Wikipedia talk:Reliable sources/Perennial sources/Archive 10

Archive 5

←

Archive 8

Archive 9

Archive 10

Merging some entries

To reduce the number of entries in the list, I propose we merge some into broader categories.

Merge entries: Academia.edu + ResearchGate, add HAL Open Archives, Zenodo into "academic repository"

Before

Perennial sources

Source	Status (legend)	Discussions			Uses
		List	Last	Summary
Academia.edu		1 2 3 4 5 6 7 8	2024	Sources from Academia.edu may or may not be reliable. Academia.edu allows anyone to upload articles, so it doesn't confer any reliability, but the articles have often been published elsewhere first in which case the reliability of an article depends on whether the original source is reliable. When possible, use the original source in preference to Academia.edu.	1
ResearchGate WP:RESEARCHGATE 📌 WP:RGATE 📌		1 2 3 4	2022	ResearchGate is a social network that hosts a repository of user-generated publications, including preprints. ResearchGate does not perform fact checking or peer reviewing, and is considered a self-published source. Verify whether a paper on ResearchGate is also published in a peer-reviewed academic journal; in these cases, cite the more reliable journal and provide an open access link to the paper (which may be hosted on ResearchGate).	1

After

Perennial sources

Source	Status (legend)	Discussions			Uses
		List	Last	Summary
Academic repositories WP:ACADREP 📌 WP:ACADEMIA.EDU 📌 WP:RGATE 📌 WP:HAL 📌 WP:ZENODO 📌		+20	2024	General repositories like Academia.edu, HAL Open Archives, ResearchGate, and Zenodo, host several academic papers, conference proceedings, book chapters, preprints, technical reports, etc. No filters exist for quality, and will host several unreviewed preprints, retracted papers not marked as such, unreviewed manuscripts, and even papers from predatory journals. Determine the actual source of what is being cited first, a paper from Physical Review D will likely be reliable, whereas a paper from the so-called International Journal of Advanced Computer Technology will likely not be.	1   1   1   1

Headbomb {t · c · p · b} 22:11, 6 July 2024 (UTC)

Merging into an "Academic repository" source makes sense to me. (Also, RG, Academia, and Zenodo have similar characteristics, based on the the Nature survey in ResearchGate thread.) Consider including MDPI and Semantic Scholar in the merge? I'll check if they are even present in our RS/Perennial sources.--FeralOink (talk) 19:44, 10 July 2024 (UTC)

MDPI is a specific publisher, but Semantic Scholar goes in Academic Repositories, yes. But there's a zillion such repositories, so it's hard to track them all. Headbomb {t · c · p · b} 20:36, 10 July 2024 (UTC)

Broaden arxiv to preprints

Before

Perennial sources

Source	Status (legend)	Discussions			Uses
		List	Last	Summary
arXiv WP:ARXIV 📌		1 2 3 4 A B	2015	arXiv is a preprint (and sometimes postprint) repository containing papers that have undergone moderation, but not necessarily peer review. There is consensus that arXiv is a self-published source, and is generally unreliable with the exception of papers authored by established subject-matter experts. Verify whether a paper on arXiv is also published in a peer-reviewed academic journal; in these cases, cite the more reliable journal and provide an open access link to the paper (which may be hosted on arXiv).	1

After

Perennial sources

Source	Status (legend)	Discussions			Uses
		List	Last	Summary
Preprints WP:PREPRINTS 📌 WP:ARXIV 📌 WP:BIORXIV 📌 WP:MEDRXIV 📌		1 2 3 4 A B	2015	Preprint repositories, like arXiv, bioRxiv, medRxiv, PeerJ Preprints, and Preprints.org, contain papers that have undergone moderation, but not necessarily peer review. There is consensus that preprints are self-published sources, and are generally unreliable with the exception of papers authored by established subject-matter experts. Verify whether a preprint paper has been published in a peer-reviewed academic journal; in these cases, cite the more reliable journal and provide an open access link to the paper (which may be hosted on the preprint repository).	1   1   1

Headbomb {t · c · p · b} 22:19, 6 July 2024 (UTC)

Seems reasonable to me. — Red-tailed hawk _(nest) 01:43, 9 July 2024 (UTC)

Done. Headbomb {t · c · p · b} 19:20, 10 July 2024 (UTC)

Wen Wei Po

Should the outcome of this RfC, Wikipedia:Reliable_sources/Noticeboard/Archive_306#RfC:_Wen_Wei_Po, be reflected at WP:RSPSOURCES? - Amigao (talk) 01:54, 11 July 2024 (UTC)

Worldatlas.com

Is world atlas reliable? Used in a lot of articles but it feels a little clickbaity to me. — 48JCL 00:47, 11 July 2024 (UTC)

@48JCL, WP:RSN is the right place for that question. RSP is just a list of of sources that have been repeatedly discussed there. Gråbergs Gråa Sång (talk) 18:30, 13 July 2024 (UTC)

Thx — 48JCL 18:31, 13 July 2024 (UTC)

Can Rotten Tomatoes be used for birth info

Why can’t rotten tomatoes be used for birth info if the same birth info is used on IMBD Tnays20 (talk) 22:19, 17 July 2024 (UTC)

No. Neither RT nor IMDB should be used for birth dates. Firefangledfeathers (talk / contribs) 22:31, 17 July 2024 (UTC)

@Tnays20 IMDb itself is not reliable. (Please see WP:IMDB.) Finding something there does not help Rotten Tomatoes' credibility at all. @Firefangledfeathers is correct. Neither source should be used for any biographical details. Eddie Blick (talk) 00:54, 18 July 2024 (UTC)

But I was told by the user:Laterthanyouthink (talk) that he remembers an earlier talk page discussion about RT bios where the editors concerned agreed that it was acceptable for DOB. And that he added some notes and other sources on the talk page of the article. Tnays20 (talk) 01:32, 18 July 2024 (UTC)

Here's the RfC from last summer regarding Rotten Tomatoes.[1] The consensus was that it's okay to use for movie reviews and ratings as it's core purpose. However it's not a reliable source when it comes to biography details as it's not a journalism site and it doesn't provide any information as to how the material is obtained or verified. Which is a huge red flag when it comes to using such pages as a reliable source for BLPs. Kcj5062 (talk) 02:08, 18 July 2024 (UTC)

Recent IP edits

Are these 2 edits correct? [2] And should the page be protected? I'm not sure I'm happy with IPs editing it. This seems to be based on the short dicussion Wikipedia:Reliable sources/Noticeboard#TechTimes. Doug Weller talk 14:01, 25 July 2024 (UTC)

@Doug Weller: I don't think that's the link you meant to inlcude in your post. As to the edits I think the IP was trying to be helpful, if misguided. The original discussion (see Wikipedia:Reliable sources/Noticeboard/Archive 364#Tech Times (techtimes.com), iTech Post (itechpost.com), Gamenguide (gamenguide.com)) shows the sources isn't reliable, but I don't think there's any need to add it to RSP. The IP reverted themselves, and started a new discussion on RSN. -- LCU ActivelyDisinterested «@» °∆t° 15:08, 25 July 2024 (UTC)

@ActivelyDisinterested Great. That's what we need to happen. I think the link is the one I wanted, the IP's discussion.

Still, should IPs and new editors be editing such an important page / Doug Weller talk 16:49, 25 July 2024 (UTC)

The first link in your post (directly after Are these 2 edits correct?) leads to a shop listing for a 'ResMed ClimateLineAir™ 10'. -- LCU ActivelyDisinterested «@» °∆t° 16:55, 25 July 2024 (UTC)

Damn iPad. Sorry. Doug Weller talk 17:44, 25 July 2024 (UTC)

swentr.site (RT mirror)

The above site, brought to my attention by this paper, is an RT mirror so should be linked under RT's deprecated section. The report also lists a number of websites which seem to re-publish RT stories verbatim, so well worth a read and possibly including these in the list too. GnocchiFan (talk) 22:03, 28 July 2024 (UTC)

word missing in WP:THESUN

This deprecation does not apply to the broadsheet publication of the same name, that existed before the Murdoch from 1964–1969. Before the Murdoch paper, acquisition, ...? -sche (talk) 21:15, 29 July 2024 (UTC)

'the Murdoch' shouldn't be there, I'll correct it. When I added the sentence I rewrote it before publishing but apparently forgot to remove that fragment. -- LCU ActivelyDisinterested «@» °∆t° 09:23, 30 July 2024 (UTC)

models.com

• Wikipedia:Reliable_sources/Noticeboard/Archive_256#Reliability_of_models.com,_just_as_a_source_for_images_of_magazine_covers
• Wikipedia:Reliable_sources/Noticeboard/Archive_358#Models.com_for_1)_interviews_and_2)_ranking_lists
• Wikipedia:Reliable_sources/Noticeboard/Archive_432#Models.com_rankings

Is it worth adding? It's used quite a lot:[3]. Their about-page: [4]. Ping @AndyTheGrump if you have an opinion. Gråbergs Gråa Sång (talk) 10:38, 3 August 2024 (UTC)

I made my opinion on models.com entirely clear in the discussion of 'rankings' linked above. A single quote from their 'about' page seems quite sufficient to demonstrate why they cannot be seen as an independent source: By joining Models.com, you can build your brand and leave a lasting legacy: Create and manage your profile page, keep your work up-to-date for clients, magazines, and other talents to see, and claim your credits for extra visibility. Paid-for self-promotion, plain and simple. The website clearly needs to be deprecated as a source. AndyTheGrump (talk) 11:47, 3 August 2024 (UTC)

Instagram pages

I know for the most part social media pages are unreliable sources. But what if the page is a relative of the subject's? I'm asking because an Instagram page is being used as a source for actress Dara Renee's full name. The page seems to belong to her father.[5] Kcj5062 (talk) 02:27, 4 August 2024 (UTC)

It shouldn't be used unless by the post is by the subject of the article themselves, see WP:BLPSPS. -- LCU ActivelyDisinterested «@» °∆t° 07:43, 4 August 2024 (UTC)

Silver Bulletin

Is the Silver Bulletin, an election projection and analysis Substack website run by Nate Silver (the founder of and former employee at FiveThirtyEight, former correspondent to ABC News, named one of the 100 most influential people by Time for his successful election predictions) a reliable source? Personisinsterest (talk) 03:03, 9 August 2024 (UTC)

You should post this to the Reliable Sources Noticeboard, this talk page is for discussions on improving the perennial sources list. -- LCU ActivelyDisinterested «@» °∆t° 09:22, 9 August 2024 (UTC)

The Telegraph again

@Hemiauchenia, the way I understood the closure review decision, the entry should be reverted to the status quo ante. Is my understanding incorrect? I'm pinging @Compassionate727 as the reviewer. Alaexis_¿question? 20:30, 9 August 2024 (UTC)

This is not how RSN discussions or RSP entries work. The people arguing for this are inveterate wikilayers who are ignoring years of precedent regarding RSN discussions. Take for instance the famous 2020 Fox News RfC, which found that there was "no consensus" for the reliability of Fox News regarding politics and science, which was closed by a panel of admins (rather than unilaterally by a non-admin as with the Telegraph RfC and reclose). By the logic of "no consensus=remains reliable/no change in the RSP entry" employed here, the Fox News entry at RSP should not have changed, yet it was changed and the change was widely accepted at the time. Many entries on RSP have "there is no consensus on the reliability of X", and yet prior to the recent Telegraph RfC the concept of "no consensus" entries on RSP has to my knowledge never been challenged. Hemiauchenia (talk) 20:44, 9 August 2024 (UTC)

This particular issue isn't about how no consensus close should be interpreted, the review closer effectively reclosed the RFC. -- LCU ActivelyDisinterested «@» °∆t° 20:46, 9 August 2024 (UTC)

Which doesn't prevent an RFC on the underlying issue, right? Selfstudier (talk) 20:47, 9 August 2024 (UTC)

I've undone the addition. Nothing on Compassionate727 talk page shows they have withdrawn their close, and the close is still in place at AN. -- LCU ActivelyDisinterested «@» °∆t° 20:44, 9 August 2024 (UTC)

I have reverted the Administrator's noticeboard close as a fail of WP:BADNAC criterion 2: The outcome is a close call (especially where there are several valid outcomes) or likely to be controversial. Hemiauchenia (talk) 21:06, 9 August 2024 (UTC)

Funny that literally the entirety of the history of how RSN no consensus closes are done can be ignored just because of a controversial topic discussion happening. Someone needs to have the backbone to actually close with a determination on one side or the other or follow RSN rules on a no consensus close, because precedent is very blatant and obvious when it comes to RSN that no consensus means no consensus on reliability for the source in question. Since RSN is about affirmative reliability determinations, whether a source is reliable needs to be something we agree on. If we don't agree, then the source can't be considered WP:GREL by definition. Follow that link, it even says in the first line "Editors show consensus that the source is reliable in most cases on subject matters in its areas of expertise". Silverseren^C 21:12, 9 August 2024 (UTC)

• I've updated the notification at RSN that the close of the RFC review has been reverted, and added the same to Compassionate727's talk page. -- LCU ActivelyDisinterested «@» °∆t° 21:18, 9 August 2024 (UTC)

  The close revert has been reverted, discussion ongoing at Wikipedia:Administrators' noticeboard#Removal of closure Hemiauchenia (talk) 21:30, 9 August 2024 (UTC)

  The close review has been closed again, and the original RFC reopened. I've updated RSN with the details, and made a placed a new close request. -- LCU ActivelyDisinterested «@» °∆t° 16:57, 15 August 2024 (UTC)

Telegraph RfC and "no consensus = no change"

• 1. As a proposal, I think it's best not to include the The Telegraph RfC until matters are resolved. There has already been some good faith reverting from myself and @ObserveOwl, and while the close remains "active" as of writing, it doesn't appear worthwhile for inclusion until matters have been resolved. In hindsight, my initial summary including a quote from the closure wasn't the best idea.
• 2. I have opened a "Request for clarification" regarding the idea that an NC RfC close would mean no change in RSP, an argument that has been repeated numerous times in the general discussion. It's only open to those who promote these idea.

