rule that Youtube itself is not an acceptable source for Youtube viewership, due to issues of COI and WEIGHT. Viewership should be based on independent sourcing, both for verifiability and to establish relevance. Ivanvector (Talk/Edits) 18:22, 16 August 2025 (UTC)
    I assumed a change to platform entries wouldn't be yellow for MREL, but instead a neutral colour indicating 'indeterminable' (or similar) based on being a platform and not a source. White would be that that colour imo with a new entry in the legend to clarify this, assuming there is consensus for such a change in a future RfC. This would benefit by maintaining/updating the current platform entries, instead of lumping them all together into one, given there are subtle nuances across these differing platforms which would be worth retaining (YouTube, Facebook, Instagram, TikTok, Medium, etc). CNC (talk) 19:14, 16 August 2025 (UTC)
    Agreed with this change. We should not say that YouTube is marginally reliable/additional considerations apply, because we shouldn't say it's of any reliability level. Loki (talk) 19:30, 16 August 2025 (UTC)
  • I've made this point before, and should probably make this point in many places throughout this discussion. Yes, there is reliable material hosted on YouTube, but it is a minority. The percentage of material that is useable as a reliable source is staggeringly low, and quite some of it should not be directly linked but linked through the publisher.
  • A news company, 'Generic News Network', is interviewing that guy that everyone knows, 'John Doe'. They have a page on their website about the interview with some metadata, and embedded the video, that they uploaded on YouTube. We should not be linking to the YouTube video (even if that is where you orginally foundit), we should link to the page on the website of GNN. Similarly for the chemistry department of Generic Technical University, maintaining a list of videos where they explain explosions of different materials. That is the case for the vast majority of reliable material. In 2020, BBC had 11944 videos on YouTube out of (give or take) 10 billion (7 billion in 2017). Not in scope here, link to the page that embeds the video, the publisher. No entry for YouTube needed, publisher could go on RSP for being generally reliable.
  • Similar, but for very unreliable 'VerySpecific News Network'. Thousands of videos. Not to be linked, not to the embedded page, not to YouTube. (Note, I don't expect this volume to be astoundingly different from the volume from reputable sources, and not all of it will be linked on either anyway). Again, publisher goes on RSP as unreliable.
  • Then there is a small fraction of material that is directly uploaded by a source that we define as reliable. Not embedded on another website. Given that BBC (in 2020) had 11944 videos out of the (give or take) 10 billion that were there I would expect that most of these would have less than 50. Are all going to be linked? No, but some yes. Worth an entry. No. (and if so, for the publisher being reliable)
  • Then there are artists uploading their music videos (with copyvio copies sometimes appearing, or those with added lyrics. With few exceptions not used as a source (but as an external link). Not worth the entry.
  • Then there is the rest. Influencers, grandma birthday videos, dogs and cats doing funny stuff, UFO sightings, AI generated videos of fiery meteor impacts on the dark side of the moon, someone doing an interpretation of 'We've only just begun', unboxing of boxer shorts, drone overview of my neighbours greener garden. Billions upon billions of video's. Unreliable home made stuff with at best a primary source that it can be seen in grandma's birthday video that 'Generic city' has a 'McGeneric' restaurant because that is where grandma lives and that is where grandma celebrated her birthday.
Should YouTube be on this list? No, with a few exceptions, youtube itself does not publish. Agree with the change (maybe use 'However, by far most videos ...'. Dirk Beetstra T C 11:34, 17 August 2025 (UTC)
  • I do think RSP needs to keep its entry on YouTube (as this has become a perennial question)… and I agree that the entry should explain that YouTube is a platform, not a source (linking to a more detailed essay that explains the distinction).
I would also agree that we need a new separate RSP category (with a different color… perhaps teal?) for platforms. Blueboar (talk) 19:19, 17 August 2025 (UTC)
I'm going to start a new section to draft a possible RFC question. Interested editors are invited to join me there. WhatamIdoing (talk) 16:10, 19 August 2025 (UTC)

"Wikipedia:Tabloid" listed at Redirects for discussion

The redirect Wikipedia:Tabloid has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2025 August 25 § WP namespace "tabloid" redirects until a consensus is reached. Left guide (talk) 21:58, 25 August 2025 (UTC)

WP:DEPD now exists, providing a collection of edit-filter and revert-list–discussion links by domain. Editors who want to see the relevant edit filter links probably want to see the revert-list–discussion links as well, so I'm no longer seeing any use for the edit-filter links on this page. What do we think about removing them? Aaron Liu (talk) 20:15, 31 August 2025 (UTC)

In the interest of limiting RSP's page size, I think it would be best to remove the edit filter links from here. When a source is deprecated and not blacklisted, its domains are subject to the edit filter. The only exception is the National Enquirer (RSP entry) due to its unique RfC result which is explained in its entry description. This means that even without the edit filter diff links, any editor who reads an entry of a deprecated source on this list can still tell whether the source is edit-filtered. The date that the edit filter was applied to the source is unimportant for the vast majority of editors who consult this list for guidance on how to use sources, and any editors who want to see the implementation details can go to WP:DEPD for the extra information. — Newslinger talk 00:37, 1 September 2025 (UTC)

Template:Press BRD

Discussion at Talk:CNN#Template:Press_BRD, but the same revert happened on this talkpage, so if you have an opinon, please join. Gråbergs Gråa Sång (talk) 16:52, 6 September 2025 (UTC)

New York Post

This is a waste of time. The opinion of a single person is not going to overturn a community consensus on the issue. Any actual discussion related to the NYPost should be held at WP:RSN, not here where it is completely inactionable. Hemiauchenia (talk) 12:12, 11 September 2025 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

By excluding New York Post as a reliable source, this "consensus" affirms that editors here are biased against a politically conservative news organization, as many people already believe about Wikipedia.

This policy should be reversed and each New York Post story should be judged on its own merits. Queens Historian (talk) 11:25, 10 September 2025 (UTC)

No. Simonm223 (talk) 11:38, 10 September 2025 (UTC)
Why? Queens Historian (talk) 16:48, 10 September 2025 (UTC)
Please see the discussions that led to it being declared a deprecated perennial source. Simonm223 (talk) 18:15, 10 September 2025 (UTC)
Specifically, there was a discussion about the NYP in 2020, and editors came to a consensus that it is generally unreliable as a source about politics, especially about New York City politics. nomen alternativum (he/him • talk • contribs) 18:20, 10 September 2025 (UTC)
Thanks! I wasn't aware that there was a discussion in 2020 on this matter that resulted in this consensus. However, I dispute the policy in which "Even in cases where the source may be valid, it is usually better to find a more reliable source instead." That's simply discrediting the messenger even when the message is entirely accurate. I'd like this policy to change, so that New York Post can be used a source when it is accurate. Queens Historian (talk) 11:37, 11 September 2025 (UTC)
Well good luck with that but I would not support such a policy change. We are strict with unreliable sources with good reason. Simonm223 (talk) 11:41, 11 September 2025 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Reflist is broken?

Hey, if I go to Wikipedia:Reliable sources/Perennial sources#References, the reflist is broken. All I see is:

Template:Reflist

And nothing else. Not sure why. Endwise (talk) 09:21, 14 September 2025 (UTC)

This is likely caused by hitting the template limit, something that's been discussed on the lists talk page before. Replacing {{reflist}} with <references/> allows some of the references to be seen but some are broken. The only long-term solution is to reduce the amount of templates in the article. -- LCU ActivelyDisinterested «@» °∆t° 10:21, 14 September 2025 (UTC)
Yes, the page is in the hidden Category:Pages where post-expand include size is exceeded. I don't have time at the moment to investigate. Someone at WP:VPT might be able to help. A lot of content is included, for example with {{WP:Reliable sources/Perennial sources/1}}. An ugly fix might be get rid of those templates and put the content directly in the page. Johnuniq (talk) 11:02, 14 September 2025 (UTC)
The transclusions are the result of the last attempt to fix the same issue. Using the reference tag rather than the reflist template shows that even #invoked cites aren't displayed correctly. -- LCU ActivelyDisinterested «@» °∆t° 19:26, 14 September 2025 (UTC)
Transcluding was actually the opposite: It pushed us up against the PEIS limit and was done due to complaints of the editor taking a long time to load and browsers crashing when attempting to find the part to edit. Aaron Liu (talk) 19:51, 14 September 2025 (UTC)
I'll try pruning the edit filter links and then the unused shortcuts. Aaron Liu (talk) 19:49, 14 September 2025 (UTC)

Request to revise Army Recognition entry – defamatory & unsourced

Moved to Wikipedia:Reliable sources/Noticeboard § Request to revise Army Recognition entry – defamatory & unsourced
 – Aaron Liu (talk) 18:32, 17 September 2025 (UTC)

Blanket suppression of stale

@AKK-700 The instructions say considered generally unreliable for being self-published or presenting user-generated content are excluded, not automatically all GUnRel sources. Regardless, please WP:BRD. (Sorry if this is WP:BITEY, I just don't have much time to compose a message right now.) Aaron Liu (talk) 02:12, 19 September 2025 (UTC)

It doesn't cover biased, fake news, opinion, or satire news websites, defunct websites, sites that copy content from other sources, blacklisted sites, or articles from websites before their reliability changed (like Newsweek pre-2013). AKK700 02:52, 19 September 2025 (UTC)
Is there consensus for any of these options except "from before their reliability changed" anywhere? Aaron Liu (talk) 12:04, 19 September 2025 (UTC)
I understand that it is current practice that sources unlikely to change status (say, due to their abysmal reputation) get this suppressed. But that doesn't seem to be the case with most of the sources you suppressed stale for. The Western Journal is also none of the above. @AKK-700 I'll be reverting this tomorrow while we discuss. Aaron Liu (talk) 00:10, 20 September 2025 (UTC)

CBR

How can I add Comic Book Resources to the list? ~Rafael (He, him) • talk • guestbook • projects 02:38, 20 September 2025 (UTC)

Please see Wikipedia:Reliable sources/Perennial sources#What this page is. This is not supposed to (and due to technical limits, should not) be a list of every source but rather just the most frequently discussed. You should seek other Category:WikiProject lists of reliable sources to add sources to. Aaron Liu (talk) 04:01, 20 September 2025 (UTC)
Does it meet WP:RSPCRITERIA? Per Wikipedia:Reliable_sources/Noticeboard/Archive_487#Reliabilty_of_Comic_Book_Resources_(CBR) there is some previous discussion, but not very much. Fwiw, it's on WP:GAMESOURCES. Gråbergs Gråa Sång (talk) 07:57, 20 September 2025 (UTC)
I replied on Rafael's talk page – per WP:VALNET, it's reliable up to 2016. ClaudineChionh (she/her · talk · email · global) 09:56, 20 September 2025 (UTC)

YouTube

The entry for YouTube blends an exception for reliable content into the "generally unreliable" rating:

  • "Content uploaded from a verified official account, such as that of a news organization, may be treated as originating from the uploader and therefore inheriting their level of reliability."

