Jump to content

Wikipedia talk:Reliable sources/Perennial sources

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Liberal bias

[edit]

Msnbc doesnt cover events that are against the democrats but you consider it reliable but fox no its the same but some times covers republucan failures but its unreliable clearly biased Random conservative guy (talk) 19:59, 14 July 2025 (UTC)[reply]

Hey. Sources are never marked generally unreliable for bias; instead, they're marked so for persistent misinformation. More importantly, this page simply summarizes past discussions and agreement on whether sources are generally accurate. On this transparent platform, you can find out precisely why editors feel a certain way about a source by clicking on the links to discussions in the giant table you see on this page! In this case, you can see e.g. Wikipedia:Reliable sources/Noticeboard/Archive 303#Also CNN & MSNBC. If you want to try and change past agreement, you can start a discussion at WP:Reliable sources/Noticeboard! Aaron Liu (talk) 01:59, 15 July 2025 (UTC)[reply]
If you're interested you can read WP:RSBIAS that explains the policy that biased sources are not unreliable. MSNBC isn't unreliable because of it's slanted reporting, and slanted reporting isn't the reason that Fox is considered unreliable for science and politics. Each of the entries in the list contain links to previous discussions, the one marked with years are past RFCs (formal discussions that get a formal close), while the ones linked with a number or letter are informal discussions. If you read them and find anything you think is wrong then you should post to WP:Reliable Sources/Noticeboard with your thoughts. By editing you're now one of the people who decides such things, if you can make a good argument and convince others editors it's right you can change anything. -- LCU ActivelyDisinterested «@» °∆t° 09:25, 15 July 2025 (UTC)[reply]

Polygon

[edit]

I would like to add to the section about Polygon post-Valnet, they have seem to have started writing advertisements that look like articles. This may significantly affect the reliability of the publisher. So long, and thanks for all the fish (talk) 23:12, 17 July 2025 (UTC)[reply]

That article is clearly labeled "advertiser content". Schazjmd (talk) 23:15, 17 July 2025 (UTC)[reply]
Correct me if I'm wrong, but many of the reliable sources on Wikipedia do not create advertised content as articles. So long, and thanks for all the fish (talk) 23:31, 17 July 2025 (UTC)[reply]
Advertorials are fine as long as they are clearly marked as such. -- LCU ActivelyDisinterested «@» °∆t° 23:40, 17 July 2025 (UTC)[reply]
Agreed. WT:VG/S#Polygon separately recently came to the same conclusion as well. Sergecross73 msg me 03:10, 22 July 2025 (UTC)[reply]
In addition to the labeled inclusion of sponsored content (which is far more common than you may think) not affecting an outlet's reliability, we already have a separate entry for Polygon post-Valnet anyways. Aaron Liu (talk) 02:03, 18 July 2025 (UTC)[reply]

BizJournals

[edit]

Seems to meet RSPCRITERIA. Ongoing discussion at Wikipedia:Reliable_sources/Noticeboard#BizJournals, feel free to join. Gråbergs Gråa Sång (talk) 17:46, 19 July 2025 (UTC)[reply]

YouTube is not a source

[edit]

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)[reply]

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)[reply]
@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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
@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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
"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)[reply]
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)[reply]
  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)[reply]
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)[reply]
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)[reply]
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)[reply]
Thanks, have updated the discussion count to 36. CNC (talk) 17:15, 25 July 2025 (UTC)[reply]
  • 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)[reply]
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)[reply]
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)[reply]
@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)[reply]
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)[reply]
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)[reply]
@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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
  • 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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]
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)[reply]

Statista

[edit]

I'm undecided on the reliability of Statista, but the reasoning given on this page needs a better explanation. It says that Statista is not the source for the statistics it displays, so it shouldn't be used as a source. But this is true of many secondary sources that we regard as reliable. Do we have empirical evidence or some other principle to support this classification as unreliable? Alec Gargett (talk) 22:58, 25 July 2025 (UTC)[reply]

The entry doesn't make clear that it sometimes uses Wikipedia as a source, and it's data could be WP:CIRCULAR. -- LCU ActivelyDisinterested «@» °∆t° 23:43, 25 July 2025 (UTC)[reply]
Statista obfuscates its sources, and has no indication there is fact-checking or editorial control. (For example, I know they have used VGChartz for video game stats, but VGChartz is known to be incredibly unreliable). Reliable sources have processes in place to assure they can support the factuality of their sources. Masem (t) 02:36, 26 July 2025 (UTC)[reply]
The above is true only if you do not use the paid version, which is accessible through most institutions. They only get half of their data from external sources, the rest is generated in-house. Rollinginhisgrave (talk | contributions) 05:02, 27 July 2025 (UTC)[reply]
Could you give us a source for "half of their data is generated in-house"?
But you do have a point. You might want to ask whether Statista's truly in-house statistics are reliable at RSN. Aaron Liu (talk) 20:41, 28 July 2025 (UTC)[reply]
Aaron Liu I posted about this at Wikipedia:Reliable sources/Noticeboard/Archive 467#Statista (again) but it didn't get much attention. I originally read the claim that they produce half their content in the Journal of Business & Finance Librarianship's review of Statista. Rollinginhisgrave (talk | contributions) 02:39, 30 July 2025 (UTC)[reply]
Thanks, that's a T&F journal that indeed should be reliable. I'd be willing to participate in a discussion focusing on in-house content that also acknowledges the GUnRel status for all other content. I think not doing so clearly enough in the linked discussion resulted in people skimming things and not discussing what you wanted to. I can help draft an opening statement if you'd like, even though we're definitely not at the RfC stage. Aaron Liu (talk) 03:32, 30 July 2025 (UTC)[reply]
I'm sure you're right. Help with drafting an opening statement would be appreciated (and clearly needed). Rollinginhisgrave (talk | contributions) 03:47, 30 July 2025 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Reliable_sources/Perennial_sources&oldid=1303290092"