That's it. If you read that, I apologise for likely wasting your time. CNC (talk) 20:00, 9 July 2024 (UTC)

Since it is disputed with regard to LGBT issues should it be updated to yellow and additional considerations apply? PackMecEng (talk) 16:29, 25 July 2024 (UTC)

To clarify, The Telegraph remain GREL excluding trans topics, but otherwise would be MREL for trans topics based on 2024 RfC. See this edit as example. [6]. Ideally the close review would be closed first prior to inclusion. CNC (talk) 16:26, 1 August 2024 (UTC)

Unfortunately no, you are either generally reliable or additional considerations apply. So until the close review overturns it we go by the rfc and if you look at the extended close its clear on the subject. PackMecEng (talk) 16:38, 1 August 2024 (UTC)

Looking at the page that is clearly not true, many of the green entries have some sort of additional consideration attatched. Just going from the top we find Al Jazeera which is green with additional considerations, Amnesty International which is green with additional considerations, the Anti-Defamation League which is green with additional considerations, Aon which is green with additional considerations, Ars Technica which is green with additional considerations... I can keep going but its pretty clear that your suggestion that its a hard line is absolutely false on its face. Horse Eye's Back (talk) 16:47, 1 August 2024 (UTC)

For opinion sections sure, that is always the case and noted. Not for when whole topics that would normally be reliable are not. We change those to yellow. For example look at WP:ADLAS. PackMecEng (talk) 16:51, 1 August 2024 (UTC)

Maybe I missed something but the Telegraph is not a case of "whole topics that would normally be reliable are not" and your handwaving that these only address opinion sections appears fallacious. Horse Eye's Back (talk) 17:03, 1 August 2024 (UTC)

Also noting that in your example you list Anti-Defamation League which is currently yellow, as it should be. There is no additional consideration on Amnesty International. I think the difference you are seeing is Telegraph is specifically not generally reliable on those topics while the examples you gave that are still green say you should attribute them. Not the same thing. Also Ars Technica has no additional considerations so yeah its green. Even your Al Jazeera one does not list areas that it was not found reliable. I am not sure you are looking at the right list since all your examples are wrong... PackMecEng (talk) 16:56, 1 August 2024 (UTC)

Please take a minute to look at the sitution before responding, we have three different entries for the ADL... But even the green one describes additional consderations. If we want to jump down to Aon we find "Editors pointed out that Aon often provides data not found in other sources, and care should be taken when using the source as it may be providing a different estimate than other sources, e.g. total economic damages, rather than property damage." which is definitely an additonal consideration. Horse Eye's Back (talk) 17:03, 1 August 2024 (UTC)

Yes we have three, the one with additional considerations is yellow. I'm sorry I think you are just flat wrong on this. Yellow is specifically for additional considerations as laid out in the legend. The fact that others have not been updated just means they should be updated, not that we should not follow the guide and legend. PackMecEng (talk) 17:07, 1 August 2024 (UTC)

Ah yes... The "everybody is wrong except me, they will figure it out eventually" argument, very strong. Its simply not as rigid as you make it out to be. Horse Eye's Back (talk) 17:09, 1 August 2024 (UTC)

The original objection was to wait until the challenged close was finished. You are the one saying we shouldn't follow the legend or page above. So no idea what you are talking about. But seeing as you don't have an argument and I am sighting the guide page, maybe its you pulling it out their butt? But seriously if you have it documented anywhere, like I do, that it should be your preferred way I would love to see it. I found these random examples that are against the legend so I must be right is a bad argument. Just sayin... PackMecEng (talk) 17:13, 1 August 2024 (UTC)

And where does it define the difference between an additional condiseration which merits being yellow and an additional consideration which doesn't? You're the one saying that the community is incapable of following the legend or page above, you just argued that most of the green sections on the page should not be so and should be updated to yellow. I'm not arguing for radical change, you are... Horse Eye's Back (talk) 17:22, 1 August 2024 (UTC)

Actually I pointed out you were mostly wrong in your examples or why they stayed green. Color and symbol are covered by Wikipedia:MREL. I have no idea what you are arguing for honestly. PackMecEng (talk) 17:24, 1 August 2024 (UTC)

You didn't do that. I'm just describing reality, I'm not arguing. In theory I actually agree with you, but thats clearly not consensus... How much more obvious does it need to get than the PinkNews (green) entry "There is rough consensus that PinkNews is generally reliable for factual reporting, but additional considerations may apply and caution should be used. Most of those who commented on PinkNews' reliability for statements about a person's sexuality said that such claims had to be based on direct quotes from the subject." Horse Eye's Back (talk) 17:29, 1 August 2024 (UTC)

Editors closing RFC and those updating the RSP don't always do so in a standard manner, which leads to a lack of uniformity in how entries are maintained. For instance the entry for South China Morning Post is green but that additional considerations apply in certain areas, this comes direct from the close of the RFC. As the RFC was closed as 'reliable but additional considerations apply' that's how the entry is maintained, and it stands as noone has challenged it (even though neither the entry or the RFC close exactly match one of the standard results). This hasn't been an issue as previous contentious RFCs have had clearer results, but ultimately it doesn't matter as editors should be reading what the additional considerations are not simply checking the colour of the entry. -- LCU ActivelyDisinterested «@» °∆t° 18:08, 1 August 2024 (UTC)

Part of the nuance appears to be that generally reliable is shorthand for "Generally reliable in its areas of expertise" so we have the nebulous issue of discerning the difference between being ruled unreliable in a given area of expertise (a change in color, not GREL) and an area being ruled as not within their expertise (no change in color, remains GREL) Horse Eye's Back (talk) 18:21, 1 August 2024 (UTC)

The point is this update to RSP should be topic specific, similar to WP:ADL and WP:HUFFPOST. The Telegraph RfC wasn't based on the general reliability of the outlet, but instead topic specific. If you want to update RSP based on current consensus then at least do it properly. Per my example of Telegraph for transgender topics as MREL, but ideally with better wording. [7] CNC (talk) 19:33, 1 August 2024 (UTC)

From what I can tell that just muddies the water and is not really helpful. No need for it to have a separate section, just make it yellow and have the description. PackMecEng (talk) 14:09, 3 August 2024 (UTC)

No consensus versus mixed consensus

Following the close of the AN thread (which has attracted its own criticism at the closer's talk page), I can't help but wonder about the current "no consensus" label. It feels to me that label is really trying to do two things and that's where we get tripped up, the same way we would if we tried to smush generally unreliable, deprecated, and blacklisted (or even just these last two) into one category. Most of the time it's saying there is a mixed consensus about the source. That is Editor's generally agree its good in some ways and less good in other ways, so use it carefully. Perhaps "Use with care" could also work for this. And then there times where there is genuinely no consensus about a source, which is a whole different thing than a mixed bag, generally reliable, or generally unreliable. I'm wondering if this kind of split might help us land on a reaosnable outcome not only for the Telegraph but others. Best, Barkeep49 (talk) 16:14, 7 August 2024 (UTC)

I'll ask the same question I asked at the talk page discussion mentioned - what is the practical difference for those evaluating (potential) sources for an article between "consensus that additional considerations apply" and "additional considerations apply because there is no consensus it's generally (un)reliable"? If there is none then I don't see any value in a split. Thryduulf (talk) 16:24, 7 August 2024 (UTC)

The things in quotes aren't what I'm suggesting. I'm suggesting there is "editors can't agree on how to use this source" and "editors agree some uses of the source are appropriate and others inappropriate, so use with care". I think the labeling of "no consensus" is what has caused a fair amount of the agita here and even if that were just renamed editors would be able to see things differently in this instance. Best, Barkeep49 (talk) 16:28, 7 August 2024 (UTC)

Easy fix: "consensus that add'l considerations apply" should remain yellow, "no consensus about reliability one way or the other" should be blank (no color). To Thryd's question, the difference between the two is this: for yellow, editors are guided as to the reliability of the source by the add'l considerations described in the RSP entry; for a blank entry, RSP provides no guidance one way or the other as to reliability. Levivich (talk) 16:27, 7 August 2024 (UTC)

I wish I'd refreshed the page rather than finish writing what I had started because Levivich says it better than I did in my reply. Best, Barkeep49 (talk) 16:29, 7 August 2024 (UTC)

I support this. Chess (talk) (please mention me on reply) 16:30, 7 August 2024 (UTC)

That seems like a thing I could live with, I have always taken yellow as attribution in line for anything contentious. Selfstudier (talk) 16:32, 7 August 2024 (UTC)

I like this idea, but closers would have to be clear about exactly what they meant. "No consensus, but editors noted that attribution might need to be used" is ambiguous in whether that's an additional consideration or a general comment about the discussion. -- LCU ActivelyDisinterested «@» °∆t° 16:52, 7 August 2024 (UTC)

Yes, this is very sensible. If it’s open season on the colours, I’d be tempted to change green to blue to break the connotation of a traffic light rating system, which seems to cause editors to think things like “green is always strictly better than amber, context be damned”. Barnards.tar.gz (talk) 17:08, 7 August 2024 (UTC)

I'm fine with splitting this with the caveat that "blank" should mean gray or white, not "not present". If a source that is normally considered to be reliable (say, a WP:NEWSORG or a book published by an academic publisher) reaches a no consensus result at WP:RSP, the fact that there was a serious dispute about its reliability is important and should be noted at RSP. If it wasn't noted, we'd be back to "no consensus means reliable", which I strongly oppose. Loki (talk) 18:31, 7 August 2024 (UTC)

Grey is already used for blacklisted sites, but no colour could be a bad idea as it will appear black in dark mode. Although equally white could be taken as a mistake in dark mode, as if the background colour isn't being handled correctly. I like the blue idea, as it prompts editors to read what 'blue' means. -- LCU ActivelyDisinterested «@» °∆t° 20:53, 7 August 2024 (UTC)

I'd also be fine with blue, but I don't see how having the same color as a page's background would be bad. Aaron Liu (talk) 22:48, 7 August 2024 (UTC)

It's just about reducing chances for misunderstandings. My concern would be that a, admittedly tiny, fraction of editors would mistake the black background as the entry being black and therefore 'bad!'. Blue is an unexpected colour amongst the other entries, which helps to highlight it. -- LCU ActivelyDisinterested «@» °∆t° 23:55, 7 August 2024 (UTC)

If there is a prior consensus that a source was reliable, then a "no consensus" result should not change the fact there was a prior consensus that a source was reliable. I see no compelling reason that there should be an exception to "status quo" just for RSP. There can be consensus on adding additional considerations or a note about bias without there being a consensus on the reliability as a whole. That said, there should really not be many "no consensus" outcomes. RSN discussions should not be closed as "no consensus" just because a significant portion of people responding are ignoring the arguments and !voting based on their opinions. Arguments should be weighted as required by the policy, and after those arguments are weighted there should virtually always be clear consensus that either the source is unreliable or that the claims of unreliability were refuted. If the "side" arguing for unreliability did not manage to convince people of their claims, or they are disputed successfully, that should not result in anything other than "generally reliable". If the "side" arguing for reliability did not refute the claims of unreliability, then it should result in either a "considerations apply" or topic-area breakouts (a la Fox News) as the discussion merits.

To summarize, it appears some editors (both during the Telegraph discussion and discussions about this after the fact) wish to use "no consensus to do anything" as an excuse to "downgrade" a source. The solution there is to more carefully evaluate the discussions and their strengths in the discussion, and to discard opinions on the source that are just that - opinions. It's not to allow this sort of "backdoor downgrading" of a source's reliability just because enough people expressed their opinion without backing it up or managing to convince others. -bɜ:ʳkənhɪmez | me | talk to me! 16:43, 7 August 2024 (UTC)

The key question you and others have failed to answer is how can a result of "no consensus that a source is generally reliable" equal "there is a consensus that this sources is generally reliable", which is the definition of a generally reliable source? Thryduulf (talk) 16:46, 7 August 2024 (UTC)

Although I'm not necessarily in agreement with it, I can see the argument being stated as "If the prior consensus was that a source was generally reliable, then a new 'no consensus' result means there's no consensus to change that status". -- LCU ActivelyDisinterested «@» °∆t° 16:49, 7 August 2024 (UTC)

There is a difference between "no consensus that a source is generally reliable" versus "there was a previous consensus that a source is generally reliable". Consensus can change, but absent compelling evidence that there is now no consensus on the reliability, there is no compelling reason to violate maintaining the status quo that is present everywhere else on Wikipedia. If editors wish to dispute the reliability of the source that was previously considered generally reliable, the onus should be on them to make clear, compelling, and unrefuted claims of the unreliability that convince other editors of its unreliability. If they are unable to do so, they should not be able to "win" through a backdoor like this. -bɜ:ʳkənhɪmez | me | talk to me! 16:51, 7 August 2024 (UTC)

I don't see it as a backdoor, it is in effect a source downgrade (perhaps less so than Mrel) because we cannot categorically say Grel. Selfstudier (talk) 17:01, 7 August 2024 (UTC)

If the people arguing that it is not generally reliable do not manage to convince editors (to the point of there being a consensus that it is not generally reliable), then it is a backdoor downgrade. I'm not arguing that new discussions resulting in no consensus can't result in a "downgrade" - but if a source previously had consensus that it was reliable, and there is no consensus that is is unreliable, then the previous consensus should stand. Anything different is in effect a "backdoor" that is only open to "one side" - the side arguing unreliability. -bɜ:ʳkənhɪmez | me | talk to me! 17:04, 7 August 2024 (UTC)

You were not convinced but that doesn't mean that the consensus of editors was not convinced, no matter how many times you state and imply otherwise. Thryduulf (talk) 17:10, 7 August 2024 (UTC)

If there's no consensus, then editors were not convinced. Downgrading a previously reliable source should not be possible through a backdoor "no consensus" result. That is saying "no consensus" is a consensus. And that's simply untrue. -bɜ:ʳkənhɪmez | me | talk to me! 17:14, 7 August 2024 (UTC)

It is a consensus that it is not Grel presently. Further discussion can take place if things change (better or worse). Selfstudier (talk) 17:19, 7 August 2024 (UTC)

That’s simply untrue. No consensus means no consensus. It does not mean it is not generally reliable. There has still been no reason that this should be an exception from “status quo” applied everywhere else when no consensus arises in a new discussion. -bɜ:ʳkənhɪmez | me | talk to me! 17:45, 7 August 2024 (UTC)

We're back to my earlier question which you still haven't answered: how can a result of "no consensus that a source is generally reliable" equal "there is a consensus that this sources is generally reliable"? Thryduulf (talk) 17:52, 7 August 2024 (UTC)

I’ve answered it multiple times. If there is no consensus to change something or not change something, it does not change. This is the standard virtually everywhere on WP except here. There is no compelling reason it should not also be the standard here other than to give people a “backdoor downgrade” possibility if the discussion becomes so large/unwieldy that there is no consensus. To make it abundantly clear for you: If there was a prior consensus that the source is reliable, then future discussions need to be seen in that light. No consensus on the source being unreliable should not result in the downgrading of a source that previously had consensus that it is reliable. -bɜ:ʳkənhɪmez | me | talk to me! 17:55, 7 August 2024 (UTC)

That's a lot of words that you've said before that didn't answer the question then and don't answer the question now. The question asked is not "Is there a consensus to change the reliability?" but "What is the reliability of the source?". Why is it so difficult for you to understand that GREL is defined as "there is a consensus that this source is generally reliable" and that anything other than a consensus that is generally reliable means it is not generally reliable? Thryduulf (talk) 18:48, 7 August 2024 (UTC)

Articles are not the same things as RS and even there the status quo thing is slightly suspect. Selfstudier (talk) 17:55, 7 August 2024 (UTC)

My point stands that if a source had a prior discussion that it is generally reliable, and the claims presented by someone in a new discussion to claim “unreliable” are not sufficient for a positive consensus of unreliability, there should not be a “backdoor” to mark it as anything less than unreliable. Either the claims succeeded at convincing people and thus a consensus formed, or they did not, and there should be no change. -bɜ:ʳkənhɪmez | me | talk to me! 18:01, 7 August 2024 (UTC)

That argument is reversible tho, instead of the claims were not sufficient, the defense was insufficient. If the source were truly Grel, the defense would have succeeded. To me, nocon just means more discussion, probably not right away tho, meanwhile the status quo has effectively changed. Selfstudier (talk) 18:08, 7 August 2024 (UTC)

The defense was not insufficient. The defense was sufficient to convince enough editors that there was “no consensus that it is unreliable”. Combine that with the prior status quo, and you have “generally reliable”. There is no reason this should be an exception to the status quo that applies everywhere else. Either the claims were sufficient for consensus or they were not. If they were not sufficient for an affirmative consensus, the status quo should not change. -bɜ:ʳkənhɪmez | me | talk to me! 18:15, 7 August 2024 (UTC)

I give up, cannot defeat repetition. Selfstudier (talk) 18:17, 7 August 2024 (UTC)

Because of the importance of WP:V to this project, I think it's very important that the usual "status quo holds" rule does not imply to WP:RSP. Otherwise, a source going downhill could have years to spread misinfo on Wikipedia even after its decline became obvious to many editors but before other editors got over its past reputation. Loki (talk) 18:25, 7 August 2024 (UTC)

There is a difference between “consensus that the source used to be reliable and now is not” (as can be seen with Cnet/etc) versus “consensus that the source was reliable, and no consensus now to ‘overrule’ that prior determination”. As is shown, when a source actually begins to publish misinformation, RSN is fairly quick at a clear consensus to change its reliability. What you fail to accept here is that the Telegraph did not publish misinformation - they published information you didn’t like and/or published attributed information that was later shown to be inaccurate but was reported on as someone’s opinion - at worst. If a source was previously determined to be reliable, then no - editors shouldn’t be able to censor it just because they get enough people yelling for it to be censored for publishing biased information or information they don’t like. -bɜ:ʳkənhɪmez | me | talk to me! 18:41, 7 August 2024 (UTC)

When we're talking about RSes, every discussion is an independent question. If discussion A says that a source is reliable, and discussion B doesn't, what happened in discussion A is and should be irrelevant.