That should be listed separately with a big "Exception" heading and different color. -- Valjean (talk) (PING me) 14:37, 21 September 2025 (UTC)

@Valjean See the Wikipedia talk:Reliable sources/Perennial sources/Archive 12#YouTube is not a source and Wikipedia talk:Reliable sources/Perennial sources#Drafting the RFC question about platforms sections above. Aaron Liu (talk) 16:22, 21 September 2025 (UTC)
That's an awful lot of discussion. Does it disagree with the current wording?
  • "Content uploaded from a verified official account, such as that of a news organization, may be treated as originating from the uploader and therefore inheriting their level of reliability."
If not, then I don't see any problem with just splitting the YouTube portion into a "generally unreliable" pink box and a box that cites ""Content uploaded from a verified official account, such as that of a news organization, may be treated as originating from the uploader and therefore inheriting their level of reliability." That portion should not be buried within "generally unreliable". That's where all these are buried, without naming them, because they all have official "verified" YouTube accounts: Reuters, AP, BBC, AFP, ABC News, NBC News, CBS News, USA Today, Bloomberg News, etc. -- Valjean (talk) (PING me) 19:57, 21 September 2025 (UTC)
It does disagree. The plan is to propose making all platforms their own category as the issue also extends to e.g. Wordpress. Aaron Liu (talk) 20:22, 21 September 2025 (UTC)
In our ample free time... I apologize for not doing my part to keep this moving forward. I agree that it's a problem that we need to address.
@Valjean, I think the view of YouTube is kind of three things:
  • Home videos, someone's favorite 10-second clip from a movie, ordinary-user-type content: Mostly don't use it/approximately the same rules as someone's personal blog (including same rules for personal blogs that violate copyrights).
  • YouTube's own 'infrastructure', e.g., "Had 30,000 likes as of 2024": Don't use it/we don't like it and think it could be faked.
  • Official verified accounts of reliable sources: It's just fine. We actually don't care where Reuters, AP, BBC, etc. post their content; that content's reliable because they're Reuters, AP, BBC, etc. and not because of the domain name shown in your browser.
The first and last are the 'platform' ones. The middle one is a WP:PRIMARY source that would be theoretically okay for WP:V purposes except we (apparently) reject it anyway (think it WP:UNDUE by default?). WhatamIdoing (talk) 23:49, 21 September 2025 (UTC)

Summary for The Washington Free Beacon

Can someone explain why the discussions all say it is an unreliable, hyperpartisan newsblog, but the summary says "Most editors considered the Washington Free Beacon to have become generally reliable during the editorship of Eliana Johnson." Is there a missing discussion that has not been included, or did someone just sneakily change the summary without anyone noticing? ເສລີພາບ (talk) 19:32, 5 September 2025 (UTC)

I found the missing discussion here, although it does not to me look like a consensus that they are generally reliable, but no consensus. Regardless, this discussion should be linked to on this page. ເສລີພາບ (talk) 20:27, 5 September 2025 (UTC)
It was updated[1]. There are now two entries for the Free Beacon, which matches the consensus from the 2025 RFC (that is also already linked in both entries). -- LCU ActivelyDisinterested «@» °∆t° 22:13, 5 September 2025 (UTC)
The link takes me to Wikipedia:Reliable_sources/Noticeboard#RFC_on_the_reliability_of_the_Washington_Free_Beacon which doesn't exist. The actual location is Wikipedia:Reliable_sources/Noticeboard/Archive_478#RFC_on_the_reliability_of_the_Washington_Free_Beacon. Now that I'm on my computer though I'm seeing a box showing up redirecting me to the archive. ເສລີພາບ (talk) 01:52, 6 September 2025 (UTC)
The link has been updated to show it's been archived, you can fix that the page isn't protected. As with any other page on Wikipedia if it's broken the best solution is to WP:JUSTFIXIT. -- LCU ActivelyDisinterested «@» °∆t° 02:59, 6 September 2025 (UTC)
Question #2 seems like consensus. Are you sure you're looking at that instead of Q1 on its status before Johnson? Aaron Liu (talk) 17:37, 6 September 2025 (UTC)

WP warns against making defamatory remarks against living persons; yet here Matthew Continetti is characterized as unreliable based on collective editorial assessments of the website's practices and reputation during has editorship. Matthew Continetti is widely recognized as the founding editor and was editor-in-chief of the Free Beacon from its launch in 2012 until 2019. He is also known as a journalist and Director of Domestic Policy Studies at the American Enterprise Institute, with bylines in many reputable publications and authorship of books on conservatism and politics.73.157.112.234 (talk) 08:40, 30 September 2025 (UTC)

There are now 2 Template:Press on this talkpage

That's not a mistake, they have a limit of 30 items. Gråbergs Gråa Sång (talk) 07:52, 1 October 2025 (UTC)

Southern Policy Law Center reliable?

Off-topic (wrong place for this discussion). Feel free to copy the contents of this section to WP:Reliable sources/Noticeboard. (non-admin closure) Aaron Liu (talk) 01:23, 7 October 2025 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

How is an organization that has had to pay out millions in a defamation lawsuit, and has been forced to retract and issue apologies for some of its claims considered "generally reliable? Fenngibbon (talk) 10:45, 4 October 2025 (UTC)

The US has a suing culture that's highly politicized. Read the actual entry WP:SPLC
The Southern Poverty Law Center is considered generally reliable on topics related to hate groups and extremism in the United States. [Which means it is not generally reliable on other topics, like that of British hate groups, like in the case of Maajid Nawaz where it settled for 3.4M and issued an apology.] As an advocacy group, the SPLC is a biased and opinionated source. The organization's views, especially when labeling hate groups, should be attributed per WP:RSOPINION. Take care to ensure that content from the SPLC constitutes due weight in the article and conforms to the biographies of living persons policy. Some editors have questioned the reliability of the SPLC on non-United States topics. SPLC classifications should not automatically be included in the lead section of the article about the group which received the classification. The decision to include should rather be decided on a case-by-case basis.
Emphasis mine. Headbomb {t · c · p · b} 11:33, 4 October 2025 (UTC)
The SPLC had to pay millions to Maajid Nawaz a Muslim anti-extremist activist because they accused him of being an "anti-Islam extremist."
That they had to pay out that claim is not because they are "generally reliable on topics related to hate groups and extremism in the United States" but only in the United State; it is because they are not generally reliable at all, having a long history of labeling as "hate" and "extremism" not just actual hate or extremism, but any point of view contrary to the predilections of the leaders of the SPLC. Fenngibbon (talk) 17:41, 4 October 2025 (UTC)
The bulk of what SPLC claims as a hate group are generally regarded as hate groups by academics and politicians alike. But they are also aggressive (in the same way the ADL is) to overclassify certain views as hate. This might account for maybe 5%, 10% at most of what SPLC outputs; the rest of the time, no one (save for the group in question) really questions their classification. For that reason, we take what the SPLC says with a grain of doubt, and as noted above, classify them as reliable for their opinion, meaning it must be attributed. If we run into a situation where multiple sources disagree with the SPLC, like with Nawaz, we can document that controversy just like as has been done at Nawaz. Masem (t) 17:48, 4 October 2025 (UTC)
Probably the right forum for this is WP:RSN. Maybe we need a stronger worded caveat about the non-US topics. Alaexis¿question? 19:55, 4 October 2025 (UTC)
It's disputable even on a number of the US topics. Fenngibbon (talk) 02:15, 5 October 2025 (UTC)
It's not a credible organization. If wikipedia wishes to demonstrate some good faith in its own credibililty, SPLC should be removed as a reliable source. It's an extremist organization that attempts to label its opponents to discredit them. The labeling has already directly led to one attack on a religious organization. The FBI has dropped it as a reference guide.
https://www.allsides.com/news-source/southern-poverty-law-center-media-bias PerseusMeredith (talk) 19:58, 4 October 2025 (UTC)
"If wikipedia wishes to demonstrate some good faith in its own credibililty"
This isn't how it works. Headbomb {t · c · p · b} 02:40, 5 October 2025 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Wikipedia:Reliable sources/Perennial sources/6 includes a link to this discussion in support of the determination on PBS (as PBS #6). This looks like a mistake to me. The linked discussion is about Voice of America explicitly; in fact, it is used as discussion #6 (correctly) for RSPVOA. A few – maybe three – editors brought up PBS only as a comparison, not as an endorsement or determination.

Why this matters: The next-newest discussion about PBS that I can find is from 2016. This determination should definitely be noted as stale. ☆ Bri (talk) 21:23, 7 October 2025 (UTC)

I agree on this one. Discussions about other source can have three different editors start evaluating another source but that indeed doesn't seem like what's happening here. Aaron Liu (talk) 00:40, 8 October 2025 (UTC)

Questionable date on Haaretz discussion

The footnoted list of Haaretz at Wikipedia:Reliable sources/Perennial sources/4 has this discussion as #10 which was closed with this comment: The article is by all indications an opinion piece...it has no bearing on Haaretz's general reliability of coverage. (emphasis mine). I don't think this discussion should be used to pin the date for this entry in the list.