This isn't the same as for text changes: if RFC A reaches a consensus on phrasing, and RFC B neither supports or overturns that consensus, then it still needs to be phrased some way, and the best "some way" to phrase it as a default is what's currently in the article. Loki (talk) 18:48, 7 August 2024 (UTC)

You are arguing this yet you still have provided no actual reason that the “status quo” (which would result in A remaining the phrasing, since B did not come up with a consensus to overturn A) should not apply here other than that you don’t like it. -bɜ:ʳkənhɪmez | me | talk to me! 18:51, 7 August 2024 (UTC)

We have repeatedly explained why RSP is different: The question asked is not the one you think was asked, and the explicit wording of the definitions. At this point it's clear you simply aren't listening. Thryduulf (talk) 19:14, 7 August 2024 (UTC)

You’ve explained why it is different. Not why it should be, and not why a backdoor way to “downgrade” sources should be permitted and is beneficial to the encyclopedia by allowing people to censor viewpoints they don’t like. -bɜ:ʳkənhɪmez | me | talk to me! 19:15, 7 August 2024 (UTC)

If you want to change the way RSP has worked for years you need to get consensus to do so, and unless and until you do then the outcome of discussions must reflect the existing definitions used. This is not a "backdoor way to downgrade sources" it's doing things in accordance with the explicit definitions presented front and centre on the page. Closer to reality is describing your preference as a backdoor way to avoid downgrading sources you like - I'm not saying that is your intention but such an intention would be entirely compatible with the arguments you are making. Thryduulf (talk) 19:30, 7 August 2024 (UTC)

RSP is a summary of editors' opinion and discussions on sources. To summarize a nocon result, which indicates concerns, as the previous status is disingenuous. Aaron Liu (talk) 22:49, 7 August 2024 (UTC)

It's not to summarize a nocon result - it's summarizing years and multiple discussions that resulted in a consensus as "this source is questionable" just because, even though nothing has changed, enough people yelled on one side of the issue that it made it "no consensus". That is, in fact, why "status quo" is the standard virtually everywhere else on Wikipedia. I stand by my view that there should not be a "backdoor" to downgrade sources in this manner just because people finally got tired of dealing with the bludgeoning from the "sources I don't like should be prohibited" side that tends to be "louder" in these sorts of discussions. If there is a true no consensus on the first discussion about an issue, sure, that can be noted as such. But when there are years of past consensus, one flimsy no consensus result should not result in a backdoor downgrade without an affirmative consensus to downgrade it. -bɜ:ʳkənhɪmez | me | talk to me! 22:53, 7 August 2024 (UTC)

enough people yelled on one side of the issue

Yeah, that's why it should be reflected. Not to mention that louder is not supposed to be how consensus is to be weighed. Knocking down a pillar and that pillar's sizable supporters to enforce one's own beliefs is how villains work.

What happens after a NOCON is the "safe" option. For most scenarios, that's the status quo; for living people, that's removal of disputed material to discourage external pressure; for RSP, that's appropriately displaying the concerns. Aaron Liu (talk) 22:57, 7 August 2024 (UTC)

I don't disagree with appropriately displaying the concerns, but marking as WP:MREL endorses that there are concerns and they are valid. That's what "additional considerations" means, and I believe we disagree on whether that distinction is necessary. Chess (talk) (please mention me on reply) 02:03, 8 August 2024 (UTC)

If a discussion is closed as NOCON, then yeah, the close endorses that there are concerns and they are valid. Aaron Liu (talk) 15:56, 8 August 2024 (UTC)

(edit conflict) RSP does not "summarise years and multiple discussions" it documents the current opinions regarding the sources reliability as of the most recent discussion. This is because the reliability of sources changes. Also, please stop repeating your rhetoric about this being a "backdoor downgrade", it's been explained to you enough times now that it is incorrect and repeating it will not change that. Thryduulf (talk) 22:59, 7 August 2024 (UTC)

even though nothing has changed -> Well, that's clearly the issue, isn't it? If there was a consensus, and now there is no consensus, something has changed. Loki (talk) 23:23, 7 August 2024 (UTC)

It might be worthwhile to have an RFC on whether 'no consensus' equals MREL, or if there has to be a positive consensus that a sources is MREL. Simply as a way of having a fixed consensus in case of future disagreement. -- LCU ActivelyDisinterested «@» °∆t° 16:45, 7 August 2024 (UTC)

Per the very long-standing definitions at RSP, MREL is both. Feel free to start a discussion if you think that should be changed, but no discussion is needed to determine what the status quo is. Thryduulf (talk) 23:01, 7 August 2024 (UTC)

I would agree, but that it has caused contention is undeniable. RSN and RSP seem to have many detractors recently, showing that common practice has community support can't hurt. -- LCU ActivelyDisinterested «@» °∆t° 00:00, 8 August 2024 (UTC)

I don't think that works because no consenus is more a spectrum than a discrete block... There being no consensus between 1 and 2 is not the same as there being no consenus between 3 and 4. If we can't decide whether something is unreliable or deprecable upgrading it to MREL based on that lack of consensus makes absolutely no sense whatsoever. Horse Eye's Back (talk) 00:27, 8 August 2024 (UTC)

What has been the precedent in the past? Including cases in the other direction?

In past cases, when a source had a prior consensus of GUNREL, if a subsequent discussion resulted in "no consensus", did that result in a change in the status quo of the RSP listing? Surely, the result of "no consensus" for a status quo GREL source should be treated the same way as a "no consensus" for a status quo GUNREL source, right?
Incidentally I can only recall one such past case (Wikileaks), and that resulted in "no change to status quo", therefore still GUNREL. (The dispute in question was over something ridiculously mundane, too. It was about whether or not some member of the Vietnamese communist party held some leadership position in a particular year. Like, who would care enough to make up stuff about something so boring!) Anyway, we can't do one thing some times and another other times, nor should we defer to the individual closers to apply the amswer to this question however they wish. Nor can there be one rule for GREL sources and a different rule for GUNREL sources: it must be either status quo for both or a change to "no consensus" for both.
One last point: "No consensus of general reliability" is NOT "Consensus of no general reliability". The difference is in the case of the latter, there is general agreement between editors, and for the former there is not. Plus the former also means "no consensus that it is NOT generally reliable" and you cannot have consensus and no consensus for the same thing at the same time. That would be nonsense. But it's an easy mistake to make, and I don't at all doubt that Selfstudier made the argument in good faith. 73.2.106.248 (talk) 01:22, 8 August 2024 (UTC)

This summarizes exactly what I've been saying about it being a "backdoor" so well. People use "no consensus on general reliability" to mean that there is a consensus that it should be "less than generally reliable" when listing at RSP. The status quo should be maintained in no consensus. If "no consensus" turns into "no consensus of general reliability" and that is noted, then it is being listed as a consensus that it is not "generally reliable", which cannot be the case if there was "no consensus". -bɜ:ʳkənhɪmez | me | talk to me! 01:25, 8 August 2024 (UTC)

@73.2.106.248 I don't recall ottomh any examples of previously GUNREL sources having a discussion that resulted in no consensus between Generally Reliable and Generally Unreliable. If such a discussion were to occur though then MREL would be the right outcome because there is no consensus that it's unreliable and also no consensus that it is reliable.

Discussions have closed with a lack of consensus between generally unreliable and deprecate. Those have (afaik always uncontroversially) resulted in a status of generally unreliable, partly because deprecation requires as much of an active consensus as generally reliable does and also because deprecation is just a stronger degree of unreliability but generally reliable and generally unreliable are opposites.

Generally reliable can be though of as assigning a score of 1, MREL a score of 0, GUNREL -1 and deprecation -2 on an integer scale. You can see that 0 is between 1 and -1, but there is no option between -1 and -2. It's also clear that -1 to -2 is just a degree of the same thing while -1 and +1 are very different (FWIW Blacklisting doesn't really sit on this scale, it's something that's applied additionally to (in theory) sources of any rating that have been significantly abused (and/or have themselves been the abuser) although in practice only deprecated or GUNREL sources have done this).

I shan't bother wasting my breath refuting berchanhimez's repetition of falsehoods yet again. Thryduulf (talk) 01:55, 8 August 2024 (UTC)

To say that a MREL sources is less reliable than a GREL source is overly simplistic. In most cases MREL mean that certain additional considerations apply, these may or may not be directly related to reliability in general or concerns about a certain context. Only by reading the details can that be discerned.

In this specific issue the Telegraph is considered generally reliable, but in the specific context of trans issue it's reliability is contested. Nothing in that says that the Telegraph is unreliable in any context. -- LCU ActivelyDisinterested «@» °∆t° 15:27, 8 August 2024 (UTC)

• At the very least, we should link to the RFC itself; its closure and impact are still being disputed and will probably take a while to fully hash out, but no matter how things end it is a relevant discussion given its massive attendance and fallout. Removing it doesn't make sense - RSP lists all significant discussions, including ones that are much more narrow and specific than this; inclusion in that part of the template doesn't carry any implications about the meaning of the RFC or its results, it just provides a courtesy link so editors who want to review prior discussions on a source before starting a new one can easily access them all in one place. ie. no matter what, if a new editor arrives on this page and is like "hey, I have an idea, let's discuss whether the Telegraph is reliable about trans issues, that doesn't seem to have been discussed lately!", that list of links makes it easy for them to see that, no, whatever the outcome, we just discussed that, it was a whole thing. --Aquillion (talk) 21:16, 13 August 2024 (UTC)

  How about we put it in a note? I agree now that it should be included somewhere, but I disagree that it should be listed as part of the discussions on why the Telegraph is generally GREL. I also think that this shouldn't count to its "last discussed" date. Aaron Liu (talk) 21:21, 13 August 2024 (UTC)

  Assuming S Marshall's close is not overturned there are two possible ways that this could be noted at RSP. Either the Telegraph's entry is split into two sections - "Daily Telegraph (except trans issues) - GREL" and "Daily Telegraph (trans issues) - MREL"; or the single entry is maintained with a note that it's "generally reliable, but additional considerations apply for trans issues" (but not that exact wording). In the former case (my preference) this RFC would definitely be listed in the second section but could also be listed in the first (as a discussion broadly confirming it GREL for non-trans issues). In the second case it would need to be listed in the single section. It will count as the last discussed date in whichever section(s) it's mentioned in because that's simply factual. If someone wants to start a discussion about some other aspect of the Telegraph's reliability (American celebrity culture for the sake of argument a topic area I have zero interest in and no knowledge of the Telegraph's reliability for) then they'd just start their section with "The most recent discussion (July-August 2014) didn't cover the reliability of American celebrity culture, it hasn't been discussed since [link]", which is preferable to the commenters saying "Didn't we just discuss the Telegraph?" and the OP not being aware of that discussion (trans issues not being relevant to what they want to talk about). Thryduulf (talk) 21:38, 13 August 2024 (UTC)

  Definitely the former case as is precedent We were discussing what to do while the entire row is in progress. I'll implement the note thing now Aaron Liu (talk) 00:07, 14 August 2024 (UTC)

Context

This drama is all about the Daily Telegraph and whether it's a reliable source about trans people. There's a broad consensus that the Daily Telegraph is, in general, a newspaper of record and a reliable source, and we've tested that consensus several times. Whether it's a reliable source about trans people is disputed, and there's no consensus among Wikipedians whether it is or isn't.

WP:RSP needs above all to tell the truth about what the consensus is. Therefore it should say that there's no consensus about the Daily Telegraph's reliability on trans issues. QED.

It doesn't matter whether it says so on a yellow background or a plain white one. It doesn't matter what the precedent is. What matters here is whether we can trust our information pages.—S Marshall T/C 07:38, 8 August 2024 (UTC)

Exactly this. Thryduulf (talk) 09:34, 8 August 2024 (UTC)

I think we all agree that it should say that there's no consensus about the Daily Telegraph's reliability on trans issues... The question seems to be what color the box is (yes I know its odd but box color seems to matter much more to some editors than what it actually says in the box). I think we're also painted ourselves into a bit of a corner by expecting papers of record to be reliable sources about everything beyond MEDRS, in reality they all have strong areas of expertise and weak areas of expertise like any other source. Horse Eye's Back (talk) 17:08, 8 August 2024 (UTC)

I do not think that Berchanhimez would agree with that. I think he'd say the discussion was no consensus but the Telegraph should be WP:GREL anyway.

Not that I think his opinion is relevant or according to policy here, but I do want to note that he's very likely to come in and express it loudly anyway. Loki (talk) 17:33, 8 August 2024 (UTC)

Maybe the easiest thing is just to reassess The Telegraph, their treatment of gender issues related to the Olypics have been shocking and in terms of mixing editorial POV with factual reporting go further than anything presented previously (to the point where I'm personally beginning to reconsider it as a reliable source at all), its getting bad[8][9] etc. Horse Eye's Back (talk) 18:14, 18 August 2024 (UTC)

I don't think there's any appetite for it but I also agree that those two articles are pretty egregious. (Though, I think it's pretty obvious that even though Khelif is cis they'd be covered by "trans issues".) Loki (talk) 18:25, 18 August 2024 (UTC)

All depends on your perspective (The Telegraph would say they're covering women's issues not trans issues), but I think this all falls into a bucket which could be called gender issues without signficant objection. Horse Eye's Back (talk) 18:32, 18 August 2024 (UTC)

Well, there’s Wikipedia:Broadly construed. Aaron Liu (talk) 18:36, 18 August 2024 (UTC)

Yes, ultimately the accusation of them being transgender was levied in the article, so under broadly construed it falls within the purview. Raladic (talk) 19:13, 18 August 2024 (UTC)

• So we could, in fact, add that on a white background and call it done?—S Marshall T/C 06:51, 9 August 2024 (UTC)

  I'd be for it. Don't know if it's feasible to do only that post close review, but I would support the eventual close being basically that. Loki (talk) 17:35, 9 August 2024 (UTC)

How suitable is Dave Leip's Atlas of U.S. Presidential Elections for a FLC?

Links: Dave Leip's Atlas of U.S. Presidential Elections, User:OlifanofmrTennant/sandbox/Oklahoma pres elections, United States presidential elections in Oklahoma.

I opened up the Oklahoma elections page and found it quite unusable and unsourced. I opened a sand box using the table from United States presidential elections in Arizona and began replacing the data with proper sources. I got to the 2000 election and was looking for a suitable source when I found the site. I did some digging and decided to ask about it. Questions? four Olifanofmrtennant (she/her) 18:32, 19 August 2024 (UTC)

Questions like this are better suited to WP:Reliable sources/Noticeboard. I would have personally be wary of the site, but the WP:USEBYOTHERS shown in it's wiki article is a strong indication of its reliability. -- LCU ActivelyDisinterested «@» °∆t° 22:58, 19 August 2024 (UTC)

My mistake I got the two confused. Questions? four Olifanofmrtennant (she/her) 01:47, 20 August 2024 (UTC)

GB News inclusion

I added GB News due to multiple discussions as well as consensus it was unreliable. [10] This was then reverted [11] based on discussion elsewhere with the edit summary "It is not a perenial issue, it has only be dicsues twice in 20 years (or more". In fact it has had 3 discussion (with one active) within 3 years (since launch), which would very much be considered regular and multiple discussions based on RSPCRITERIA. The user appeared to think that because it's GUNREL, it could be deleted without discussion? CNC (talk) 12:54, 18 August 2024 (UTC)

I can find 3 discussions of GB news (including the current one), one started by a sock (and closed as such). I disagree this constitutes " two or more significant discussions about the source's reliability in the past, or an uninterrupted request for comment on the source's reliability that took place on the reliable sources noticeboard.". Slatersteven (talk) 13:15, 18 August 2024 (UTC)

Per WP:RSPCRITERIA, as a continuation from your select quote, as your interpretation of significant appears to be the issue here:

"For a discussion to be considered significant, most editors expect no fewer than two qualifying participants for RSN discussions where the source's name is in the section heading, and no fewer than three qualifying participants for all other discussions. Qualifying participants are editors who make at least one comment on the source's reliability."

Are you suggesting the discussion created by the sock included fewer than two qualifying participants, or was off topic? Or that the most recent discussion involves less than 2 participants and wasn't about reliability of the source? I assume we can agree on the section headers at least? If not obvious here, the significance is based on the basic threshold of a discussion on the reliability of the source, not the subjective important of such as discussion. Finally, would you consider three discussions not multiple? CNC (talk) 13:28, 18 August 2024 (UTC)

No I am saying its close reason does not mean it was significant, as it was not a valid discussion. As to the recent one, its is current not "in the past". But I have raised my objections, and will not go round in circles, time for others to chip in. Per the below, we have had one actual discussion in the past that came to a conclusion (not RS). Slatersteven (talk) 13:33, 18 August 2024 (UTC)

I won't bother arguing with your interpretation then, given it doesn't reflect RSPCRITERIA. You should know that other "bad" RfCs & discussions are otherwise included in this list, even if not under the RfC categorisation, as long as the discussion is on-topic etc.

For reference sake, as I see this is your first edit to this list and probably it's not obvious; per the legend, discussions that are in progress are also added with an active parameter, creating the italic styling of the link, to signify an active discussion. By past discussion I meant that discussion has occurred (past tense), even if still inprogress. CNC (talk) 13:54, 18 August 2024 (UTC)

Depends, GB News is okay for general stories, but their problem is when it comes too opinion pieces, I would place it at MREL. General news reports are okay for citation but it maybe recommended to use a more reliable source. Then also point out, the case of reporting when it's a GB News opinion piece, etc. Reliable for some stuff, maybe not for other. Anyway, that's my two cents. Govvy (talk) 13:30, 18 August 2024 (UTC)

It sounds like this opinion is better placed in the discussion about GB News on reliability over at RSN. This discussion is intended to be based on it's inclusion or lack of at RSP. Unless of course this opinion is based on the interpretation of the 2022 RfC? To me it seems like that discussion was snow consensus for GUNREL, but each to their own. CNC (talk) 16:40, 18 August 2024 (UTC)

The 2022 RfC was a bit irregular as there had been no previous discussion. Some editors questioned it and as far as I understand it was never closed. I think the right thing is to hold a new RfC in case there are issues with this source. Alaexis_¿question? 14:32, 18 August 2024 (UTC)

Correct it was never closed. The interpretation up until now has been GUNREL, at least based on previous RSP inclusion and interpretation by others. Ideally this RfC would be formally closed and discussion could continue based on the current consensus. This could help to avoid an unnecessary RfC if there is lack of opposition to this categorisation (as there has been in the past). I'm not sure how long GB News was included in RSP without any opposition at RSN or otherwise, does anyone know? This would be useful information, given the RSP entry wasn't removed due to inaccuracy, but simply because there was only one discussion listed. CNC (talk) 16:47, 18 August 2024 (UTC)

GB News should be included. It has been discussed sufficiently considering how new it is. Cortador (talk) 12:49, 20 August 2024 (UTC)

What does Uses section in the table mean?

What does the Uses section and then 1, 2, 3 in the table mean? NamelessLameless (talk) 06:18, 5 August 2024 (UTC)

If you click on the number it will take you to search results of Wikipedia articles using the source. -- LCU ActivelyDisinterested «@» °∆t° 09:07, 5 August 2024 (UTC)

Oh. NamelessLameless (talk) 21:19, 29 August 2024 (UTC)

AntWiki reliability

@YoungForever has mentioned that AntWiki is not a reliable source and should not be used to cite Wikipedia articles due to it being user-generated. However, I believe that it can be used because contrary to Wikipedia which can be edited by everyone, it can only be edited by ant experts confirmed by administrators as stated on its website. See full discussion on User talk:YoungForever. Should it be considered reliable or not? 2003 LN6 07:04, 2 September 2024 (UTC)

This page is for discussing changes to the perennial sources list, you should post this to the reliable sources noticeboard. -- LCU ActivelyDisinterested «@» °∆t° 10:52, 2 September 2024 (UTC)

Done. 2003 LN6 16:01, 2 September 2024 (UTC)

Cartoon image near lead

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.

---

Current leadimage

Alternative?

Another option

Yet another option

I removed the cartoon image near the lead. I think this image is sarcastic and distracting to the point of this article, and takes up a lot of space, especially when you read the page from a phone. Per WP:ONUS, The responsibility for achieving consensus for inclusion is on those seeking to include disputed content., so I have removed the image for now. I think it might be better positioned in the beginning of the "Legend", but personally I do not think it is beneficial to the article, even if it has been on this page for 5 years as mentioned by another editor. Iljhgtn (talk) 12:33, 7 May 2024 (UTC)

I don't think it's distracting for the point of this project-page, and since it's not an article, there aren't really any rules apart from consensus. Until I see an alternative I like better, I'm at keep it there. Previous discussion at Wikipedia_talk:Reliable_sources/Perennial_sources/Archive_6#reliable_sources. Gråbergs Gråa Sång (talk) 13:09, 7 May 2024 (UTC)

I think it's a great example of the dangers of poor sources. American Apple Orchards PAC aka O3000, Ret. (talk) 13:45, 7 May 2024 (UTC)

I don't particularly care for the image either and would be fine with its removal. Some1 (talk) 23:13, 7 May 2024 (UTC)

It's not all serious, we're allowed a little bit of humour. Keep or replace with the xkcd. Anarchyte (talk) 09:15, 8 May 2024 (UTC)

This xkcd image is far better than the current image. I would support that as a replacement if outright removal of the current cartoon does not reach consensus. Iljhgtn (talk) 20:19, 1 June 2024 (UTC)

The image is perfect as is and illustrates clearly that not all sources are equal. Margaret Hamilton next to a stack of code illustrates nothing, and the XKCD one illustrates that sources are needed, not that reliable sources are needed. Headbomb {t · c · p · b} 09:43, 8 May 2024 (UTC)

But she has surely assembled the good sources! ;) That said, you make sense to me. Gråbergs Gråa Sång (talk) 09:49, 8 May 2024 (UTC)

I agree with Headbomb that neither of the proposed alternatives are better. I don't have any particular attachment to the current image so if someone can come up with a better alternative I wouldn't mind changing it, but neither of these are. Caeciliusinhorto-public (talk) 09:51, 8 May 2024 (UTC)

I think no image is the best. Iljhgtn (talk) 12:39, 8 May 2024 (UTC)

I like having an image, so long as it's material to the text. The use of an image, especially one as pertinent as this one, sets up the content in a helpful way.

The four-panel cartoon is is the only image identified that is material to the text it accompanies: it shows the use of sources, but also that not all sources are reliable. That's the page's entire point.

The Hamilton photo is a great photo, for Margaret Hamilton (software engineer); not so much for this page. It says nothing at all about sources or their quality (except perhaps a pun about source code, which would be too arcane for a large number of non-programming Wikipedia editors who are the target audience for this page). The XKCD cartoon is is about the need for sourcing, and would be fine for WP:Reliable sources or WP:Citation needed; but it makes no point about the reliability of sourcing, so does not carry the point here. So far, the four-panel cartoon is the only one identified that is apt for this page. TJRC (talk) 16:09, 8 May 2024 (UTC)

• Support removal of the cartoon. Ridicule is not part of the scientific method. The kind of mockery that the cartoon engages in is not helpful in discouraging pseudoscience: [12]. James500 (talk) 11:51, 10 May 2024 (UTC)

  RSP isn't very much about fighting pseudoscience, is it? Gråbergs Gråa Sång (talk) 12:20, 10 May 2024 (UTC)

  It's very much about fighting it, though it's not its exclusive mission (that would be fighting bad information in general, of which pseudoscience is part of). But this page is also not about changing minds. I couldn't care less about the feelings of WP:LUNATICS offended that no one is taking seriously their claims that water has memory, or that CNN is controlled by reptilians.

  What it's about is explaining and documenting that not all sources are equal. And the cartoon explains that very succinctly. Headbomb {t · c · p · b} 04:17, 15 May 2024 (UTC)

  The cartoon does exactly what @James500says it does which is antithetical to the headline, "The key to fighting pseudoscience isn’t mockery—it’s empathy". The cartoon could not be more mocking if it tried. Thankfully, we are free to remove it in the absence of evident consensus for the image being retained. Iljhgtn (talk) 20:16, 1 June 2024 (UTC)

  "The key to fighting pseudoscience isn’t mockery—it’s empathy" That's one opinion, and it certainly isn't universal. Moreover the point of the cartoon is to illustrate that not all sources are equal, and that if you bring a non-reliable source, people will ignore it, and rightfully so. Headbomb {t · c · p · b} 21:03, 1 June 2024 (UTC)

  The cartoon does not explain "that not all sources are equal". The cartoon does not say that any sources are reliable. The cartoon does not explain which sources are unreliable, or what they are unreliable for, or why they are unreliable. The cartoon does not say whether there is any difference between being unreliable and being less reliable, and does not admit that uncertainty sometimes exists. The cartoon does not explain that a source may be reliable for some things even though it is unreliable for others. The hypothetical source described in the cartoon bears no resemblence to most of the sources listed at RSP. There are no ancient books that claim that apples cause cancer. RSP should not include jokes about the World Elephant or (even if only by implication) the World Turtle. I do not think that showing that cartoon to someone who "brings" an unreliable source is going to help to convince them or anyone else it is unreliable. I think it is more likely to cause offence and prolong disputes. And individuals who cannot be reasoned with need to be blocked, not insulted with a cartoon. James500 (talk) 01:16, 27 June 2024 (UTC)

  @James500 makes some excellent points about how the cartoon is actually potentially offensive, in addition to being unnecessary and in no way actually clearly didactic. I had not considered the offensive nature of the cartoon, and it is with this added reason that I see the building consensus for removal is becoming even more compelling than I had initially expected. Iljhgtn (talk) 04:15, 1 July 2024 (UTC)

  It is a good illustration for WP:AGEMATTERS. For an article about perennialy unreliable sources, not so much. Would it be a COPYVIO to use the front page of this article from The Sun (UK, Murdoch stable, now there's a surprise): Freddie Starr ate my hamster. Fair use? It is the archetypal example regularly cited in UK commentary, though there are more egregious examples. --𝕁𝕄𝔽 (talk) 09:33, 15 May 2024 (UTC)

  Not a bad idea, but would fail the strict demands of WP:NFCC/WP:NFCI. Got anything good pre-1929? Gråbergs Gråa Sång (talk) 09:41, 15 May 2024 (UTC)

  I found a sort-of-alternative which I added, but IMO it doesn't really fit. This is more an example of "Even generally reliable sources sometimes get it wrong." I wonder if they published a redaction? JMF's example is a much better fit, arguably even fake news per "The man behind the hamster story was the British publicist, Max Clifford, at that time Starr's agent, who concocted the story as a practical joke." Gråbergs Gråa Sång (talk) 10:18, 15 May 2024 (UTC)

  I'm not convinced it's really illustrating WP:AGEMATTERS. Yes, the books is described as ancient, but "apples cause cancer" isn't a previously-respectable but outdated theory.