The next-newest discussion appears to be from 2018, which would make this determination very stale. ☆ Bri (talk) 21:40, 7 October 2025 (UTC)

The participants seem to affirm that it is a reliable source. Furthermore, the "RS status not in question, and yet..." section addresses the source in general in detail. Aaron Liu (talk) 00:33, 8 October 2025 (UTC)
Thanks for reply. I'll still mark it as stale since the discussion was closed over four years ago, in August 2021. ☆ Bri (talk) 16:04, 8 October 2025 (UTC)

Semi-automating template RSPLAST

Currently, the date value passed to template WP:RSPLAST is hand-curated, and subject to going out of date, or to data entry error. This could be partly automated, based on a proposed module that would return the timestamp of the most recent comment in a set of given discussions. To my knowledge, there is no module that currently does this, but archiving bots already do something very like this; for example, Lowercase sigmabot calculates max(stamps) in function parse_stamps. If such a module is forthcoming, we could replace the current invocations of WP:RSPLAST with a new template which would return the latest date given the list of discussions already present in the list column. It still depends on that list being kept up to date manually, which is why it is semi- and not completely automated, but it will still save work going forward, and minimize errors. P.S., If we invoked the module directly instead of through a wrapper template, that would save us several hundred template expansions in the table, and thus gain us a little breathing room regarding PEIS capacity, as discussed § above. Mathglot (talk) 01:50, 23 September 2025 (UTC)

@DLynch (WMF), is there an API or something that would let a script/bot grab the date (at least year) of the most recent comment in a discussion? The input would be a link to, e.g., Wikipedia:Reliable sources/Noticeboard/Archive 371#Deutsche Welle, and the thing we want back is the thing that produces "Latest comment: 3 years ago" in DiscussionTools. WhatamIdoing (talk) 00:21, 28 September 2025 (UTC)
Ideally, something usable via template (i.e., most likely a module, rather than a user-script which is not template-accessible). The idea is not to limit this to conversion, for which a bot or script would be fine, but to become an integral part of the table (or whatever approach is chosen) so that you have a new template that does {{latest comment|discussion-1|disc-2|disc-3}} and it returns the date (or year) of the most recent comment in the given discussions. (The output of that could be used as input to a staleness-calculator that would remain up-to-date live, as opposed to the current implementation as the very rough, < 4 years or > 4 years binary measure, which is entered manually on the page and becomes inaccurate with the passage of time.) Mathglot (talk) 00:40, 28 September 2025 (UTC)
Not trivially right now -- though there's a patch that'll be going out on WP:ITSTHURSDAY which will add it to the headings in the DiscussionToolsPageInfo threaditemshtml API output. DLynch (WMF) (talk) 15:44, 29 September 2025 (UTC)
(The patch just gives you some convenient summaries in the API response. You could right now compute all that yourself from it, it'd just be more work.) DLynch (WMF) (talk) 15:45, 29 September 2025 (UTC)
Thanks! WhatamIdoing (talk) 18:21, 29 September 2025 (UTC)

In case it's not evident why having this feature would be a good thing, note that six of these 8 contributions, plus these two are exclusively about changing the "staleness" value in five different SRP numbered subpages. We should not have to edit these files to accomplish this; the staleness value should be rendered automatically based on a date calculation. Mathglot (talk) 05:28, 8 October 2025 (UTC)

I agree. I also believe that it would be better to collect the necessary ~500 dates via script than by hand. WhatamIdoing (talk) 05:36, 8 October 2025 (UTC)
Script not needed; I just threw a regex together, so no promises (there might be conflation of two rows, missing data, etc.) and too tired to check it now, but this is at least a start, even if there are bugs. See Wikipedia:Reliable sources/Perennial sources/1/Last. But having the dates doesn't help much by itself. Mathglot (talk) 06:07, 8 October 2025 (UTC)
With the date, the calculations are easy. I mean, even I could figure out how to make that parser function work, with no more than one trip to VPT. ;-) WhatamIdoing (talk) 16:56, 8 October 2025 (UTC)
Yes, the date calculations are easy once you have that date, e.g., as in the parameter in {{WP:RSPLAST|2021}}, in the row. The point is, how did '2021' get in there? A: somebody added it by hand. What I meant above is, having the dates isolated in a key=value list like this doesn't help, because it is still the result of extracting the manually entered date from the row, and manual date entry is what we are trying to automate. All occurrences of the template should have the date parameter removed, and the invocation should instead be {{WP:RSPLAST|discussion-list}}, pointing to (or listing) the discussions already available in the row, and the last date should be derived from them, not by manually shuffling through twenty discussions and trying to pick out the most recent date among all of their comments. Do editors who update existing rows by adding a new discussion to the list for a source always know enough/remember to update LAST as well? With this change, they would not have to. Mathglot (talk) 17:54, 8 October 2025 (UTC)

Are you sure?

Moved from User talk:Aaron Liu/Archive 5 § Are you sure?
 – Aaron Liu (talk) 22:17, 8 October 2025 (UTC)

I'm not sure that the logic [2] change to WP:RSPLAST is correct, it seems to me you'd have to have month, year, and day all supplied as parameters to do it right. But it's your call. ☆ Bri (talk) 22:11, 8 October 2025 (UTC)

2025-3=2022, so anything less than 2022 would mark it as stale. Sure, there's the edge case of a discussion from November 2021, but I'm not really concerned about the practical implications of that and am more concerned about the "stale=y" making it harder for dumber editors to update entries. Aaron Liu (talk) 22:17, 8 October 2025 (UTC)

Drafting the RFC question about platforms

Here's a starting point for the RFC:

Some websites listed in Wikipedia:Reliable sources/Perennial sources are not WP:SOURCES, but are instead digital platforms that host both reliable and unreliable sources. Examples of these platforms include YouTube, which hosts reliable TV news shows (example) as well as many unreliable home videos, and Flickr, which hosts reliable photos from government agencies such as NASA (example) as well as many unreliable personal photos.

Proposal: Shall we expand Wikipedia:Reliable sources/Perennial sources#Legend to include a new category, "platform", with a description that says the website itself does not determine whether the source is reliable, and instead editors must look at the publisher. For example, on YouTube, editors must base their decision about reliability on the uploader or user. For example,; https://www.youtube.com/@BBCNews is (the verified official account) is reliable because https://www.bbc.com/news is reliable.

Note:

  • Some websites will need to have two entries. For example, YouTube is only a platform for news videos, but editors discourage citing it as a primary source for views and subscriber counts, so it could have one entry for "platform" and another saying "generally unreliable" for views and subscriber counts.
  • Which items would get placed in this category is subject to case-by-case editorial consensus. Initially, _______ would get re-classified as platforms.

What should we change first? Are the examples good? What websites (if any) should we put in the blank? WhatamIdoing (talk) 16:34, 19 August 2025 (UTC)

I think we should just merge the entries into a category first (c.f. Wikipedia talk:Reliable sources/Perennial sources/Archive 10#Merging some entries), giving it the status-quo GUnRel status, before starting an RfC on the status of that category. Aaron Liu (talk) 16:49, 19 August 2025 (UTC)
The problem there would be something like WP:ACADREP where the hosting site is already marked as MRel. Moving those to GUnRel before the RFC could muddy the waters. -- LCU ActivelyDisinterested «@» °∆t° 22:25, 19 August 2025 (UTC)
The description at ACADREP is pretty close to what we're saying about platforms: don't look at the hosting site; look at the real source. WhatamIdoing (talk) 19:25, 20 August 2025 (UTC)
Ah, I should've realized ACADREP was essentially the same thing. I agree with Void then that focusing on the concept is probably the way to go before turning the RSP stuff into a category. If the PaG thing passes, we should be able to Boldly add a MRel category. Aaron Liu (talk) 19:29, 20 August 2025 (UTC)
That looks like a good RFC question, its a bit on the long side but I think it needs to be to illustrate the issue. Personally the only clear area I see for improvement would be some sort of footnote after "the website itself does not determine whether the source is reliable" explaining that these platforms use proprietary content standards very different from that of publishers. Horse Eye's Back (talk) 16:54, 19 August 2025 (UTC)
I'm not sure what you mean by "proprietary content standards very different from that of publishers". Does that mean that YouTube lets publishers post almost anything (that's legal), where as some publishers (e.g., the BBC) have higher standards and other publishers (e.g., a vanity press) do not? WhatamIdoing (talk) 20:08, 19 August 2025 (UTC)
IMO it means that Youtube hosts what it wants to host (note that they choose not to host a lot of legal content), I believe that even vanity presses generally have more control over and liability for the material than a hosting platform does. Its not key to the RfC though, it can be a discussion that happens later if this proposal goes through. Horse Eye's Back (talk) 16:00, 20 August 2025 (UTC)
Youtube hosts what it wants to host, but bbc.com/news does not?
In the US, a vanity press can refuse to print anything they choose to refuse, because Freedom of the press belongs to those who own one. (Under US law, they actually shouldn't have liability except in specified circumstances such as child porn, as they aren't the publisher – legally, it'd be like blaming the manufacturer of a Photocopier if someone makes photocopies of something confidential.) But the same's true for YouTube and the BBC, so I'm not sure that it's relevant. The relevant point is that the BBC has higher standards than any vanity press. WhatamIdoing (talk) 19:25, 20 August 2025 (UTC)
Vanity presses run a gamut, in some cases they are the publisher and in others it really is much more akin to the photocopier analogy. There certainly are aspects in which these platforms and vanity presses are similar though, especially with promoted posts. I agree that the relevant point is that the BBC has higher standards than any vanity press, but I think its also relevant that they have a fundamentally different mentality/tradition. Amazon also seems to fit into a few different boxes here. Horse Eye's Back (talk) 19:48, 20 August 2025 (UTC)
I think that's a good RFC question. ~ ONUnicorn(Talk|Contribs)problem solving 17:07, 19 August 2025 (UTC)
This seems good assuming we start after "Proposal:". The stuff before seems to assume the conclusion in ways I'm worried would violate WP:RFCNEUTRAL.
I also think we ideally shouldn't repeat "For example" twice. Loki (talk) 19:18, 19 August 2025 (UTC)
Agree with this. First paragraph should be part of a !vote, or otherwise adding "Editors have argued" at the start of the paragraph and linking to this discussion in order to include it. Best not to draw any conclusions into it. CNC (talk) 19:40, 19 August 2025 (UTC)
The problem with omitting the explanation is it leaves us with a proposal to create a category called 'foo', with no explanation of what's intended to be in that category. I would expect that to produce confusion.
"Editors have argued" sounds like weasel words, and besides, I don't think that anybody is seriously arguing the opposite (e.g., that the reliability of anything on YouTube can be determined merely by saying 'well, it's on YouTube, so it is/isn't reliable'). WhatamIdoing (talk) 20:11, 19 August 2025 (UTC)
I agree with this proposal in principle but I think it should be proposed as an addition to WP:SOURCES first, and that perennial sources should follow suit. Void if removed (talk) 20:31, 19 August 2025 (UTC)
I'm not sure how that would fit into WP:SOURCES: "When editors discuss sources...they are usually talking about the work itself, the creator of the work, the publication, or the publisher of the work, but we frankly don't care which website the work has been posted to; a book on Google Books or on Perlego or on Amazon is still the same book"? WhatamIdoing (talk) 20:38, 19 August 2025 (UTC)
I'd suggest pretty much your proposed text be added as a subsection titled "Platforms", alongside WP:NEWSBLOGS, with a WP:PLATFORMS shortcut. Eg.
Some websites are digital platforms that host both reliable and unreliable sources. Examples of these platforms include YouTube, which hosts reliable TV news shows (example) as well as many unreliable home videos, and Flickr, which hosts reliable photos from government agencies such as NASA (example) as well as many unreliable personal photos. In these cases, it is the reliability of the publisher which should be assessed, rather than that of the platform itself. Where no reliable publisher can be identified, such sources should be considered self-published.
If there's consensus for that then Perennial sources should follow suit. I'd suggest doing it this way round because IMO it is more clearly a policy amendment/interpretation that should cascade out from WP:SOURCES. Void if removed (talk) 20:49, 19 August 2025 (UTC)
I think I agree with Void, or at least I agree that we should do both of these at the same time.
I'd also maybe include a line saying something along the lines of Information from the platform owner itself (including metadata like view counts as well as announcements directly from the platform) is considered to be published by the platform owner. In general this information should also be considered self-published unless it's reviewed by independent sources. Loki (talk) 21:01, 19 August 2025 (UTC)
I think that both of these ideas are feasible.
That means:
Does that sound about right? WhatamIdoing (talk) 19:31, 20 August 2025 (UTC)
Yes! Loki (talk) 21:29, 20 August 2025 (UTC)
Support this. My main opposition previously was the lack of "doing things properly", whereby this is very much a thorough approach, along with UGC referenced below by Sunrise, so there would be no contradictions or varying interpretations of policy leftover. CNC (talk) 09:59, 21 August 2025 (UTC)
I also think this is a better approach. My feedback is:
  • I would add Wikipedia:Reliable Sources#User-generated sources as well, given that it describes all such sources as "generally unacceptable". The list of examples may need to be updated, e.g. it seems that an organization's official Facebook page would be covered by this change as well.
  • I would emphasize that there must be some confirmation of the publisher's identity (e.g. uploads by other users are likely to be copyright violations). This could be an expansion or follow-up to Void's text of Where no reliable publisher can be identified, such sources should be considered self-published.
  • Since this is a broad change to policy, editors from the relevant pages should be invited to comment once text for all the changes has been proposed.
--Sunrise (talk) 23:27, 20 August 2025 (UTC)
  • I regret to say that I agree with you about needing to restructure Wikipedia:Reliable sources#User-generated content. I'm currently thinking that we need to separate collaborative content (e.g., Wikia's Fandoms, Reddit discussions) from individual user posts (e.g., a teenage boy posting videos of himself and his friends on skateboards). Collaborative content is user-generated, and the rest is perhaps better filed under this 'platforms' concept. (But where to put a multi-person Twitter thread [as opposed to an individual tweet]]? Obviously I haven't thought about this enough yet.)
  • I'm not sure that Void's text is appropriate. Unreliable publishers (e.g., National Enquirer) are not self-published, so they should not be considered self-published. They should be considered unreliable (in the case of the National Enquirer, due to a reputation for the opposite of fact-checking and accuracy). Also, you should be able to determine the publisher on most of these platforms: it's the username of the uploader. The publisher will very frequently be engaged in self-publishing (e.g., that teenage boy on his skateboard) or potential copyvios (e.g., video clips), but IMO the policies and guidelines should normally consider these unreliable (meaning that they have one or more of a long list of disqualifying problems) rather than specifically self-published.
  • As I said above, I expect this to be a very widely advertised RFC.
WhatamIdoing (talk) 00:09, 21 August 2025 (UTC)
On point 2, that's a good point that I overlooked when quoting that. The wording would need to be adjusted - I think the key idea is to clarify that a self-published source does not lose that status simply by being published on one of these platforms. On point 3, to be clear, I was recommending getting input from those pages before actually starting the RfC, since RSP is rather specialized in comparison. Sunrise (talk) 02:11, 21 August 2025 (UTC)
Half of the usual suspects are already here, but I've no objection to anyone posting notices.
The thing on my mind right now is that this really is a technical adjustment (nobody who understands the problem/proposal thinks we are proposing new rules), but with every additional "Oh, and this other section", it risks sounding so complicated that we may get fear-driven opposition ("I can't support anything unless you show me exact wording changes" – followed inevitably by "Thanks for the exact wording changes, but I can't support this because it's too complicated to change so many separate pages at the same time"). Maybe we should consider an explicitly "in principle" proposal. WhatamIdoing (talk) 02:33, 21 August 2025 (UTC)
In principle is fine by me and arguably simpler, as the exact wording (based explicitly on the interpretation of the consensus) can be devolved to local consensus. For here, it'd be no different than summarising any other consensus from RSN for example. Being thorough does not mean everything has to be complete(d). CNC (talk) 10:11, 21 August 2025 (UTC)
Yes I agree, I've struck "reliable", that wasn't my intent. My intention was that if there's no obvious publisher, it is SPS. Void if removed (talk) 07:52, 21 August 2025 (UTC)
@WhatamIdoing Per advertising the RfC (it's come up enough times including from me), I'm happy to notify relevant talk pages, purely so there is assurance that it will happen. Ping me and I'll do it, out of respect for the effort that has gone into developing this RfC. (It'd be lazy for me to push for an RfC and then be unwilling to notify editors). CNC (talk) 10:07, 21 August 2025 (UTC)
Yes, it would have to be clear that by "publisher" we're talking about what on YouTube would be the officially identified channel of the publisher. So if it is on the BBC channel, the BBC are the publishers. If it is a clip from a BBC programme on some random user's channel, it isn't, and is self-published and probably a copyright violation. Void if removed (talk) 10:51, 21 August 2025 (UTC)
The proposal will ultimately impact many different hosting sites. Maybe there should be less focus on specific sites, and more on the concept. -- LCU ActivelyDisinterested «@» °∆t° 22:27, 19 August 2025 (UTC)