  As for a non-free image: per WP:NFCC#9, non-free images are only permitted in articlespace. Suggesting non-free images here is an absolute non-starter. Caeciliusinhorto-public (talk) 09:41, 15 May 2024 (UTC)

  Quote from book: "Put ye sliced apples on a beach, and behold, crabs shall appear!" Gråbergs Gråa Sång (talk) 09:52, 15 May 2024 (UTC)

  I do not see how WP:AGEMATTERS really applies here as well. Also, it looks like some people want the image removed, some don't really care, others want a new image, but there is no consensus to keep the existing image. Iljhgtn (talk) 20:13, 1 June 2024 (UTC)

  I see strong support for keeping it from:

  • Objective3000 "American Apple Orchards PAC aka O3000, Ret"
  • Headbomb
  • TJRC

  And weak support from:

  • Gråbergs Gråa Sång
  • Anarchyte
  • Caeciliusinhorto-public

  Solomon Ucko (talk) 20:29, 1 June 2024 (UTC)

  No consensus to change = no change. Basic wikipedianism. Gråbergs Gråa Sång (talk) 05:17, 2 June 2024 (UTC)

Support removal and inclusion of second alternative. It's supposed to be a witty cartoon, but really isn't that funny or relevant to the source list that intends to be a more formal summary of source reliability. The cartoon in question would be better suited for WP:RS or WP:FRINGE rather than this list of perennial sources, specifically the latter. The second alternative (Margaret Hamilton) is a lot more relevant, as it indirectly represents the concept of "stacking" source discussions from an enormous archive, which is very much reflective of this project page and purpose. Will have a search on commons and see if I can find any more alternatives. CNC (talk) 12:22, 27 June 2024 (UTC)

Support removal. As the kids say, the current image is pretty cringe. I'm not sure having an allegedly funny image in the lede contributes anything. Apocheir (talk) 16:56, 27 June 2024 (UTC)

Comment None of the alternative images seem compelling. Margaret Hamilton standing next to papers doesn't tell us anything. The Wikipedian protestor comic is actually kind of funny, but it's about a lack of sources at all rather than unreliable ones. The Dewey Defeat Truman press photo is pretty exclusionary to anyone not well versed in midcentury US presidential politics. Bremps... 03:28, 1 July 2024 (UTC)

I would say that is why outright removal, without a replacement, is looking like the best option supported by the largest consensus at the present time. Iljhgtn (talk) 04:17, 1 July 2024 (UTC)

Support removal. None of the three images are clearly relevant to this particular page rather than, say, WP:RSN or many other source-related guidance pages. I support removal without replacement. Also, removal will reduce page size... a bit.--FeralOink (talk) 06:26, 12 July 2024 (UTC)

Oppose removal. The cartoon we have currently is the best of the four choices presented, and does a good job of showing that not all sources are reliable. Enoryt nwased lamaj (talk) 08:28, 8 August 2024 (UTC)

I don't think any of the suggested alternatives are better. FunLater (talk) 20:44, 29 August 2024 (UTC)

Do you think it is best to leave the current image though? Or remove it and replace it with nothing? I support removing the current image which I think badly promotes a negative view, and replacing it with nothing. Iljhgtn (talk) 22:10, 30 August 2024 (UTC)

Support removal The "flat earth" comment reinforces the false notion that ancient civilizations all thought the earth was flat. See Flat Earth#Greece: spherical Earth. Ancients didn't know about cancer so it seems unlikely that an ancient book would claim that apples cause cancer. See this review of studies about whether Apples causes cancer:

Foods of plant origin have received particular interest over the years as potential cancer-preventive components of a healthy diet. Fruit and vegetables contain a myriad of bioactive phytochemicals that, through various molecular mechanisms, show chemopreventive properties in both in vitro and in vivo models of carcinogenesis( 4 ). However, from an epidemiological point of view, while early data from case–control studies suggested a clear preventive role for fruit and vegetables on cancer in different sites, recent large prospective studies have questioned this conclusion( 5 ). Indeed, while an expert panel report from the World Cancer Research Fund/American Institute for Cancer Research published in 1997 stated that there was ‘convincing’ evidence that a high intake of fruit and/or vegetables prevents cancers, an updated report published 10 years later downgraded the evidence to either ‘probable’ or ‘limited-suggestive’( 6 ).

Because of the peculiar chemical composition and the potential molecular mechanisms involved, it is possible that some types of fruit/vegetable may be much more strongly associated with cancer risk than others. This may be hidden in epidemiological studies examining the association of cancer risk with total fruit/vegetable intake. In this respect, our interest was attracted by apples considering that they are the most consumed fruit in European countries and they are a rich source of bioactive phytochemicals (phenols and flavonoids) possessing strong chemopreventive and antioxidant activities( 7 ). We therefore conducted a systematic review of the literature on the relationship between apple intake and cancer risk, and for the first time undertook a meta-analysis to provide quantitative estimates of the association.

The meta-analysis here found 41 previous studies on whether apples cause cancer and finds that they likely do not. One could say that whether the opinion that apples cause cancer is reasonable is irrelevant since the actual point with respect to RSP is that the source is reliable, not that its contents are accurate. But this point itself is clearly ignored by the comic, because it does not tell us whether the ancient books are reliable. Instead, we are meant to assume that they are not due to the contents, which brings us back the the first point that some of the contents (the apple) are in fact at least somewhat worthy of serious question and study. Additionally, Euclid's Elements, despite being written in ~300 BC, is still correct today, as are many ancient mathematical works. So the comic fails on all points except being antagonizing to the person that goes to look up a source here, reads that comic and then finds out that their source was deemed unreliable by Wikipedia's editors, placing them on the same level as the idiot that believes apples cause cancer. J2UDY7r00CRjH (talk) 21:49, 3 September 2024 (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.

List purge

I've purged all entries that only had one discussion from the list (26 entries out of ~435, a ~6% reduction in list size/3% reduction in page size). I've put them below in case someone finds other discussions. I think I've got them all here. Headbomb {t · c · p · b} 20:59, 12 August 2024 (UTC)

I've also redirected the shortcuts to this discussion so they don't get lost/point to nowhere. Headbomb {t · c · p · b} 21:02, 12 August 2024 (UTC)

I've reverted for now. Those entries were all (at a glance) the result of RFCs whose clear intent and consensus was to add them to this page; removing them would require discussion and consensus. More generally, the argument that something must actually have perennial discussion before being discussed and added to this list was actually raised at several of those RFCs and rejected, so removing them under the argument that they're not actually perennial has been considered and rejected; while it's the original premise for this page and part of its name, it's not a hard-and-fast requirement. And finally, most importantly, it's just not an improvement. We list the prior discussions, so anyone who is concerned can look at them and determine how frequently something has been discussed in the past; but it's still valuable to be able to have a go-to list of such RFCs and the consensus they reached. Deleting them from here and putting them on talk (???) serves no purpose beyond making the established consensus for those sources harder to find, which I don't think we ought to be doing and which makes no real sense anyway - there still is a consensus in each case, after all. (Also, while I know it wasn't intentional, please don't make a sweeping controversial change like this to a page with no discussion, then immediately follow it up with other sweeping improvements all over the page; because you changed so many things, I had to restore an older revision and then manually go through and re-instate all the changes you made afterwards in order to revert you cleanly.) --Aquillion (talk) 01:26, 13 August 2024 (UTC)

Just went to use one..... Our shortcuts don't go to the main page anymore.... They still redirect here. Can we get some more input on this whole topic. Moxy🍁 02:25, 15 August 2024 (UTC)

As the list changes have been reverted I've also for the moment reverted the changes to the redirects. -- LCU ActivelyDisinterested «@» °∆t° 16:56, 15 August 2024 (UTC)

putting them under a new list below the current one as "niche sources" or something would be better NotQualified (talk) 20:47, 9 September 2024 (UTC)

List collapsed for length

Perennial sources Source Status (legend) Discussions Uses List Last Summary Asian News International (ANI)  2021 2021 Asian News International is an Indian news agency. For general reporting, Asian News International is considered to be between marginally reliable and generally unreliable, with consensus that it is biased and that it should be attributed in-text for contentious claims. For its coverage related to Indian domestic politics, foreign politics, and other topics in which the Government of India may have an established stake, there is consensus that Asian News International is questionable and generally unreliable due to its reported dissemination of pro-government propaganda. 1   Australian Strategic Policy Institute (ASPI)  2021 2021 There is consensus that use of Australian Strategic Policy Institute should be evaluated for due weight and accompanied with in text attribution when used. Editors consider the Australian Strategic Policy Institute to be a biased or opinionated source that is reliable in the topic area of Australian defence and strategic issues but recommend care as it is a think tank associated with the defence industry in Australia and the Australian Government. 1   AVN (Adult Video News, AVN Magazine)  2021 2021 AVN is considered generally reliable for the adult industry. Editors should take care to ensure the content is not a republished press release (which is marked as such in search). 1   Bustle  2019 2019 There is consensus that the reliability of Bustle is unclear and that its reliability should be decided on an instance-by-instance basis. Editors noted that it has an editorial policy and that it will issue retractions. Editors also noted previous issues it had around reliability and that its content is written by freelance writers – though there is no consensus on whether this model affects their reliability. 1   California Globe  2021 2021 There is consensus that The California Globe is generally unreliable. Editors note the lack of substantial editorial process, the lack of evidence for fact-checking, and the bias present in the site's material. Editors also note the highly opinionated nature of the site as evidence against its reliability. 1   Center for Economic and Policy Research (CEPR)  2020 2020 The Center for Economic and Policy Research is an economic policy think tank. Though its articles are regularly written by subject-matter experts in economics and are frequently cited by reliable sources, most editors consider the CEPR biased or opinionated, so its uses should be attributed. 1   Coda Media (Coda Story)  2021 2021 A 2021 RfC found consensus that Coda Media is generally reliable for factual reporting. A few editors consider Coda Media a biased source for international politics related to the U.S., as it has received funding from the National Endowment for Democracy, though not to the extent that it affects reliability. 1   Correo del Orinoco  2023 2023 There is consensus that Correo del Orinoco is generally unreliable because it is used to amplify misleading and/or false information. Many editors consider Correo del Orinoco to be used by the Venezuelan government to promulgate propaganda due to its connection to the Bolivarian Communication and Information System. 1   Daily NK  2022 2022 The Daily NK is an online newspaper based in South Korea that reports on stories based inside of North Korea. There is no consensus as to if it should be deprecated or used with attribution. There is a consensus that this source, as well as all other sources reporting on North Korea, is generally unreliable. However, due to a paucity of readily accessible information on North Korea, as well as a perception that Daily NK is not more unreliable than other sources on the topic, it can be used as a source, albeit with great caution. 1   The Dorchester Review  2024 2024 There is consensus The Dorchester Review is generally unreliable, as it is not peer reviewed by the wider academic community. It has a poor reputation for fact-checking and lacks an editorial team. The source may still be used in some circumstances e.g. for uncontroversial self-descriptions, and content authored by established subject-matter experts. 1   Encyclopedia Astronautica  2023 2023 Encyclopedia Astronautica is a website on space history. A 2023 RfC found no consensus on the reliability of the site. There appears to be a consensus that this is a valuable resource, but it lacks editorial oversight, contains errors, and is no longer updated. Caution needs to taken in using this source. 1   The Globe and Mail  2021 2021 In a 2021 RfC, editors found a strong consensus that The Globe and Mail is generally reliable for news coverage and is considered a newspaper of record. 1   The Indian Express  2020 2020 The Indian Express is considered generally reliable under the news organizations guideline. 1   International Fact-Checking Network (IFCN)  2020 2020 The Poynter Institute's International Fact-Checking Network (IFCN) reviews fact-checking organizations according to a code of principles. There is consensus that it is generally reliable for determining the reliability of fact-checking organizations. 1   Mail & Guardian  2021 2021 The Mail & Guardian is a South African newspaper. There is consensus that it is generally reliable. 1   Metal-experience.com  2021 2021 Metal-experience.com was determined to be generally unreliable for factual reporting. 1   Newslaundry  2020 2020 There is consensus that Newslaundry is generally reliable. Some editors have expressed concerns regarding possible bias in its political narratives and reporting on rival publications; in cases where this could reasonably apply, attribution is recommended, and sufficient. 1   Our Campaigns  2021 2021 Our Campaigns is considered generally unreliable due to its publishing of user-generated content. 1   People Make Games  2023 2023 There is consensus that People Make Games is generally reliable for the topic of video games, although care should be taken if using the source for WP:BLP-related information due to concerns that they have no clear editorial policy, and they are a WP:EXPERTSPS. — Pride.com  2020 2020 There is consensus that Pride.com is marginally reliable and that its articles should be evaluated for reliability on a case-by-case basis. Editors consider Pride.com comparable to BuzzFeed in its presentation. 1   Quadrant  2019 2019 Most editors consider Quadrant generally unreliable for factual reporting. The publication is a biased and opinionated source. 1   RTÉ (Raidió Teilifís Éireann)  2023 2023 RTÉ is an Irish public service broadcaster. There is consensus that RTÉ is generally reliable. 1   Sherdog  2020 2020 In the 2020 RfC, Sherdog was determined to be not self-published and can be used for basic information on MMA fighters and matches. However, it is considered less reliable than ESPN and other generally reliable sources, so use with caution. 1   Sky News Australia  2022 2022 In the 2022 RfC, there is a consensus that additional considerations apply to Sky News Australia, and that it should not be used to substantiate any exceptional claims. The talk shows for Sky News Australia engage in disinformation and should be considered generally unreliable. The majority of articles labeled as "news" contain short blurbs and video segments, which should similarly be considered unreliable. For articles with significant written content, caution is advised. Sky News Australia is not to be confused with the UK Sky News; the two are presently unaffiliated. 1   XBIZ  2021 2021 XBIZ is considered generally reliable for the adult industry. However, it publishes press releases/sponsored content without clearly delineating the distinction between their own journalism and the promotional content of others. Thus, editors should take care that the source is not used for content obviously or likely to be promotional. 1

Press coverage on this page

I just added another one to the template, there's quite a few there now. I'm not saying a WP-article about this page would be a good idea, but it might survive. Gråbergs Gråa Sång (talk) 10:16, 11 September 2024 (UTC)

Are these sources appropriate for a private limited company specialising in Fintech?

Do these sources meet all 4 of the criteria to be considered for a private limited company: in-depth, reliable, secondary and strictly independent of the subject?

Which?

Best prepaid travel money cards 2024 Discover which are the best prepaid currency cards to save money on your trip

https://www.which.co.uk/money/credit-cards-and-loans/credit-cards/prepaid-euro-and-dollar-card-reviews-ag0Bt7D7bxKL

Finextra

Currensea smashes crowdfunding target

Travel debit card Currensea has raised over £1.7m from 760 investors in just four hours on crowdfunding platform Seedrs.

https://www.finextra.com/newsarticle/44400/currensea-smashes-crowdfunding-target?utm_medium=newsflash&utm_source=2024-7-2&member=81820

Fintech Finance

Travel Debit Card Currensea Smashes Crowdfunding Target in Just Four Hours as It Raises Over £1.7m on Seedrs

https://ffnews.com/newsarticle/funding/travel-debit-card-currensea-smashes-crowdfunding-target-in-just-four-hours-as-it-raises-over-1-7m-on-seedrs/#:~:text=In%202022%2C%20Currensea%20raised%20over,a%20total%20of%20%C2%A311m.

City AM

Fintech advised by former Amazon and Visa executives to launch travel card

https://www.cityam.com/fintech-advised-by-former-amazon-and-visa-executives-to-launch-travel-card/

https://www.cityam.com/london-fintech-currensea-launches-product-for-small-businesses/

Electronic Payments International

Travel debit card Currensea raises over £1.7m on Seedrs

Currensea smashes its crowdfunding target in just four hours raising the company's value to £28.5m

https://www.electronicpaymentsinternational.com/news/travel-debit-card-currensea-raises-over-1-7m-on-seedrs/

Yahoo Finance!