This is very good. I've slightly rewritten for brevity and clarity, but I have no intent to change the meaning.—S Marshall T/C 08:23, 17 September 2025 (UTC)

Thanks for this. I've been thinking about expanding the first sentence: "are not individual WP:SOURCES, but digital platforms that host both reliable and unreliable sources".
I've also been thinking about the "in principle" idea, and wondering if we ought to say something like "A digital platform is a website like YouTube that hosts content published by both traditional, reliable publishers and ordinary people. For example, YouTube hosts videos from traditional television news shows (which are reliable sources) and home videos of teenagers showing off their skateboarding tricks (which is not). In principle, should we differentiate between ordinary self-published and user-generated content (e.g., a Wikia Fandom site) and a digital platform (e.g., YouTube, which is used by traditional news outlets as well as self-publishing individuals)?"
This might be too abstract for some people. WhatamIdoing (talk) 18:09, 17 September 2025 (UTC)
Also, I don't think we should use Twitter/X as an example, because my impression is that most of our uses are for primary sources (e.g., Joe Film announced that he's engaged to be married), and the "but a tweet is not reliable SIGCOV IRS!!!1!" could be a distraction. WhatamIdoing (talk) 19:05, 17 September 2025 (UTC)
Wordpress or Blogspot might be a good one; I remember participating in an AfD a few months back where we were deciding if a post by a history centre/museum/archival-type-project associated with a local university was reliable for a local history matter. The fact that they hosted part of their site on Blogspot was a deciding factor for one participant. GreenLipstickLesbian💌🦋 19:23, 17 September 2025 (UTC)
I like that idea. Do we have a couple of good examples of that? WhatamIdoing (talk) 00:51, 18 September 2025 (UTC)

Any follow-ups? I hope this proposal can come to fruition soon. I plan to start a similar thread on zhwiki RSN as well, but I'm hoping to see the "platform" rating implemented first on enwiki, to establish a strong precedent for other wikis to follow. SuperGrey (talk) 21:51, 7 October 2025 (UTC)

I've been thinking that we should postpone this until the WP:PEIS problem with Wikipedia:Reliable sources/Perennial sources has been addressed. WhatamIdoing (talk) 19:58, 8 October 2025 (UTC)
You are right, the PEIS problem is more imminent. SuperGrey (talk) 20:21, 8 October 2025 (UTC)
To chime in belatedly, I don't feel vested in one idea versus another, but I agree strogly with the notion that we need to take steps against the frequent confusion of "source" with "platform/venue/hosting provider/conduit/access-method", not to mention "medium/format". What actually matters is the nature of the content and the authoritativeness [on that specific subject] of the person(s) who created it.  — SMcCandlish ☏ ¢ 😼  21:59, 13 October 2025 (UTC)

Microsoft Edge user here ( 141.0.3537.71 )

Every time I visit this project page (not talk page which I'm visiting), I cannot keep reading after 30 seconds because RAM is used to display a long table on that page is over 10GB. So, the browser (Microsoft Edge 141.0.3537.71) displays a message saying the page cannot be displayed and asks you to refresh the page. --Chukje (talk) 11:38, 14 October 2025 (UTC)

Edit: after 30 seconds -> after 20 seconds Chukje (talk) 11:40, 14 October 2025 (UTC)
Update: I turned off "Urban VPN" extension app and this problem is fixed Chukje (talk) 11:54, 14 October 2025 (UTC)
Yeah, I don't think that's a problem with this page. It's not anywhere close to taking up 10 GB in memory. Aaron Liu (talk) 12:46, 14 October 2025 (UTC)

CSV/PDF file is best alternative

Can anyone make CSV/PDF file of this source list? --Chukje (talk) 11:49, 14 October 2025 (UTC)

Yes, but we have some important tasks to deal with at the moment. Please ask again in a few weeks or so. In the meantime, I believe there is a Print page feature in most browsers, and if you do Print to PDF on your machine, then you will at least have it in that format. Mathglot (talk) 21:52, 14 October 2025 (UTC)

CBS News

Moved to Wikipedia:Reliable sources/Noticeboard § CBS News
 – Aaron Liu (talk) 18:34, 16 October 2025 (UTC)

Mobile testing for RSPGIANT

Marcocapelle, Laterthanyouthink, Johnpacklambert, Mediocre Legacy, Bluethricecreamman, Sbaio, anyone else who edits from the mobile site:

Would you please make a test edit to the huge table in WP:RSPGIANT, and tell us whether that's feasible for mobile editors? (It's a copy, so don't worry. Just do any basic change, as if you were fixing a typo or adding a simple link.)

It'll be useful to know whether you're using the mobile app or the mobile website. I don't expect it to be pleasant, but it'd be helpful to know whether it's actually impossible (e.g., timed out before it could load the whole page?), and also how you would rate the experience on a scale that runs from, say, "painful" to "manageable". And, I suppose: Can you even read that table on mobile? WhatamIdoing (talk) 06:20, 16 October 2025 (UTC)

@WhatamIdoing: done. It is slow, as expected, but it works. It is readable too, of course both horizontal and vertical scrolling is needed. Marcocapelle (talk) 06:27, 16 October 2025 (UTC)
@Marcocapelle (or anyone else), from the mobile POV, would you prefer a page like /Deutsche Welle or /The Daily Telegraph be better? (Those links show two different styles.) WhatamIdoing (talk) 22:57, 16 October 2025 (UTC)
Unfortunately WP:RSPGIANT has "Post‐expand include size: 1978935/2097152 bytes" which leaves very little expansion room. Some of the helper templates such as WP:RSPSHORTCUT and WP:RSPLAST are used hundreds of times and each occurrence includes a large number of bytes. Tweaking them and dropping some of the functionality would give more breathing room. Johnuniq (talk) 06:56, 16 October 2025 (UTC)
Yes, but it's a test page. We're actually trying to figure out how big we can make it without breaking t. WhatamIdoing (talk) 08:10, 16 October 2025 (UTC)
@Johnuniq That's only because Mathglot copied and pasted the table a few times to see how much space is remaining. IIRC the current estimate is over 600 new rows can be added. (1978935 is after those 600 new rows were added.) Aaron Liu (talk) 22:49, 16 October 2025 (UTC)
At the moment, RSPGIANT might let us add 600 new rows, the row-builder module might let us add about 200 new rows,[3] and the subpages approach would allow thousands (though they wouldn't be rows any longer). WhatamIdoing (talk) 22:54, 16 October 2025 (UTC)
I think WP:RSPGIANT might let us add about 300 new rows. See this edit and this one, which added 150 rows each, for a total of 300 rows added, 800 total. The '600' was probably quoted from an earlier comment by me, and was a mistake. Mathglot (talk) 10:01, 18 October 2025 (UTC)

Middle East Eye

There have been numerous discussions of MEE on RSN. Would anyone care to summarise them on the RSP list? BobFromBrockley (talk) 19:49, 19 October 2025 (UTC)

Last time I checked, there weren't enough RSN discussions about the source as a whole, instead of e.g. a specific opinion article. Aaron Liu (talk) 19:52, 19 October 2025 (UTC)
There are 20 mentions of Middle East Eye, and of those, I count four discussions that might qualify as significantly about them (1, 2, 3, 4). Mathglot (talk) 20:52, 19 October 2025 (UTC)
1 and 4 do not discuss the reliability in general (4 is the "a specific opinion article" one I mentioned), 2 says Grel but is questionable as the prevailing argument was Middle East Eye, here, is used as a source for an interview with the victim's mother. There isnt any indication that they did not faithfully and accurately report what her mother said. (though the source was evaluated), and while 3 counts I'm unsure what consensus the disucssion has; numerically it's a no-consensus but I'm unsure if the "unreliable"/"attribution" camp was sufficiently grounded in policy.
IAR I would only add this if another significant discussion naturally occurred. Aaron Liu (talk) 23:58, 19 October 2025 (UTC)

Encyclopedia.com

Can Encyclopedia.com be placed on the list given the closure of the recent RFC? ―Howard • 🌽33 15:21, 5 October 2025 (UTC)

I think we should, but I don't know if it would cause this page to crash at this time. Gråbergs Gråa Sång (talk) 15:27, 5 October 2025 (UTC)
We could always draft the content as a note here, and then add it when the page is fixed up. WhatamIdoing (talk) 02:28, 6 October 2025 (UTC)
A halfway-house solution might be the following: go ahead and write it up, and include it in the row in the */3 file, where sources beginning with 'E' are stored. Place it in between <noinclude>...</noinclude> tags, and point the shortcut at its row in the */3 file, *not* the main file. This way, it will not blow up the main page when it is transcluded there, and people using the shortcut can still directly view it, in the */3 file. The cost: it won't be seen if you browse or sort the main table; but it also won't contribute to crashing it. Then, when the problem with the main page is fixed, just remove the <noinclude> tags. Mathglot (talk) 05:58, 7 October 2025 (UTC)
Oh, looks like someone already added it. We are now back up to 99.76% of capacity, and 4,959 PEIS bytes short of full. Do we want to add an WP:Edit notice to the eight numbered files, to wave people off of editing them, or at least, warning of the possible consequences? Edit notices show up above the preview window when editing. (Not sure what happens in VE.) One of the things we could mention in the edit notice, if desired, is the noinclude workaround. In the meantime, it would be a good idea if we all added numbered files WP:Reliable sources/Perennial sources/1 – WP:Reliable sources/Perennial sources/8 to our watchlists. Mathglot (talk) 06:07, 7 October 2025 (UTC)
I won't add a WP:Encyclopedia.com shortcut then. Gråbergs Gråa Sång (talk) 07:52, 7 October 2025 (UTC)
In general, as mentioned in the inclusion criteria, do not add a shortcut unless it is used elsewhere. Use {{RSP entry}} instead of creating a shortcut until the entry becomes really really popular. Besides what we already feared about it bumping against technical limitations, ti turns out that the shortcut template is the biggest non-textual contributor that bumps us against the PEIS limits. Aaron Liu (talk) 00:30, 8 October 2025 (UTC)
How can a shortcut to WP:RSP at this stage be used elsewhere unless I add it somewhere else the same time? The point of the shortcut here is to show editors it exists so they can use it when talking to other editors. I don't think every RSP-item needs one but I have seen this one come up in discussions now and then, so it seemed a reasonable one to me. But noted that Template:RSP entry exists. Gråbergs Gråa Sång (talk) 06:48, 8 October 2025 (UTC)
Feel free to create a shortcut if you are using it now. Shortcuts should be created on the same day that you want to use it for the first time. Please do not create shortcuts on the off-chance that someone might someday want to use it. WhatamIdoing (talk) 16:59, 8 October 2025 (UTC)
Noted, but my guess is that most of the shortcuts on WP:RSP were created with the "someone might someday want to use it" motivation. It was certainly mine when I made WP:RSPSCRIPTURE and WP:RSPYT (I also had a "I will certainly use this" motivation). I was right, too. Gråbergs Gråa Sång (talk) 17:18, 8 October 2025 (UTC)
Gråbergs Gråa Sång, there is another solution, which avoids increasing PEIS which involves just adding [[WP:NEWSOURCELINK]] to your row instead of using the WP:RSPSHORTCUT template. Your link will go red, but it won't look like the little bordered box with the map pin. It's possible to get that look-and-feel without impacting PEIS and with not too much extra work; if you want to try that, add [[WP:NEWSOURCELINK|<span class=wp-rsp-sc2>WP:NEWSOURCELINK</span>]], which renders as WP:NEWSOURCELINK, and once you create the link, it will go green (as a redirect). It's fine to use that, if you want. Mathglot (talk) 01:40, 11 October 2025 (UTC) I've gone ahead and added it for you. Note that it doesn't behave identically with the other ones, which if you click them, go to the redirect page and are blue; this is green, and just comes back to the table row again if you click it. But, it gets you a visible, copyable link that users can use. Mathglot (talk) 02:00, 11 October 2025 (UTC)

For the record: the recent change to the template makes the above hack no longer needed, and it could be backed out in favor of just using the shortcut template. But it's probably not worth doing it just now, until the result of the Rfc is in. Mathglot (talk) 17:46, 21 October 2025 (UTC)

RFC drafting section

 Courtesy link: Wikipedia:Requests for comment/Restructuring RSP     (this is the Rfc resulting from this discussion)

Here's my draft:

Which option should be used to fix the technical limitations that will prevent us from expanding Wikipedia:Reliable sources/Perennial sources (RSP)?

RSP currently lists about 500 sources, with a growth rate of about 50 new entries per year. With the current format, the page has reached the WP:PEIS template limits. Only templates within the limits are displayed; templates (and their contents) past that point on the page are not displayed. We need to restructure RSP to reduce the PEIS problem and accommodate more entries.

Editors have identified three main approaches to solving this problem. We are calling these three options "One giant table", "List of subpages", and "Row-building module". All options have advantages and disadvantages. Before we invest more hours in developing the options, we want to know which option is most appealing to the community. 02:46, 18 October 2025 (UTC)

Details

All of these are in the prototype stage. Volunteers are already lined up to implement any of them. All of them are expected to be implemented via semi-automated conversion from the existing RSP pages. There are many details yet to be settled, but this is our best estimate of what's possible:

  • One giant table: All the table rows in WP:RSP are currently split among eight subpages (example), which are transcluded into the main RSP page. In this proposal, the subpages would be combined into a single giant table in the main RSP page. See WP:RSPGIANT for an editable example.
    • Appearance: It would look the same.
    • Capacity: We estimate that we could add about 300 new sources (about six years) with the "One giant table" approach. After that, we'll have the same problem again.
    • Performance: Loading speed for the page would be similar to what we have now.
    • Editing experience: It would be more difficult to edit the entries because there would be so much wikitext in the editing window. However, it wouldn't be impossible, even on mobile or with the visual editor, for most editors to make changes or add new rows.
  • List of subpages: Put each entry (table row) into its own, shorter subpage. WP:RSP would have a list of all sources plus a search box. See WP:RSPINDEX and the linked subpages on it for an editable example.
    • Appearance: It would look significantly different. RSP would have a simple bullet list with links to separate pages (mockup). Clicking on the link will take you to a subpage with information about the source. Shortcuts such as WP:RSPBBC would take you directly to the relevant subpage (example). The exact style of the subpages has not been settled, but see examples for BBC (generally reliable), Ballotpedia (no consensus). California Globe (unreliable), CNET (complicated source), Deutsche Welle (shorter format). On most of these sample pages, you can scroll down to see a copy of the original row table for comparison.
    • Capacity: We could add thousands of new sources (many, many years; essentially a permanent solution) with the subpages approach.
    • Performance: Loading speed for both the main RSP page and each individual subpage would be much faster than what we have now. However, you might have to load two pages (RSP to find the link to the subpage, and then the subpage). The reading experience is probably better for mobile users (no scrolling sideways on a wide table).
    • Editing experience: It would be easier to edit and to add new sources. However, you'll have to remember to add any new pages to the main list in RSP.
    • Other: Use Special:RecentChangesLinked for the main RSP page to see all changes to all subpages (example; scroll down to 6 October or earlier). The function currently provided by table sorting can be replaced with categories; see test pages in Category:Wikipedia perennial sources.
  • Row-builder module: RSP and its associated templates will be re-written to use a Lua module.
    • Appearance: It would look similar to what we have now. See this example.
    • Capacity: We estimate that we could add up to 200 new sources (up to four years) with the "Row-building module" approach. After that, we'll have the same problem again.
    • Performance: Loading speed for the page probably would be similar to what we have now.
    • Editing experience: Instead of editing a table, each row would use a template or module invocation, so the wikitext will look different. (A somewhat similar approach is used by {{Episode table}} in articles such as List of The Brady Bunch episodes#Episodes.)
    • Other: Lua programming skills are not common in the community, which means that we might struggle to find a volunteer to make any future changes.

If more than one approach seems acceptable to you, then for the convenience of the eventual closer, we suggest naming your first choice and a second choice.

Discussion and questions
Survey

@Mathglot, Aidan9382, Aaron Liu, anyone else: Please read this and especially check whether I've got the correct estimates for expansion sizes. As it's so long, I have highlighted the facts that I think will be most relevant to most editors, and I'm curious about what you think of that.