Travel debit card Currensea raises over £1.7m on Seedrs

https://finance.yahoo.com/news/travel-debit-card-currensea-raises-091509093.html

Pymnts

Currensea on Offering the Ability to Decouple From Banks While Giving Back

https://www.pymnts.com/next-gen-debit/2022/currensea-offers-ability-decouple-banks-while-giving-back/ SarahHunnings24 (talk) 16:29, 19 September 2024 (UTC)

@SarahHunnings24 Please go to the WP:RSN. They also have a search bar to look in the archives and see if anyone has talked about those too. Apenguinlover(🐧!) 23:09, 20 September 2024 (UTC)

Thank you - I managed to find two of them (Pyments and City AM) and they are okay. However the others have not come up - what should I do and if someone could let me know whether these sources are credible that will be great. SarahHunnings24 (talk) 09:47, 23 September 2024 (UTC)

This is the page for discussing how to improve the perennial source list, such as how it should be formatted. You will likely have better luck if you post your question to WP:Reliable sources/Noticeboard, which for discussing the reliability of sources. -- LCU ActivelyDisinterested «@» °∆t° 12:21, 23 September 2024 (UTC)

Could we talk some horse sense re Discogs?

Executive summary: images of records and record jackets at Discogs should be allowed as refs, and this noted in the Discogs entry. (I'm not talking about the text in Discogs. All or virtually all of that is no good, mustn't be used, fine. I'm talking about images.)

I believe this is a maintenance issue and so fit for being posted here. Willing to be corrected ofc.

Argument : first, let me lay out some background statements -- truths, I believe:

1. Our #1 remit is to serve the reader.
2. We can't hope for say 99.9% accuracy in our honestly ref'd statements of fact. One wrong fact out of a thousand is far more than humanly possible I'd say, and in fact our actual number must be worse. Few of our sources are that accurate for starters, and being human we will sometimes misread them and so on.
3. Works of art and artifice are their own references for their own contents, I believe. So, to verify basic facts about a book, film, record, etc., the reader may obtain a copy of the work and check for herself, and we consider this to be sufficient reffing. After all, most all track listings etc. etc. etc. are reffed this way -- no link to a secondary source used, expected, or usually possible. (More easily obtainable refs are a welcome bonus of course -- and that's why I'm here.) For good or ill (and there's some of each), this is our de facto rule.
4. We are vulnerable to elaborate hoaxes -- well-faked refs I mean -- for all subjects, and surely some get thru (many (tho not all) are to push some POV, I would guess). But faked refs are probably very rare, and we assume no hoax refs absent some reason to suspect. Otherwise we'd be paranoid, and paralyzed.

Most everyone on board with this so far?

So. Images of record labels and jackets uploaded to Discogs (by an anon person and not formally checked by anyone, understood), can we use those? How accurate are they for their contents and how accurate are they for facts?

For contents -- absent a deliberate elaborate hoax, which would involve skilled photoshopping of the image (often several), it must be very high. Fewer than one in a thousand of Discogs label images are not photoshopped hoaxes, I am confident. Why would some artist go to this trouble? There's no ideological reason.

As to statements of fact -- who the songwriter is, who really wrote most of the script, etc. -- yes, the label, screen credits, book forward, etc can and sometimes do get this wrong, by mistake or deliberately. But, we are recommending the source to the reader as sufficiently reliable already. So providing a much easier way the reader to access an image of the source and verify our statement without trolling thru used record shops and eBay and so forth... this is a considerable service to the reader, n'est-ce pas? (BTW Discogs has URLs for the images alone, so no worries about the link target being polluted with unreliable text.)

I can't think of any good, none-kneejerk or circular counterarguments. I could be wrong, of course, and am willing to be educated. Otherwise I recommend we move forward with this, mnmh? Herostratus (talk) 03:37, 3 September 2024 (UTC)

The only websites I know that give any reliable provenance and sourcing to images are museums, direct academic publishers, professional artists' own work, and usually-but-sometimes-not newspapers. Everywhere else, including Wikipedia, seems more or less content to have pictures speak for themselves as far as reliability. If you can permalink to it, I don't see how images on Discogs (or similar UGC+ sites) can be much worse than the images posted most anywhere else online (including uploaded here) that are reused far too often. Since they are photos of a published work, the auditing process is straightforward -- somebody who has access to that published work available will complain or edit if it's inaccurate. (This kind of thing happens with photographs, maps, diagrams, etc all the time.)

Support your suggestion as it's essentially on par with image quality control at Commons. In future I'd want to fix that and get some standards for using images on WP, but not many others seem to support the idea. SamuelRiv (talk) 04:33, 3 September 2024 (UTC)

Support since I don't see how using a scan from Discogs is worse than Wikipedia relying on an editor copying the information from an album they have a home or found on Google image search using no sources at all. Cortador (talk) 08:28, 16 September 2024 (UTC)

Comment – I don't really see how "this isn't worse than another thing we probably shouldn't do" is a compelling argument. In any case, the point is verifiability – it seems Discogs is only serving as the host of the image file, and not as the source per se (an actual source would seem to be the album or a reliable source discussing it that includes the image within). That is to say, since we're not using Discogs as a source, we shouldn't be treating this as a RS question at all or why we should bother including Discogs links as a matter of course. Remsense ‥ 论 08:38, 16 September 2024 (UTC)

I agree with Remsense, in that it's the album being referenced not Discog. The question then becomes can the images at Discog be trusted? Probably a question for RSN not RSP. Also I would say that 3/. is a bug not a feature and such listing are regular the target of subtle vandalism, it would be a boon if we did have a reputable online listing for tracks. Albums are a reliable source for themselves, but secondary sources are preferred. -- LCU ActivelyDisinterested «@» °∆t° 10:14, 16 September 2024 (UTC)

I'd like to plug the {{cite object}} subtemplate that I created for cases like (3) (or mainly for image captions/descriptions, museum catalog items, internal secondary references to specific objects, etc.) -- in-text of course one should carefully beware of the guidelines for wp:primary historical sources. One should take caution that while I might agree that a work of art is a source for its contents (the definition of a primary source), a photograph or mp3 is a reproduction or representation. Without the uploader explicitly noting the where when and what, or the illustrator giving the source material for an svg, there is a lot that can go wrong in citing the uploaded photograph, if the uploader doesn't give the basic required image info per policy, as representative of the object. SamuelRiv (talk) 17:51, 19 September 2024 (UTC)

It looks like maybe we should have a proper WP:RFC on this to get a definite decision, and that's fine.

For the rest, some of us will have to agree to disagree til then I guess. I stated my case. But to respond to recent point made:

Re "it's the album being referenced not Discogs", no it's not, often enough. Discog label photos are used as a source, a lot. I mean I do it, and a lot of other editors do I'm sure. Where do you think editors are getting this info for obscure records? And "supposed" to do it or not, you can't really stop people from doing it, because it's convenient, it's easily reliable to our standards, it would be pretty pedantic and overly rule-bound to not do it, it's an asset to the article, and you can't prove it anyway. Rules are supposed to codify common good practice, and all I'm suggesting is a small rule change that does that. No point in pretending that people don't do it. And if we stop pretending, we can actually point the reader to the source we ourselves are using.

Re "secondary sources are preferred", why on earth would that be. We're not establishing notability, we're not cherrypicking primary sources to make some point or create some synthesis that might be wrong, we're not using primary sources that might not be reliable, we're not finding only a partial set of primary sources that might skew the material, and in fact we're slightly reducing errors that might be introduced by a secondary source. A secondary source might be an article or book listing tracks or songwriters or whatnot, and the writer might miscopy or whatever (and most of these are not fact-checked to that level, to be realistic).

Re this being "Probably a question for RSN not RSP", well, RSN threads are more emphemeral I think?, so I don't know. I'm thinking to have the RfC here on the page where the change would be made, and make a notice at RSN pointing to it. The other way would be OK too tho.

Re "Without the uploader explicitly noting the where when and what, or the illustrator giving the source material for an svg, there is a lot that can go wrong in citing the uploaded photograph, if the uploader doesn't give the basic required image info per policy", what could go wrong? Asking. For a photograph of a building or person or anything else except a label or jacket, yes, and I'm not advocating using Discogs for anything like that. But unless the label or jacket is a photoshopped hoax, who cares if it was taken by a drunk cat or whatever, it's accurate on its face. Unlike a photo of a person or building etc, it's not like a photo of a label that says "Angry Young Popes' Greatest Hits" could actually be a photo of the label for "Best Of The Beagles". Flat impossible, so not worth worrying about.

Re what we need to be "carefully beware of", it'd be letting this or that or the other rule stop us from using common sense to serve the reader. That way lies pedantry. I'm not suggesting that using Discogs images be required, but if an editor is crafting an article to best serve the reader, stay out of her way I'd say.

BTW it now occurs to me that you're probably going to reduce errors a little bit by using these photos -- subtle errors introduced by vandals for lulz in particular. This happens -- subtle date changes etc, cos those can sneak under the radar sometimes. If a passing reader or editor has a "wait, what?", moment, a quick link to check the source (rather than having to go to another website and doing some searching to get (probably) the same webpage), is more likely to be used and the error corrected, I am sure. "Having fewer errors" is a worthwhile goal.

Nobody has argued yet that the images are actually unreliable in real life, because they can't. That matters, a lot, and here's one case where WP:IAR (one of the foundational Pillars, which supersede all those other rules) could be applied. We want to be judicious when invoking WP:IAR, but here's a situation that it's made for.

OK. Is there any more WP:RFCBEFORE to be done before we move forward? Herostratus (talk) 04:58, 22 September 2024 (UTC)

I don't see what there is to have an RfC about: I doubt we'll opt to establish the fiction that Discogs is magically reliable in a meaningful sense despite being user-generated. It's often just functioning as the host for an image, but there is no logically consistent case for its acceptance as a useful source to verify claims against. There's really nothing more to it than that. Remsense ‥ 论 05:06, 22 September 2024 (UTC)

A couple of things if your referencing a photo of a real world object, then you're referencing that real world object. It doesn't really matter how you do the referencing or how it's setup, your referencing the primary object. Also RSN discussion are the exact opposite of ephemeral, there's an archive search box on the header and the discussion are also listed in the RSP for easy of use. As to why secondary sources are preferred, because that's what policy and guidelines say. Discussing that issue is better done elsewhere, as it's so far reaching.

I'm not saying you can't just reference the primary object by using a photo, that would be allowed by multiple different policies / guidelines. The only issue I could see is one of reliability, e.g. if the photos at Discog can be trusted which if you want to have an RFC about I would suggest RSN. -- LCU ActivelyDisinterested «@» °∆t° 11:49, 22 September 2024 (UTC)

OK. Herostratus (talk) 20:42, 27 September 2024 (UTC)

"I'm not talking about the text in Discogs. All or virtually all of that is no good, mustn't be used, fine." I've always found most of discogs "text" to be 99.9% correct. Could you provide, say, 10 examples of where you've found the text to be "no good"? Many thanks. Martinevans123 (talk) 21:01, 27 September 2024 (UTC)

Yes of course the text is accurate. It's more accurate than say the Wall Street Journal most likely. BUT, anybody can upload any text to Discogs, and there's no (formal) fact checking operation (tho they do have their hivemind, and its a busy site). We don't use text like that, and should not (for one thing anybody can change text which isn't true for books and articles and refs that change are bad.) And we basically never make exeptions, and this is functional, most editors by far would agree. I mean, surely some editors do use text info from Discogs and just don't give a ref, and that can pass for decades or longer. I do not recommend doing this though. It's just different from the pictures, which self-reference. Herostratus (talk) 01:51, 28 September 2024 (UTC)

Any examples? Maybe even a single one, of where it's wrong? I fully understand the principle for not using. I also agree about using the images. I have used some myself for some album articles, e.g. for verifying personal line-ups and instrumentation. Martinevans123 (talk) 08:04, 28 September 2024 (UTC)

So, just to finish up, and review. I'm not super suprised there was some objection . Many editors are quite tradition-oriented, and I mean really it can be harder to change a Wikipedia rule than the Catholic Mass [hyperbole for effect]. And it is good to be conservative about changing stuff, so I'm not upset or anything. But this is such a slam-dunk, I figure it for a go, and there weren't any very good counterarguments given I would say.

• "Support your suggestion as it's essentially on par with image quality control at Commons" (SamuelRiv).
• "Support since I don't see how using a scan from Discogs is worse than Wikipedia relying on an editor copying the information from an album they have a home..." (Cortador)
• Oppose, as "[I don't] see how 'this isn't worse than another thing we probably shouldn't do' is a compelling argument..." "the point is verifiability – it seems Discogs is only serving as the host of the image file, and not as the source per se (an actual source would seem to be the album or a reliable source discussing it...)" "I doubt we'll opt to establish the fiction that Discogs is magically reliable in a meaningful sense" (Remsense)
• "Oppose, I agree with Remsense, in that it's the album being referenced not Discog. The question then becomes can the images at Discog be trusted?..."Albums are a reliable source for themselves, but secondary sources are preferred." (ActivelyDisinterested)
• User Martinevans123, can't tell... seems to suppport using Discogs text so you'd think they support the proposition, but not stone sure.
• And I support as proposer, I gave my reasons above.

One editor did say "since we're not using Discogs as a source, we shouldn't be treating this as a RS question at all or why we should bother including Discogs links as a matter of course". This is actual circular reasoning (we shouldn't use any Discogs data because we don't use any Discogs data) which enh.

So anyway, 3-2 at this point.

One of the things that is going to stick people, and I get this, is that we should try to avoid primary source. Nobody can say why because in this case anyway there isn't a reason beyond "It's just not done". And how the album itself is not a primary source I can't figure... Hmmm. Herostratus (talk) 01:51, 28 September 2024 (UTC)

I'm not a traditionalist, you just haven't articulated any actual reason this case should be carved out as an exception to the principle that "user-generated sources are generally unreliable for verifying claims". You also misunderstand what we can cite primary sources for: what its cover looks like is clearly one of those things. I'm sorry, this just seems like a fundamental confusion about what citation accomplishes and what verifiability means in practice. As far as I can tell there's no actual problem here, but you want to solve it by blowing a big hole in the logical boundaries of who we verify claims against. Remsense ‥ 论 03:58, 28 September 2024 (UTC)

Page size

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.