For any folks who see this but haven't been following the huge discussion above: I don't need to know which one you would !vote for, but I would very much appreciate knowing whether you feel like this description is clear enough that you could figure out why we're asking this question and whether you feel like you could choose one of the options. If it's incomprehensible, then please tell me, so I can do better. WhatamIdoing (talk) 03:10, 18 October 2025 (UTC)

WhatamIdoing, great job. I would not have thought of using this format, but I like it. Not too short and mysterious, not too long and tedious, but just right (the Goldilocks format). A couple of minor points:
  • re: emergency measures, we haven't actually been forced to use the WP:RSPOVERFLOW table yet to keep RSP readable; the four 'Z'-items in there now are copy-paste models for ease of use. However, we could mention that all eight chunks or subtables or whatever we call them are now frozen, and the next row will have to go into overflow; it is only through the emergency measures (some of them sketchy) that we have lasted this long. (I wouldn't mention all that, and maybe none of it, but just wanted to correct the record.)
  • for 'list of subpages', where it now says "Split each entry", I would write, "Split each entry (i.e., each row in the table)" (or, "spin off"). Wearing my bot-pinged-editor hat, "Splitting an entry" made me pause a second to make sure I understood.
  • estimate for row-builder module: where did '200 rows' come from again? That could've even been me; you're right that the thread is long and it's hard to keep track, plus things have evolved. I recall two estimates, high/low, from Aidan of 135 to 185 or some such – depending on some technical issue I don't recall – which to me is "150-ish", and if we are using 50 rows/year, that works out to 3 years. But don't take my word for it, plus there are also Aaron and Audiodude's row builders as well, so maybe that's where it came from.
I think you asked the important question for this rfc question-building discussion at the end, namely, not which one do you favor, but do you understand the question as posed above and the options, or could we do better? I think it's fair to ask a couple of participants who were here early on for feedback on that, namely @Johnuniq and SuperGrey:. If you wanted to have a small number of additional opinions I can think of two more (uninvolved) editors I could ping who I think have an especially good ear for this sort of thing, but I'll hold off for now until I see how this discussion goes. Well done, it's been a long road, I know, but it's been worth it, and here we are on the brink. Kudos. Mathglot (talk) 03:54, 18 October 2025 (UTC)
I'm making copyedits (in this edit) based on your comments. I have removed the reference to WP:RSPOVERFLOW as unnecessary. I got the "200" from Aidan's recent estimate of 188, which I rounded to 200; I'm changing that now to "up to 200", but if you want to change it to "about" 150/3 years, that's okay with me.
I'm happy to have you (or anyone else) ping as many (or as few) other editors as you want. I'd rather have a dozen pings now than have a dozen confused respondents later. WhatamIdoing (talk) 04:58, 18 October 2025 (UTC)
As far as the row count estimate for the row-builder module, my recollection is that 150-ish fits better than 200-ish, but I'll wait till Aidan chimes in. Regarding pings: @Sdkb and Rjjiii: – imagine you were randomly summoned to an Rfc containing the text above, starting with the horizontal rule under "Here's my draft", spanning the portion headed "Details", and ending above the bold header "Discussion and questions". After reading that, your feedback on the following question would be helpful: "[D]o you feel like this description is clear enough that you could figure out why we're asking this question and whether you feel like you could choose one of the options". This is not about asking for your preference or !vote at this time, but about drafting a clear Rfc question. Your views as uninvolved editors would be very helpful. There is a lonnnng backstory; if you want the gory details, see § Addressing hard template limits, but kindly formulate your response before peeking! Thanks, Mathglot (talk) 08:08, 18 October 2025 (UTC)
I read the bit above and made these notes before looking at the technical aspect:
  • "the WP:PEIS problems in"
    • Will most readers understand WP:PEIS and do they need to understand it here? Maybe something like "the technical limit reached in" as WP:PEIS is in the next paragraph.
  • "RSP currently lists about 500 sources, with a growth rate of about 50 new entries per year. With the current format, the page has reached the WP:PEIS template limits." ← looks good to me
  • "When the template limits are breached, content in templates is not displayed on the page." I think it would be better to phrase this in terms of this specific discussion and not templates in general.
    • Maybe, "Beyond those template limits, new entries will not be displayed on the page."
  • Is "For the last couple of weeks, we have been using emergency stop-gap measures, such as removing shortcut templates, to keep RSP readable." necessary? It might give the wrong idea that there are more of these tricks in the bag, when the reality seems to be that the bag of tricks has been fully emptied.
    • Maybe just remove this line?
  • "Editors have identified ..." ← This paragraph seems very clear to me. I like bolding on the three options for folks who have started to skim.
  • "Details" I think the highlighting works really well here. This is inherently technical and the highlighted parts are what will be relevant to most. You might even consider using nearly the same language just for the highlighted parts to lean into that and make things easier to read. So "The visual appearance will be significantly different"→"It would look significantly different" and "easier to make changes"→"easier to edit".
After looking at the technical aspect, I tried several stopgap measures to buy some time for discussion and planning, since the page was already bumping against the limit.
Post‐expand include size was at:
2075765/2097152 bytes (99%)
After removing some template calls it dropped to:
1873301/2097152 bytes (89%)
And after removing the unused r parameter for reversing display order ("r=y") in {{duses}} it dropped to:
1828451/2097152 bytes (87%)
If the show/hide logic were removed (I reversed this and was demonstrating), it would drop slightly more to around 1778271/2097152 bytes (85%), but there might be objections to that change.
That should buy some time, but it might also mean the math needs to be redone above in the above explanation. Rjjiii (talk) 11:54, 18 October 2025 (UTC)
I'll note that your changes to WP:RSPSHORTCUT aren't perfect, as it loses the invisible wikilink normally produced by {{no redirect}} to generate "what links here" information, as well as making it not work standalone due to the removed templatestyles. Aidan9382 (talk) 18:59, 18 October 2025 (UTC)
As an emergency stopgap measure, it's fine. As a long-term solution, it isn't the best. WhatamIdoing (talk) 19:09, 18 October 2025 (UTC)
Thanks for catching that, Aidan9382. I have manually added the invisible wikilink into WP:RSPSHORTCUT now. PEIS is still around 89%. Rjjiii (talk) 20:01, 18 October 2025 (UTC)
In light of Rjjiii's space-saving change to WP:RSPSHORTCUT, I think we should undo this sketchy edit made in an emergency attempt ten days ago to get down below 100%, which entirely replaced the RSPSHORTCUT template with its expansion, and which eventually made it to Wikipedia:Reliable sources/Perennial sources/2 after a swap between it and its sandbox version. Undoing it will likely increase PEIS slightly, but now we have room to play with; there's no reason the shortcut code uniquely in chunk #2 should remain so ugly. (Also, we should merge history from the original subtable 2, which now occupies the sandbox, but that should await the result of the Rfc to avoid possibly wasted effort depending what approach is chosen.) Mathglot (talk) 02:40, 19 October 2025 (UTC)
About "new entries will not be displayed": It's not "new" ones; it's whatever's towards the end of the wikitext, which could be very old.
I have taken your advice to remove the "emergency stop gap measures" line as unnecessary and standardize some of the language. [4] WhatamIdoing (talk) 19:20, 18 October 2025 (UTC)
Yes, and what's at the end of the wikitext template-wise are the notes and references, they would begin to degrade first. (This has already happened a few times, and could be linked from the rev history of the RSP page if desired.) When no more notes and references were visible, the content in the 'Z'-rows would start to degrade, and so on. Mathglot (talk) 19:45, 18 October 2025 (UTC)
It's probably a good idea to have links to a few of those revisions handy. WhatamIdoing (talk) 19:52, 18 October 2025 (UTC)
Oh, you are right. I was thinking of it wrong. Also, Archive.org snapshots show multiple times this year where the notes and references broke down when the limit was hit. Rjjiii (talk) 20:34, 18 October 2025 (UTC)
That link should be sufficient. We can keep it in our back pocket and pull it out if anyone asks. WhatamIdoing (talk) 21:01, 18 October 2025 (UTC)
Just wanted to underscore the wise choice of using an Internet Archive link to demonstrate this, rather than a link out of the page history of WP:RSP. To preempt possible questions about this, the crucial distinction is that IA preserves the Html of the page as it was at a given time, which means, of course, after all templates were expanded. However, the page history preserves the wikicode of the page as it was then, before they were expanded, and if you click a history link from, say, a month ago, all the templates will be expanded as they are now, not as they were then, thus not a valid picture of what the table looked like in the past. Only a full capture of the page Html can do that. Mathglot (talk) 21:10, 19 October 2025 (UTC)
@Aidan9382, would you rather have the row-building presented as a 150 rows/3 years or as 200 rows/4 years? I don't think our estimates are precise enough to justify being more specific than to the nearest 50 rows/1 year. WhatamIdoing (talk) 19:28, 18 October 2025 (UTC)
@WhatamIdoing: Well, I decided to go all out on my sandbox and try converting all of table 5. Obviously with the RSPSHORTCUT stop-gap in place my estimates might be a bit off, but if my notes are correct, table 5 was 252,249 bytes in its old form (it's 223,128 right now), vs 182,799 for the module approach, which puts the estimate (before the stop-gap measures mentioned above) at ~190 extra rows, which makes 200 a pretty decent estimate. Aidan9382 (talk) 20:02, 18 October 2025 (UTC)
Thanks. Mathglot, unless you strongly object, then I'm inclined to leave it like it is. WhatamIdoing (talk) 20:15, 18 October 2025 (UTC)
Given these details, leaving it as is sounds exactly right. Mathglot (talk) 20:31, 18 October 2025 (UTC)
The RFC draft looks great. Addressed all the important points -- I don't see anything needed to add. SuperGrey (talk) 04:02, 18 October 2025 (UTC)
Thanks. WhatamIdoing (talk) 04:58, 18 October 2025 (UTC)
A minor point: the first two options link an example; the row-builder module approach links a Lua help page, but we could link an example; should we? Mathglot (talk) 04:07, 18 October 2025 (UTC)
I'd be happy to have a link there, but I'm not sure what looks like an example of the output. WhatamIdoing (talk) 04:58, 18 October 2025 (UTC)
WhatamIdoing, I think the relevant link there would be User:Aidan9382/sandbox3, but I would like to hear from Aidan9382—can you confirm that that would be the best link to include as an example of possible rendered-table output from a row-builder module, if we include just one such link in parallel with the other two (non row-builder) options? Mathglot (talk) 19:56, 18 October 2025 (UTC)
@Mathglot: That's my testing page for the module implementation, yes. If you do want to link it, make sure to use a permalink (I'd recommend Special:Permalink/1317468803), but the key part about how I designed the lua implementations for my testing were to be 1:1 in html to the templates (within reason), so there's no visual difference beyond what the new wikitext would look like. Aidan9382 (talk) 20:14, 18 October 2025 (UTC)
Ah, yes; good point. WhatamIdoing, I would recommend adding this text:
See [[Special:Permalink/1317468803|this example]] module.
after the words "Lua module" (and maybe unlinking "Lua module"). Mathglot (talk) 20:41, 18 October 2025 (UTC)
I've added it to the Appearance line. WhatamIdoing (talk) 20:58, 18 October 2025 (UTC)
Thanks WhatamIdoing, that is excellent. However, you might think about revising the draft for a punchy RfC question with the information currently in the introduction moved to Details. Example question: What option should be used to fix the WP:PEIS problems in Wikipedia:Reliable sources/Perennial sources?. Re the options, very briefly, what arguments exist to avoid a list of subpages? One would be that having everything on one page allows searching for half-remembered text. Another is that one page makes monitoring changes much easier. An unmentioned advantage of subpages is that it would use familiar wikitext and would allow exceptions from standard formatting if needed. The giant and module options would require uniformity. I wonder if such arguments should be mentioned in Details. An example of subpages is seen at WP:LTA. Johnuniq (talk) 04:19, 18 October 2025 (UTC)
I'm dropping your "punchy" question right at the top, where I think it fits in nicely.
The subpages item is already longer than the others, so I'm hesitant to add any more to it. Also, I think for most editors, the main disadvantage is that it's the biggest change from the POV of the editors who use (rather than edit) RSP. They might like it eventually, but for right now, the cost of the transition is high. WhatamIdoing (talk) 05:18, 18 October 2025 (UTC)
Johnuniq, regarding the advantage of non-standard wikitext in subpages, I had considered this earlier and wrote it up as part of Q3 in section WP:RSPINDEX § Demo FAQ, but I don't see a good way to deal with that at the top level of an Rfc presentation, pretty much for the reasons WaId expressed. Flipping that around: analogous to Demo FAQ Q5, it would be natural for someone in the ensuing rfc discussion to ask about the migration paths, and clearly that is easiest for the Giant table as it is basically already done (or can be redone to pick up late changes), but again, where is the point of TMI in the initial question? Maybe that has to be left until someone asks about it? Mathglot (talk) 20:28, 18 October 2025 (UTC)
On second thought: what if we added italicized bullet Migration path to each approach? I think we could afford to add a sentence about each one without overloading the description. Mathglot (talk) 20:45, 18 October 2025 (UTC)
Are you ready and willing to implement whatever is chosen? If so, then I don't think that the cost of migration should be emphasized. WhatamIdoing (talk) 21:02, 18 October 2025 (UTC)
Ideally, migration for the landing-page approach should be scripted. It would be too hard for me working alone to implement the entire table as landing pages, given that I don't have an off-wiki scripting environment, and implemented the landing pages by executing a regex once per row. That wasn't too bad for the ~80 items under 'B', 'C', and 'D', and I think too me a week or ten days, but would be too much for one person using the regex another 420 times or so. I could do it, but it would probably block all my other Wikipedia activity for months, and I don't know that I'm willing to do that.
In theory, the existing row-builder module, which already does the necessary work to parse out the row fields, could be modified to generate the landing page-style format instead; we just have to specify what that format is (i.e., something that looks like Ballotpedia, or something that looks like Deutsche Welle, or whatever format—that could be decided by separate Rfc). The other major change is that it would have to write each resulting transformed output onto its own page, and I don't know Lua, but I don't think that is too onerous. Aidan9382, can you comment on the feasibility of such a module? (Input: table with rows like WP:BALLOTPEDIA; output: set of individual landing pages, like Ballotpedia, or some other format, t.b.d.)
One thing to keep in mind, is that unlike the row-builder approach, say, where the builder would be a permanent feature of the new table, the migration path for the landing page approach is needed only once: for the conversion of the existing table; after that, it could be thrown away. This has two major implications: it doesn't have to be anywhere near as robust, codewise, as any problems with a given row could simply be patched up after conversion by editing the wikitext of the emitted landing page. Secondly, it does not have to be done on-wiki in Lua—it can be written in any language and done offline; if someone wants to do it in Perl or Python, that works. If there's an advantage of doing it in Lua, it's probably that Aidan's module does a good part of the work already, but it isn't a requirement. So I think we could say this method requires a one-time conversion script that could be run on- or off-wiki, and could be similar to the existing row-builder module, but doesn't have to be. (Note: the Index table links could be generated by a single regex operation off the "id" attribute in the table rows, so that is not an issue.) Mathglot (talk) 21:42, 18 October 2025 (UTC)
The module itself obviously couldn't create the pages, but it would certainly be possible to make a module that made a simple layout in whatever way you'd be looking for based on the existing row data in one way or another and for you to just {{subst: it so it turns into a permanent raw text form. You'd still have to go page by page substituting it for each one, but it'd skip a lot of the work involved. If you really wanted to save time, you could make a bot task to do all of the substitutions with a module (or just have it do the logic itself externally), but that might be overkill depending on the scale in question here. Aidan9382 (talk) 22:15, 18 October 2025 (UTC)
The table has 500 rows; not sure which would make more sense. Sounds like modifying the module for a new format is not too hard, and then a very simple bot that would loop on table rows invoking the module once per iteration with a single row and writing the resulting module output would be a pretty simple bot, but someone still has to do it. It sounds generic enough, maybe there's something like that already out there. Grabbing a single row from the table is trivial. Mathglot (talk) 23:52, 18 October 2025 (UTC)
If we were in a situation in which we have a volunteer to implement "A" but not "B" (no matter those items are), then we'd want to think harder about whether "B" is a viable outcome. Remember, e.g., WP:LUGSTUBS2, in which the community decided on a particular approach, but then nobody actually implemented it, because "my" part is complaining, and actually doing the work is somebody else's problem.
My impression is that we don't have that situation here. Sure, WP:RSPGIANT requires less effort to implement than the other two (though more effort to maintain), and there's always a chance that we'll need to find help (e.g., a bot, another Lua coder...), but in principle, the transition costs are not a blocking factor on any of the options. WhatamIdoing (talk) 23:55, 18 October 2025 (UTC)
A few comments:
  • The first option has a note about reversing the 2024 decision, but the second option does that as well, as there would no longer be a fully searchable/sortable table.
  • I would add searchability/sortability as one of the bullet points, as that's the second of the two results from the previous decision.
  • I would also include the option for a "standard" page split into 2-3 subpages (or explain why it isn't feasible), since it's one of the options that attracted support in the last discussion. Arguably it's a version of option 2, but it would be very different in practice, as the appearance would remain similar (instead of changing drastically) and it would be partly searchable/sortable with extra steps (instead of only searchable by title or through the search box).
--Sunrise (talk) 05:26, 18 October 2025 (UTC)
In order:
  • The first option's purpose is (more or less) to reverse the 2024 decision (to split the wikitext into eight tables and transclude them back into the main page). With the second option, it's more of a side effect than a goal.
  • The second option doesn't provide a ⌘F compatible version for arbitrary text strings, but it does provide a search box (so it's "searchable" in a different way).
    The "sortable" function may be replaceable with categories (e.g., if you want to see all the GUNREL sources). It would be useful to know what people are trying to accomplish when they sort (available for only three columns: the name of the source, the status, and the date of the last discussion). If the answers are things like "make a list of all GUNREL sources" or "see how many NC sources there are" or "find outdated entries", then a category is likely to be more effective.
  • I'm not sure what you mean by a "standard" page split. Just chop the table in half, and put half on each of two pages, like we once did with List of colours: A-M?
WhatamIdoing (talk) 06:06, 18 October 2025 (UTC)
Sunrise, I echo WaiD's comment about sorting, and just wanted to add that categories are already available in the demo at WP:RSPINDEX. You can view categories by date by status (and more could be added to view it the other way round). Top cat is Category:Wikipedia perennial sources – see the subcats for an idea how that goes. Or, bottom-up: click any source link at WP:RSPINDEX to view it and find the status & date category it is in, and then see other sources sharing the same category (e.g., this one). This is analogous to sort-on-status-column, except there is room for a lot more links on a category page, than table rows that will fit in a screenful. For search, you can search on any text from the search box, not just source names. Mathglot (talk) 06:51, 18 October 2025 (UTC)
In order (and @Mathglot, thanks for the comment about categories):
  • I see the two main outcomes (meaning the two bolded portions of the close) as roughly co-equal in importance, in addition to representing the concerns of the two main viewpoints in the discussion. Those who gave opposing rationales might consider the strict condition clause (the result they argued for) to be the key outcome and the impact on performance from splitting to be the side effect. I don't think it makes sense to indicate when one RfC result is being reversed but not the other one.
  • Those are important differences, in my view, and I don't think it's clear which system would be more effective. To an extent, the information is already included implicitly, e.g. the inclusion of you might have to load two pages under "Performance". However, as in the first point, this relates to one of the two main viewpoints from the previous RfC, so I would describe it as its own point.
  • Yes, exactly. That said, I see that it’s been discussed in Demo FAQ #1c. I don't really agree with the response, since the primary table is not itself split, and it's citing the same RfC that we're already talking about reversing. However, I haven’t spent nearly as much time thinking about this as all of you clearly have (and I might always be misunderstanding something about templates), so I'm satisfied that you’ve considered the topic. :-)
--Sunrise (talk) 08:41, 18 October 2025 (UTC)
The 2024 discussion was not an RFC. There was only one RFC on this page in all of 2024, and it was about a sticky header gadget for the table. Maybe we just shouldn't bother mentioning it.
There are other options that could be considered (e.g., removing entries, so that only the 500 most 'important' are listed), but these are the three that these editors are prepared to offer to the community. WhatamIdoing (talk) 19:06, 18 October 2025 (UTC)
Wonderful draft! I like the idea and execution of this. A few notes though (please excuse me if these are already addressed in comments below):
  1. I strongly disagree that making changes would be approximately the same. The Audiod approach can be summarized as "instead of filling out a table, you'd be filling out a template" with an example row syntax, and the Aidan approach would be filling out a JSON, which is a very different experience thanks to the lack of Wikitext tools in the JSON editor.
  2. Is "about 50 new entries per year" accurate? It feels quite a bit less than that to me.
Aaron Liu (talk) 19:26, 19 October 2025 (UTC)
Aaron, where do you see Aidan9382's approach as having anything to do with JSON? See User:Aidan9382/sandbox3 and Module:Sandbox/Aidan9382/RSP; are you looking at something else? Mathglot (talk) 19:51, 19 October 2025 (UTC)
Okay, I didn't see that Aidan's approach had changed... It indeed pretty much preserves the existing editing experience. I don't see the reasons to use it though. I'll ask above. Nevermind, me thinkie loopy. Aaron Liu (talk) 19:56, 19 October 2025 (UTC)
@WhatamIdoing We might have restore the earlier wording, as while the current wording is correct for Audiod's approach, the earlier wording is correct for Aidan's, which we seem to be trending towards. Sorry about that. (Also, I'm convinced now that the new rows estimate is indeed correct.) Aaron Liu (talk) 18:45, 20 October 2025 (UTC)
Okay. I'll change it to "instead of filling out a table, you'd be filling out a template".
For #2, the net growth in the table per archive.org copies was 60 rows last year and 40 rows the year before, ergo about 50 rows per year. WhatamIdoing (talk) 21:03, 19 October 2025 (UTC)
I was just reviewing #2 while you did, and I came up with 65 new entries since a year ago, by taking this capture from IA of 16 October 2024, and got 435. However, on that date, there was a banner reporting a merge request from Wikipedia:Deprecated sources#Currently deprecated sources that had been there since June, and presumably that happened in the interim, so the 65 rows added in the last twelve months may be inflated due to that. Your two calendar year analyses before the merge should be more accurate. We could count 2025 and pro-rate for the last three months if we wanted to add this year to the estimate. Mathglot (talk) 22:18, 19 October 2025 (UTC)
I think that we should be thinking in very broad terms this purpose – big round numbers, half an order of magnitude. If we say 50, and it turns out to be 25 or 100, then that's a bit awkward. If we say 50 and it's 35 or 65, then that's no big deal. WhatamIdoing (talk) 22:35, 19 October 2025 (UTC)
In the details section, I would modify the bullet for One giant table as follows:
  • One giant table: All the wikitext 500 rows currently split among eight subtables would be merged into a single enormous page table.
The table isn't the only thing on the page now, and wouldn't be after, hence enormous 'table'. If desired, we could link 'subtables' to the first one at Wikipedia:Reliable sources/Perennial sources/1, or add a '(1–8)' parenthetical with both numbers linked, which should illustrate the point sufficiently. Mathglot (talk) 21:44, 19 October 2025 (UTC)
Yes, updated version looks fine. Mathglot (talk) 22:21, 19 October 2025 (UTC)
Okay, are we about ready to post this? Any objections? Any last-minute changes?
Since this talk page is already uncomfortably large at the moment (well above 300,000 bytes), and since there were discussions above about the One True™ Location for this discussion, I think we should put it on a separate subpage (e.g., Wikipedia:Requests for comment/Restructuring RSP). As far as I'm concerned, we should give folks at least a few hours (half a day?) to reply, and then anyone who agrees that it's ready can start the RFC. The RFC cats for the tag could be |tech|policy|prop (or a subset of them). WhatamIdoing (talk) 22:40, 19 October 2025 (UTC)
Minor one: Under Details, Row-builder module, Editing experience:
Instead of editing a table, each row would use a template or module invocation, so the wikitext will look different.
The addition is if we go with something that looks like User:Aidan9382/sandbox3. For that matter, in whatever flavor of row-builder module that is developed, invoking the module directly (i.e., using #invoke rather than a template wrapper) will offer some savings in PEIS, and thus a somewhat longer life. But either method (template or direct invoke) would work with any row-builder that was developed, so it isn't really part of the row-builder approach, but how you employ it once you have it. Other than that (and it is minor) I think we are ready to post. (edit conflict) Mathglot (talk) 22:46, 19 October 2025 (UTC)
Okay, I'm adding that. NB that I've deliberately added it outside the highlighted text, because I want editors to focus on the overall concept and not on "what the heck is a module invocation?!" WhatamIdoing (talk) 23:04, 19 October 2025 (UTC)
No objections, and thanks all for putting so much into this, Rjjiii (talk) 23:30, 19 October 2025 (UTC)
I'm launching this now. Join me at Wikipedia:Requests for comment/Restructuring RSP. Sometime during the next 24 hours, the Wikipedia:Feedback request service will send out announcements. Responses from that will slow down in about two days, at which point we should consider posting some manual announcements (e.g., to Village pumps). We don't need to do all the announcements on the first day, but we should make sure that this is widely advertised during the next week or two. WhatamIdoing (talk) 22:48, 20 October 2025 (UTC)