---

This page is presently 407kB long. It is difficult to load this page with a browser, and even more difficult to edit it. Are there any sources that are not sufficiently perennial, or sufficiently important, that they need to be included? Alternatively, could this page be split? James500 (talk) 14:48, 30 June 2024 (UTC)

• I think splitting a totally reasonable idea and have tagged the page accordingly. Other extremely long lists have done this, usually alphabetically. Just Step Sideways ^{from this world ..... today} 17:42, 30 June 2024 (UTC)
• I am opposed to cutting sources just to make the page shorter. If a source has been discussed repeatedly, and a consensus reached, there should be an easy way for editors to learn about that. But splitting is completely appropriate. John M Baker (talk) 19:57, 30 June 2024 (UTC)
• There have been repeated proposals to this effect, but frankly there is no easy division that does not greatly reduce usefulness as a quick reference guide. It's long because there's a lot of entries and they need to be somewhere. I must note that a short while ago you were proposing this page's complete removal as useless, so it's not clear the degree to which your advice on its modification for usefulness should be considered - David Gerard (talk) 20:48, 30 June 2024 (UTC)
• Oppose. Splitting the table means you can't sort it properly any more with all entries. Cortador (talk) 21:41, 30 June 2024 (UTC)
• Agree to split. The obvious choice would be A-M and N-Z. Otherwise sectioning alphabetically (A,B,C,etc), and collapsing each section could help with browser loading time if not mistaken? Similar to the French wiki version. CNC (talk) 20:57, 30 June 2024 (UTC)
• Oppose – Agree with David Gerard and Cortador that the usefulness of a single-page presentation is paramount in terms of sorting and searchability, so I discourage arbitrary splitting or shortening of this page. There are many more readers of this critical page than editors. Therefore, as long as it loads in an adequate manner we should not alter the page to cater to editors. - Fuzheado | Talk 23:51, 30 June 2024 (UTC)
• On a technical level something should be done to alleviate WP:CHOKING concerns. Mfko (talk) 14:13, 11 July 2024 (UTC)
• Oppose per Fuzheado The entire point of this page is to centralize information. We could have a distilled parallel list of some commonly discussed sources on a user subpage, though. Bremps... 03:23, 1 July 2024 (UTC)
• Comment What do people think about splitting combined with transclusion like at List_of_Latin_phrases_(full) (except with a single sortable table)? 174.92.25.207 (talk) 04:48, 1 July 2024 (UTC)

  How does this work? JuxtaposedJacob (talk) 02:41, 10 September 2024 (UTC)

  Read on, man. Transclusion, a live-ish view of one page from another page, is one of the fundamental functions of MediaWiki. Aaron Liu (talk) 03:01, 10 September 2024 (UTC)

• Oppose. No good way to do it.—Alalch E. 17:02, 5 July 2024 (UTC)
• Since the page will keep growing, some sort of split seems the standard response for pages that get so big they cause technical problems. Would it be possible to make a searchbox on the first page that will cover the entire RSP? WP:CHOKING seems to be the written guidance. Gråbergs Gråa Sång (talk) 17:07, 5 July 2024 (UTC)

Also, if we do split it, we could perhaps divide Sources into a few subsections, that might make editing easier. Gråbergs Gråa Sång (talk) 17:28, 5 July 2024 (UTC)

The Sources section on this page could look something like

Searchbox

• link to sources 0-L
• link to sources M-Z

A little similar to 2019_in_film#2019_films, articles like that used to have big-ass lists of films, mostly American ones. They still have big-ass lists of dead people... Gråbergs Gråa Sång (talk) 13:07, 6 July 2024 (UTC)

• Comment Is this really a question about the inclusion criteria for this list, and whether it should be adjusted? I'd say that if there hasn't been an RFC for a source in X years, then it's not really perennial anymore. (Not sure what X should be, but it's at most 10.) Maybe we should add something about that. Apocheir (talk) 03:01, 6 July 2024 (UTC)
• Oppose. Æo (talk) 23:57, 7 July 2024 (UTC)
• Comment I like Gråbergs Gråa Sång's partitioning idea! An alternative source division schema could be a link to each of the four major source reliability categories of Green, Yellow, Red, and Grey. Each would have its own page.--FeralOink (talk) 19:30, 10 July 2024 (UTC)

I don't think we should partition by color. I think that would reinforce the incorrect idea that these "reliability categories" are absolutes, rather than depending on context. There is no black-and-white line between reliable and unreliable sources; rather, different sources are reliable for different kinds of claims. —Mx. Granger (talk · contribs) 14:41, 11 July 2024 (UTC)

• There probably are some frivolous entries though maybe not enough to make a big difference. I was thinking of removing the entry for "bestgore", who would ever try and cite that really? Hardly perennial. VintageVernacular (talk) 03:44, 11 July 2024 (UTC)

You'd be surprised as to what people will try to cite. Some entries, particularly the blacklisted ones, have useful context as to why they were blacklisted in the first place. But I agree that there are some entries that are stretching the definition of "perennial". Mfko (talk) 14:11, 11 July 2024 (UTC)

• Comment A split is treating the symptoms rather than the root cause. This page has attracted an enormous amount of cruft over the years and desperately needs to be pruned, the inclusion criteria adjusted, and sources aggregated (as seen below with the medical preprints for example).

For example, several entries could simply be aggregated under a generic banner of "this is user-generated content, don't use this". (Stack Overflow, Quora, TV Tropes, Ethnicity of Celebs, Land Transport Guru, SourceWatch, WhoSampled). There are plenty more entries in the list but these stuck out the most.

There are also problems with the definition of "perennial" as there are several entries that haven't been discussed for 14+ years such as Spirit of Metal. Of course on the flip side, you have entries like "The New Yorker" that haven't been discussed as well but are obviously reliable. Where is the line drawn? Are there really that many people attempting to cite Spirit of Metal to warrant its inclusion on the list? Mfko (talk) 14:08, 11 July 2024 (UTC)

Agree with merging the user-generated sources together in most cases. Some entries that are blacklisted or deprecated might warrant still being listed on their own. There are also some edge cases, like sites that host both UGC and professionally written content (e.g. Atlas Obscura), or UGC sites when used by reliable news organizations (e.g. YouTube) that we might have to handle on a case-by-case basis. But generally, the run-of-the-mill generally unreliable UGC sites could all be condensed together. See also the § Merging some entries section below.

Your comment about the definition of perennial echos my earlier comment about the inclusion criteria for this page. Apocheir (talk) 18:44, 11 July 2024 (UTC)

• Generally disagree. This sounds like trying to put a plaster over the problem, if we are too look at the root problem which is size. At the moment the page is at 2.5KB, it should be around 2KB per CHOKING issues referenced above. Grouping together entries like user content isn't going to solve the problem here, not by a long shot. I'm not opposed to sources that haven't been discussed in many years being taken out, but not convinced even this would see the minimum required 20% decline in size either, probably more like 10% at best. Also it's complicated as some sources haven't been discussed in a long time, but otherwise are regularly used on WP, so not being discussed isn't necessarily a barometer of having no benefit for inclusion either. Sometimes you have to look at some of these entries and simply consider "could someone find this useful"; the answer is someone probably could, even if not regularly, regardless of if it's inclusion is necessary. Overall I'm generally opposed to unsustainable solutions such as "trimming" in these cases. It reminds me of when articles reach 20,000 words; instead of simply splitting as required, editors suggest trimming unnecessary content; as if it's going to reduce the article size by 50% to a readable size (which never happens). It'd be a shame to get to the point of trimming entries, which naturally proceeds into excluding further entries based on similar criteria, simply because we are unwilling to split the article in a conventional manner. CNC (talk) 19:18, 11 July 2024 (UTC)
• Split alphabetically – While we're at it, given the size of the page we might as well split into three (I propose 0–F, G–M, and N–Z, eyeballing roughly equal length), not two. We can have likely have a transcluded (full list) page too. Remsense诉 19:30, 11 July 2024 (UTC)

Sounds good. Gråbergs Gråa Sång (talk) 19:47, 11 July 2024 (UTC)

That works for me.--FeralOink (talk) 06:12, 12 July 2024 (UTC)

• Oppose: This page is very useful for pressing "ctrl-f" and splitting into multiple pages would double the work required every time the page was used. Slow editing is an acceptable price given that the article does not have to be edited that often. Support merging the use-generated sources. Mrfoogles (talk) 07:34, 12 July 2024 (UTC)

It's not just slow to edit; did you notice the transclusion proposal? Remsense诉 18:36, 12 July 2024 (UTC)

I wonder is there's an "information page" on en-WP that gets edited more often than this one. Gråbergs Gråa Sång (talk) 18:58, 12 July 2024 (UTC)

• Oppose alphabetical split as this would create a new problem after solving another. — PerfectSoundWhatever (t; c) 17:48, 15 July 2024 (UTC)
• Support per @Remsense and @Ivanvector's proposals below. voorts (talk/contributions) 20:41, 19 July 2024 (UTC)
• Oppose splitting that doesn't result in a single list that editor can check against. So I would support the transclusion option. As to inclusion criteria the change that should be made is the comment about needing an RFC for inclusion, this causes editors to jump start RFCs on RSN for sources with no prior discussion just so they can get the source listed here. -- LCU ActivelyDisinterested «@» °∆t° 10:55, 21 July 2024 (UTC)

@ActivelyDisinterested, I commented on that here: Wikipedia_talk:Reliable_sources/Perennial_sources/Archive_9#Starting_WP:RSN-discussions_with_the_purpose_to_include_stuff_on_WP:RSP Gråbergs Gråa Sång (talk) 11:08, 21 July 2024 (UTC)

Completely agree with Levivich, although I don't know how to stop it from happening. -- LCU ActivelyDisinterested «@» °∆t° 11:15, 21 July 2024 (UTC)

I do too. I've seen more than one RSN RFC where people ask "Is there actually an issue?", and that sometimes help. Gråbergs Gråa Sång (talk) 11:34, 21 July 2024 (UTC)

Maybe the inclusion criteria should explicitly exclude discussions/RFCs that only seek RSP inclusion, but that might be better discussed in a separate thread. -- LCU ActivelyDisinterested «@» °∆t° 14:09, 21 July 2024 (UTC)

• Oppose any split. This is a list. Using the list does not mean reading the whole text. The list is mostly, if not exclusively, being used to check if a specific source has been deprecated. It goes without saying that a single search is the efficient choice. -The Gnome (talk) 16:39, 31 July 2024 (UTC)

  Is this still your position given the proposal below, which would allow users to seamlessly search the full list on one page as before? Remsense诉 16:43, 31 July 2024 (UTC)

It still is, because in the case of lists such as this, there is, by definition, no issue of page clutter. Almost all uses of this list concern a single source being checked. Suggestions to change something very successful are all well meaning but misguided: they're about lists generically. -The Gnome (talk) 21:15, 1 August 2024 (UTC)

That's slightly confusing, I'm not sure why anyone would be proposing a complication like this on a lark or because we don't know article guidelines don't apply.

The concrete reason expressed by multiple editors is that they have considerable difficulty reading and editing this page due to its length and breadth being enough to slow down their system.

(I'm one of those, but I haven't explicitly said as much because I always have weird computer problems and don't like complaining about them.) Remsense诉 21:42, 1 August 2024 (UTC)

Same. RSPS sometimes takes like 20 seconds for me to publish an edit. voorts (talk/contributions) 00:17, 2 August 2024 (UTC)

• Break table and switch to a subpage format, where each subpage gives a nuanced profile of a source. The table has always been cluttered and cramped, and now it's still too big. theleekycauldron (talk • she/her) 17:50, 1 August 2024 (UTC)

Do you really believe that creating "a nuanced profile of [every] source" would be anything less than a Herculean task? Why not simply opt for simplicity, consolidation, and practicality? -The Gnome (talk) 21:15, 1 August 2024 (UTC)

I'm not clear either on that proposal or what it would do either—surely the current format of entries is generally sufficient for what the list is designed to accomplish? i.e. figure out how mean to be to who you're reverting, and maybe grab an example quote as to why a source is good or bad. Remsense诉 21:35, 1 August 2024 (UTC)

For sources that are GREL, GUNREL, or deprecated, a brief entry describing why ought to suffice. I think a nuanced profile of a source would only be needed if the source has additional considerations or there's no consensus about the source. If we go this route, however, I worry about editors warring over the contents of a subpage when they have a content disputes, rather than working it out on the article talk page.

Additionally, as someone who has closed a couple of RSN RfCs and added entries to the chart, I would not have closed those discussions if that had meant that I would have needed to write a nuanced profile of the source based on the consensuses that emerged from those discussions. voorts (talk/contributions) 00:33, 2 August 2024 (UTC)

• Oppose It does what it needs to do and I've never noticed any issues with it loading on browsers. I'm going to oppose on WP:AINT principles. The C of E God Save the King! (talk) 17:37, 7 August 2024 (UTC)

Several editors say that it's broken and loads slowly. Just because you don't have issues doesn't mean it ain't broken for others. voorts (talk/contributions) 20:59, 7 August 2024 (UTC)

We are only talking about it because it is indeed broken. Remsense诉 21:05, 7 August 2024 (UTC)

The assessments about the page being "broken" are, of course, entirely subjective. Substance is of the essence. As are assessments to the contrary, though the burden of proving "not broken" ain't on us. Do not ask us to prove a negative. -The Gnome (talk) 18:00, 12 August 2024 (UTC)

In fact they are not subjective, but instead well defined. As referenced above the in discussion, per WP:CHOKING: "The maximum limit for Wikipedia is via the MediaWiki software's wgMaxArticleSize to 2 MiB (specifically, 2048 kibibytes or 2,097,152 bytes). Exceeding the post-expand limit will result in templates in the article appearing incorrectly." At 2.5MB, it's clearly above this threshold by 536 kB currently. There's no valid argument against this. CNC (talk) 18:15, 12 August 2024 (UTC)

I have no idea where you got your "2.5MB" figure from. The entire page is only 0.43MB or 0.41MiB. Aaron Liu (talk) 18:30, 12 August 2024 (UTC)

From HTML document size, not wikitext. CNC (talk) 18:33, 12 August 2024 (UTC)

According to the NewPP report, that would be 1.42MiB. You're probably including the images and the skin. Aaron Liu (talk) 18:38, 12 August 2024 (UTC)

Assuming the burden is on people trying to prove the page is broken, several people have stated in this discussion that the page loads slowly for them. The only way to unbreak that is to fix the page. We're not asking you to prove a negative, but to show that there's some other compelling reason for not fixing the loading issue that a lot of people are having. voorts (talk/contributions) 18:59, 12 August 2024 (UTC)

One of the sillier comments I've ever read. Remsense诉 19:07, 12 August 2024 (UTC)

Try editing the page with syntax highlighting. Aaron Liu (talk) 18:32, 12 August 2024 (UTC)

• Oppose. The ability to easily search for a source at a glance to get a quick temperature check on how the community feels about it is too important to lose; while other search options exist, none of them are as easy or convenient (as anyone who has had to dig through searches for talk page archives can attest), so we should try to keep everything on one page for as long as possible. --Aquillion (talk) 01:35, 13 August 2024 (UTC)

  Perfunctory nudge about the below transclusion possibility—as the temperature check is no longer quick for many users, for whom the as long as possible period has already expired. Remsense诉 01:40, 13 August 2024 (UTC)

• Comment I've requested a close of this split proposal at WP:CR. voorts (talk/contributions) 21:40, 29 August 2024 (UTC)

Oppose anything other than a technical split - Just because editors that participate in RfCs understand the generally-accepted reliability or unreliability of a source doesn;t mean that everyone does. I use this page all the time and its usefulness would definitely be diminished if we removed even one source; however, I get the point of splitting up the page for technical reasons and am all for that if we can get it to work in a ctrl+f style manner. JuxtaposedJacob (talk) 02:45, 10 September 2024 (UTC)

• Oppose any split that would dissolve the contents of Wikipedia:Reliable sources/Perennial sources#Sources. This section needs to have all of its sources intact; I would not oppose splitting this page into two separate pages: one with only information regarding the list, and one with the actual list of sources.