Tangent: when the Rfc is launched, we should update the wording at WP:RSPTableFull, which appears at the top of each chunked 1–8 subtable currently, in order to add a link to the Rfc. Feel free to do so, or {{ping}} me and I'm happy to do it. Mathglot (talk) 08:37, 18 October 2025 (UTC)

Or given the recent change to WP:RSPSHORTCUT reducing table PEIS, perhaps we should "unfreeze" the subtable chunks, and remove the Table Full message? Mathglot (talk) 20:56, 18 October 2025 (UTC)
I'm going to be honest but I've started noticing a bit of an alarming trend with WP:RS/P which is that people think of it as being the final word for the reliability of all sources. IE: I see a lot of editors argue a source is not contextually unreliable either because it's not WP:GUNREL or, even worse, because it's not mentioned. This argument seems to often be put forward to support the use of marginal sources for controversial BLP statements. For instance, today's example is someone arguing we should use the Toronto Sun (which has the worst reputation of any major Canadian newspaper) to call a Canadian political candidate a genocide denier. With this in mind I'm wondering if we shouldn't be considering mothballing RSP and, instead, maintain a list of deprecated sources only. Then people might have to actually engage with context on discussions of reliability. Simonm223 (talk) 18:51, 20 October 2025 (UTC)
Yup. This was predicted before RSP was created (that expectation is why RSP didn't exist until 2018) and the prediction has come to pass. We do what we can to educate editors (e.g., Wikipedia:Reliable sources/Perennial sources#What if a source is not here?) but Wikipedia:Nobody reads the directions, especially if invoking RSP lets me "win". WhatamIdoing (talk) 22:41, 20 October 2025 (UTC)

The fruits of our labor, now live ripe

 You are invited to join the discussion at Wikipedia:Requests for comment/Restructuring RSP. (Using this template for better visibility.) Aaron Liu (talk) 21:48, 21 October 2025 (UTC)

I wouldn't say that we necessarily have to close this discussion, but I added a convenience link at the top linking the Rfc. Mathglot (talk) 00:01, 22 October 2025 (UTC)

Icon image licensing

I noticed that a few of the icons used here with |link= have licenses that require the link to the file description page for attribution and notice of license. These should be replaced with public domain or CC0 icons. See MOS:BLANKALT and MOS:EMPTYALT for details.

Icon License Notes
GFDL / CC BY-SA is a similar PD file.
CC BY-SA / MIT Could very likely be declared {{PD-shape}} on Commons, or replaced with one of .
CC BY-SA Could maybe be declared {{PD-shape}} on Commons, or replaced with one of .
CC BY I didn't find any existing alternatives in a quick search, but it should be easy for someone to combine something like and one of to make something similar.

I didn't want to just change things without discussion, so I'm starting a discussion here. Anomie⚔ 14:35, 21 October 2025 (UTC)

Are we sure that the copyright for the OOUI icons aren't already covered by that of MediaWiki, which incorporates them?
An easier solution for the first and last icons is to just link them and unset the alt. I don't think their links would be particularly interfering, and I'd be surprised if their alt is blank since they're not purely decorative. Aaron Liu (talk) 20:46, 21 October 2025 (UTC)
My take is that if we use the Commons version, then we need to follow the conventions to comply with the license that's stated on Commons. Many are probably PD-simple of some variety rather than the MIT or CC license they're tagged with, but someone needs to actually make that change on Commons and make it stick. If they give us a parser function or <span class="..."> or something that uses the version that's actually embedded in MediaWiki, then it's covered by MediaWiki.
By dropping the |link= on the first and last, you'd lose the icons linking back to Wikipedia:Reliable sources/Perennial sources#Deprecated and Wikipedia:Reliable sources/Perennial sources#Stale discussions that explain what they mean here. But if that's fine with people here, it'd work to resolve the licensing issue. Anomie⚔ 21:51, 21 October 2025 (UTC)
I personally think WMF's/MW Contributors' licensed us the permission for the OOUI images, but I do see that it's quite vague. I think we can replace them with the suggested alternatives № 2 and 2 for now; there's little perceptible difference.
Would using the unset link in the legend linked to work for attribution (and, of course, since there's no links in the subpages making them have the unset link too)? I couldn't find the exact policy about attribution, though it's referenced in BlankAlt. Aaron Liu (talk) 22:08, 21 October 2025 (UTC)
I don't know to what extent we want to count having to click through multiple pages as satisfying the attribution and notice of license requirements, particularly when we normally use a direct link. Someone might well argue that the CC licenses' "reasonable to the medium" for Wikipedia has been established as "link the image to the file description page" and therefore an indirect link isn't enough. 🤷 Personally I'd rather just use images that don't need such hoop-jumping in the first place instead of trying to justify it. Anomie⚔ 22:47, 21 October 2025 (UTC)
Well, I was thinking unsetting the links in the legend at the same time, so you'd just need to jump to a different section on the same page... Aaron Liu (talk) 23:10, 21 October 2025 (UTC)
Click on the icon in the table → click on the same icon in the legend is still multiple clicks to get to the attribution, although technically the same page. Anomie⚔ 23:17, 21 October 2025 (UTC)
Can't the authorship info and the licensing information just be added to the local image? We don't need to provide credit in every image caption, so why do we need to here? This just reads like tedious pedantry honestly. Hemiauchenia (talk) 22:16, 21 October 2025 (UTC)
We normally provide credit for images by linking the image to its file description page that contains the necessary information. I don't understand what you're proposing as an alternative. Anomie⚔ 22:40, 21 October 2025 (UTC)
Several of these images are literally by WMF employees and are so simple they'd be public domain by default anyway. You're making a mountain out of a molehill about a situation that is currently fine. Literally nobody but you, including the authors of the images has objected to the current situation, so why do anything? Hemiauchenia (talk) 22:45, 21 October 2025 (UTC)
"We can violate the license as long as no one complains" isn't a good argument at all. Try proposing "ignore copyvios until someone complains" at Wikipedia talk:Copyright problems and see what response you get. "Made by WMF employees" doesn't mean we don't have to follow the license either; if the copyright holder wants to give an exception for use on Wikipedia, they can make that part of the license. "So simple they'd be public domain by default" is an ok argument, but in that case why not just update the file description page to reflect that instead of asserting we can ignore the claimed license? I don't because I don't want to potentially get involved in discussions on Commons, I had enough of that years ago. Anomie⚔ 22:56, 21 October 2025 (UTC)
I think we should centralize this above in my reply chain which discusses the same thing you discuss. Aaron Liu (talk) 22:59, 21 October 2025 (UTC)
This isn't about providing credit in the caption. It's about whether you can click the picture to get to the File: page.
I think this should be postponed until Wikipedia:Requests for comment/Restructuring RSP is over, because everything on this page will change as a result of that discussion. We can just make a mental note to get the image licensing right this time. WhatamIdoing (talk) 01:56, 22 October 2025 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Reliable_sources/Perennial_sources/Archive_12&oldid=1321811203"