𝚈𝚘𝚟𝚝 (𝚝𝚊𝚕𝚔𝚟𝚝) 19:50, 23 September 2024 (UTC)

@Yovt Are you opposed to a method that would split the list but also allow it to be viewable in entirety? Aaron Liu (talk) 20:30, 23 September 2024 (UTC)

• Oppose. This page serves as the directory for Reliable sources lookup. I do not think it would be a good idea for a split here. Cfls (talk) 15:35, 29 September 2024 (UTC)

  Did you look at the transclusion option? voorts (talk/contributions) 15:58, 29 September 2024 (UTC)

Transclusion possibility

Sorry if an annoyance, but does the proposal to split, while also transcluding the segments onto a (full list) page move the needle for any of the oppose !‍votes? As I see it, this would seem to enable equivalently easy or easier editing and searching for all users. Remsense诉 18:41, 12 July 2024 (UTC)

I was just going to say that - we already have plenty of pages that work that way. WP:RFPP comes to mind, as do the main pages of both WP:AFD and WP:RFD. For AFD and RFD all active discussions are transcluded to a main page that can be Ctrl-F searched, while clicking on an edit link takes you to edit the subpage of each individual discussion for AFD or to the daily page for RFD. WP:EDR also works that way, and is a page with multiple long tables. I think it would even be technically possible to transclude subpages onto a main page in one continuous sortable table, I think our highway exit list and election results tables work that way, but with as much info as is on this page it would be a nightmare to edit. Anyway, we wouldn't be reinventing the wheel here. Ivanvector (^Talk/_Edits) 14:31, 13 July 2024 (UTC)

In fact I just tried to demo what it would look like in my userspace, see User:Ivanvector/RSP split proposal. It doesn't work because there's something with include tags on the original page that break how I think the transclusions should work, and relative links to templates that of course don't exist in my userspace. I'm sure I could figure it out but this isn't how I was planning to spend my Saturday - if someone else wants to run with it then please be my guest, otherwise I will probably look at it again next week. This did get the page from 426kB down to 28kB, though.

Another thought I had while doing this is that we could change the page so that there's a master template for entries, and then each entry transcludes that template. That would be a task for someone who knows more about templates than I do. Ivanvector (^Talk/_Edits) 15:05, 13 July 2024 (UTC)

Ah, never mind all that, I got it working by just bulk removing the "onlyinclude" tags that I don't know what they're supposed to do (nothing about it in the instructions). Works now at least superficially; take a look. Ivanvector (^Talk/_Edits) 15:16, 13 July 2024 (UTC)

Looks fine, the tags are to do with #Duplication at Wikipedia:Deprecated_sources#Currently_deprecated_sources. When it comes to transclusion, @174.92.25.207 would be the editor to ask around here it seems. CNC (talk) 15:26, 13 July 2024 (UTC)

Support transclusion along with merging entries - but it sounds like that will be a separate project. Mfko (talk) 18:00, 14 July 2024 (UTC)

I generally dislike transclusion, but it seems like a sensible solution in this situation. -- LCU ActivelyDisinterested «@» °∆t° 10:56, 21 July 2024 (UTC)

Support This was what I was thinking as well. @Ivanvector Would only transcluding the table body work? Aaron Liu (talk) 01:43, 2 August 2024 (UTC)

Sounds like a good idea. Gråbergs Gråa Sång (talk) 08:17, 2 August 2024 (UTC)

Comment: Splitting sources into alphabetical subpages is a good start, but RSP should list every source without being too lengthy and it should be more practical for mobile users. Instead of transcluding all content, the main table could have colored rows with just a source column (linking only to subpage entries) and a status column. The bulk of the content would be on the subpages and source-specific shortcuts would also go to the complete entries on subpages. We could also have a "RSPFULL" page that would transclude every subpage for people that prefer the current format. Daniel Quinlan (talk) 01:11, 7 August 2024 (UTC)

That would require a lot more maintenance due to the duplication. Aaron Liu (talk) 01:18, 7 August 2024 (UTC)

I think it would be more maintenance, but changes would only be required for additions, removals, and status changes. Clarifications, additional discussion links, etc. would not require updates. Of the last 10 edits to WP:RSP, only 1 would have required a table update (an addition). The others were changes to other parts of the page, details that would be on subpages, or were reverted. Longer-term, the summary table could be automatically generated and updated by a bot (similar to what User:Rick Bot, User:ChristieBot, and other similar bots do for various pages). Daniel Quinlan (talk) 02:08, 7 August 2024 (UTC)

Links etc. will still require updates in the subpage. Bots still require maintenance. I don't see any reason to do what you said; for mobile people, they can already see the first columns. Aaron Liu (talk) 02:14, 7 August 2024 (UTC)

Some mobile users can't even load the current page. It's too large for some browsers. (@PARAKANYAA: If you want to add anything about the crashing issue you mentioned you were experiencing?) Using a summary table would improve the accessibility of the page, especially for mobile users, users with slower internet connections, and those who use assistive technologies. It's overwhelming even with a good computer, a fast connection, and large screen.

While some updates will require editing two pages instead of one, that affects only a minority of updates. Editing the summary table will be much simpler than editing the main entries. The page currently stands at about 35,000 words, which is 233% above the point where an article should Almost certainly should be divided or trimmed according to WP:SIZERULE. Although this isn't an article, it's simply too much content for a single page. Daniel Quinlan (talk) 03:06, 7 August 2024 (UTC)

To all whom may concern: I've made a better demo of this at User:Aaron Liu/sandbox/RSP Demo, which generates a full, uninterrupted table, as well as a separate table of deprecated sources generated from transcluding a labeled section from the full table. I can convert the entirety of RSP into this format if someone wants.

If an issue here is that loading the full table is too slow, we can make it so that the landing page is just the ToC and description (everything that isn't that table) of RSP, and the full table is at something like RSP/F (with a shortcut from RSPF as well). Aaron Liu (talk) 02:32, 13 August 2024 (UTC)

Support - I got some WP:TROUT by accidentally adding back the cartoon without concensus and almost crashed my tablet trying to revert my own edit. As long as mobile can edit it I support. Apenguinlover(🐧!) 11:25, 7 September 2024 (UTC)

Entries that might be trimmed

Some people above have proposed trimming the list. I think that we can start by cutting the following entries:

• Inter Press Service
• The New Yorker
• Pew Research Center

Each of those sources are well-known institutions and considered to be generally reliable, and the last discussions occurred over 10 years ago. voorts (talk/contributions) 23:53, 2 August 2024 (UTC)

Since none of these have no (noted) discussion for more than a decade, making them arguably not "perennial", and they do seem uncontroversially generally reliable, I'm ok with that. Gråbergs Gråa Sång (talk) 09:13, 3 August 2024 (UTC)

There a similar issue with sources that were only added due to discussions about adding them, for instance the 'ABC News (United States)' entry. If you check the discussions they are only about adding sources to the RSP, not any real dispute about the sources reliability. The same is true of 'The Age' and 'The Sydney Morning Herald', which only had an RFC to green list.

I've wondered whether the criteria should exclude discussions or RFCs started solely to include sources on the list. -- LCU ActivelyDisinterested «@» °∆t° 13:07, 3 August 2024 (UTC)

Honestly, I think that's worthwhile. Stop and really think about it: has anyone in good faith ever asked in a talk page discussion whether ABC is reliable? If that happens, this page won't help—as there's either particular context or the question is purely rhetorical. Remsense诉 13:09, 3 August 2024 (UTC)

Fine with me. Gråbergs Gråa Sång (talk) 14:19, 3 August 2024 (UTC)

IMO we should add those to footnote [a] if/when we remove it from the primary list. Aaron Liu (talk) 14:27, 3 August 2024 (UTC)

We can rewrite that footnote a bit, but I'm not sure all potential removals need to go there. Gråbergs Gråa Sång (talk) 15:08, 3 August 2024 (UTC)

There is potential harm with removing the mentions from the page entirely. Aaron Liu (talk) 17:37, 3 August 2024 (UTC)

What's the harm? voorts (talk/contributions) 17:39, 3 August 2024 (UTC)

Sorry, I didn't word that the best. While keeping them in the footnote probably wouldn't do much than inflating it with a couple dozen characters, readers might not see these sources (especially people who haven't heard of them) on the main page and question its reliability if they were entirely removed.
I'm also fine with archiving or moving to a separate page. Aaron Liu (talk) 17:42, 3 August 2024 (UTC)

The same type of issues still apply, though. Presumably, the entries were created because editors thought that others might question them, and we can't exactly know if there were any debates that didn't happen because an entry already existed here. Alternatively, some editors might prefer to have a general reference list for sources even if they don't meet the dictionary definition of "perennial", meaning that even uncontroversial entries still have value to them (although that may be an argument to create a separate list). Unless we have reason to believe they're no longer accurate, I think moving or archiving them will always make more sense than removal. Sunrise (talk) 17:33, 3 August 2024 (UTC)

Archiving or moving is fine with me. voorts (talk/contributions) 17:38, 3 August 2024 (UTC)

We could absolutely make some sort of "Non-perennial perennials" archive/page. Gråbergs Gråa Sång (talk) 17:43, 3 August 2024 (UTC)

That's actually not a bad idea. CNC (talk) 17:44, 3 August 2024 (UTC)

Getting paradoxical, eh?
Personally I'd call it "stale". Aaron Liu (talk) 17:51, 3 August 2024 (UTC)

NPP already has a much more comprehensive list than RSP, which includes everything here and more. If there's a desire for a general list of sources a secondary page based on the NPP list could be a place to start. -- LCU ActivelyDisinterested «@» °∆t° 19:32, 3 August 2024 (UTC)

• Oppose: I don't believe this will move the bar much and removing sources is moving the page in the wrong direction. Daniel Quinlan (talk) 01:11, 7 August 2024 (UTC)
• Oppose - Chicken or the egg. Lack of repeated discussion about included sources? That's a feature not a bug. Dcpoliticaljunkie (talk) 09:40, 11 August 2024 (UTC)

The page is called "perennial sources" because it's intended to be about sources that are the subject of repeated discussion. voorts (talk/contributions) 12:56, 11 August 2024 (UTC)

I would argue that the fact that inclusion has prevented redundant discussions of previously perennially discussed sources is good. Perhaps the introduction could be edited to reflect that, something like "The following presents a non-exhaustive list of sources whose reliability and use on Wikipedia were frequently discussed resulting in a clear consensus emerging." Or something a bit more elegant. Dcpoliticaljunkie (talk) 18:17, 11 August 2024 (UTC)

"reliability is frequently discussed or used on Wikipedia" would be the useful clarity here. I agree with above opinions that despite no discussions in the past 14 years for say Pew Research, the use of the source on Wikipedia remains frequent, and thus it's likely that it's inclusion on RSP has avoided further discussions at RSN. In summary it's as useful to include sources that are frequently used, that have previously been discussed enough so as to ascertain consensus on their reliability, rater than just including those that are frequently discussed. I'm aware that's not the original intention of the list, but it has by default become one of it's purposes. CNC (talk) 18:35, 11 August 2024 (UTC)

Articulated far more eloquently, thanks. I agree with the perennially "used on Wikipedia" addition. Dcpoliticaljunkie (talk) 08:56, 12 August 2024 (UTC)

• Oppose split. I agree with those above who are hesitant to remove anything; if anything, I'd like to see more added. Andre🚐 21:52, 23 September 2024 (UTC)

  This is a separate proposal from the split proposal. The split proposal does not involve removing anything. voorts (talk/contributions) 22:51, 23 September 2024 (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.

Tranclusion split partition scheme

Thanks S Marshall for the close of the page size discussion above. Now, we need to discuss a splitting scheme for the entries.

Currently, the section is about 400k chars long in Wikitext. WP:TalkCond, recommends archiving talk pages at 75k characters to solve editing. Since RSP is expected to grow, let's leave room for it to be around 50k per split subpage. That means we should have about 8 subpages for the total of 433 sources currently listed. From there, I counted the frequency of the first letter of the tables sortkey (basically what we assume is the first letter of the source), and made 8 groups with roughly 433÷8=54.33 repeating sources each. (#-A is a bit far from 54, but otherwise i'd have to lump B in, which would result in a really short X−Z group of 8 sources with no good way to solve. # is a much bigger range of characters anyway.)

With that in mind, here's the partitioning scheme I came up with:

Frequency of each first letter, grouped by my partitioning proposal
#-A: 41 #: 1 A: 40 B−D: 58 B: 19 C: 14 D: 25 E−G: 55 E: 13 F: 18 G: 24 H−L: 52 H: 15 I: 18 J: 7 K: 3 L: 9 M−N: 56 M: 26 N: 30 O−R: 55 O: 7 P: 22 Q: 4 R: 22 S−T: 57 S: 36 T: 21 U−Z: 49 U: 4 V: 13 W: 24 X: 2 Y: 2 Z: 4

To prevent having to move a ton of pages when having to change this scheme in the future, I would recommend giving each group's subpage numbered titles (e.g. Wikipedia:Reliable sources/Perennial sources/1 for the #−A group) while still listing out the groups on the full RSP page (and as a navbar template on top of each subpage) like this, with each subpage linked:

• #−A
• B−D
• E−G
• H−L
• M−N
• O−R
• S−T
• U−Z

Also, hopefully we can agree with using labeled section transclusion to mark deprecated sources, as demonstrated at User:Aaron Liu/sandbox/RSP Demo.

• Remsense
• Ivanvector
• CommunityNotesContributor
• Mfko
• ActivelyDisinterested
• Gråbergs Gråa Sång
• Daniel Quinlan
• Apenguinlover

Aaron Liu (talk) 21:06, 2 October 2024 (UTC)

Very good ideas. Just make sure that what is first seen is a table of contents and a link to the full list. Apenguinlover(🐧!) 23:26, 2 October 2024 (UTC)

I think separating the table and the other stuff that's currently on RSP is a good idea but out of scope for the splitting discussion. We can discuss that after finishing the transclusion stuff. Aaron Liu (talk) 23:37, 2 October 2024 (UTC)

Actually, ${433 \over 8}=54.125\neq 54{1 \over 3}$. So there would be, on average, 54.125 sources per subpage. Anyway, I support the split to make the list easier to navigate from places where hardware is often outdated, but also from places where Internet connections are not everywhere reliable. Alfa-ketosav (talk) 13:46, 7 October 2024 (UTC)