This is an archive of past discussions about Wikipedia:Reliable sources. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
That is potentially confusing. I would change it to "No consensus" for clarity, and then the readers can click on noticeboard links to get more info. Also, my reading of the The Daily Wire discussion suggests that it's at best "Questionable"; i.e. more editors speak against it as RS than for it. --K.e.coffman (talk) 18:40, 28 July 2018 (UTC)
OK, I will change the middle category name. Feel free to change the Daily Wire to "questionable" if that's how you read the discussion.- MrX 🖋 21:58, 28 July 2018 (UTC)
Perennial sources
Can we have some guideline on what is a "perennial source". We know them when we see them, but a guideline would still be useful. If we don't have a guideline, we will have to debate every new entry (or deletion) on its own. Work permit (talk) 22:36, 30 July 2018 (UTC)
Good question. The guideline that I used was at least two discussions at WP:RSN, or at least one WP:RSN discussion and multiple talk page discussions about the source. I believe that all of the sources currently on the list meet those criteria, and probably several more that have not yet made it to the list. - MrX 🖋 22:45, 30 July 2018 (UTC)
Please add TechCrunch
Please could those with time add an entry for TechCrunch. Easily meets the requirements above, with a comment today suggesting that the consideration is consistent now. MPS1992 (talk) 00:27, 4 August 2018 (UTC)
Sure thing. I'll try to find time to dig through the RSN discussions tomorrow morning if no one beats me to it. GMGtalk00:30, 4 August 2018 (UTC)
Done Incidentally, I'm not really sure we're providing much value added here if we simply say in the notes section "there is no consensus", and we should instead be trying as much as possible to summarize the nuances of the particular discussions. I may try to go back over time and provide more in depth summaries. Of course, anyone is welcome to help with that. GMGtalk12:43, 4 August 2018 (UTC)
Democracy Now
A friend remembered me of DN tonight and as such I searched RSN and realized that it has a number of threads. Maybe enough to fit on this list? Thanks, —PaleoNeonate – 08:03, 9 August 2018 (UTC)
‘Popular press’, ‘Media concentration’, ‘State-sponsored or controlled’
Given that only a fraction of the ‘popular press’ (a term used in The NY Times entry) are covered and that ‘media concentration’ is not, it would seem useful to have an entry for the latter. Also, a pointer to discussions of ‘state-controlled’, state-sponsored, and other such groupings might also be appropriate. Thoughts? Humanengr (talk) 01:35, 12 August 2018 (UTC)
It contains the advice or opinions of one or more Wikipedia contributors. This page is not an encyclopedia article, nor is it one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community. Some essays represent widespread norms; others only represent minority viewpoints.
Personally, I prefer the last choice. None of the other three seem to "fit" was this page really is: A summary list of previous discussions. Work permit (talk) 02:24, 13 August 2018 (UTC)
I agree with the general intent, but I think we should try to come up with a common grouping that applies to other similar pages. There's supposed to be a discussion to that purpose at Wikipedia talk:Project namespace, although not very many people have commented there yet. Tamwin (talk) 00:09, 14 August 2018 (UTC)
Telesur
Telesur does not seem to be a subject of recurring discussions. How do we avoid this list becoming an index to every source that’s ever been discussed? Any objections to deleting it? Work permit (talk) 18:51, 11 August 2018 (UTC)
Maybe a link to WP:Potentially unreliable sources would work. I think that's probably the right place for Telesur right now, even if we can already be fairly confident that more extensive discussion would result in treating it like similar sources. A specific note to clarify that we can refer to similar sources already on the list might also help, since otherwise there's a (reasonable) incentive to copy existing descriptions multiple times as clarification. Another approach might be to divide the table into subcategories if it gets too large. We could possibly set explicit minimum requirements for amount of discussion, though they'd probably need to be set fairly low to account for all the possible situations. Sunrise(talk)11:08, 13 August 2018 (UTC)
Good point. WP:PUS is as you say different then WP:RS/P since the purpose of this page IS to be a list of prior discussions. When a controvery comes up, and editor can come to this page and find the link of actual discussions about sources. Quoting those sources in an article's talk page can carry some weight. WP:PUS is truly an essay. Different use case.Work permit (talk) 16:47, 14 August 2018 (UTC)
And the use case for the essay given the presence of WP:RS/P? The latter seems (potentially if not already) of higher value to me as a reference. I'm wondering about whether and how to better integrate. Also, in your view, how well do the 'summary' comments on WP:RS/P comport with matching items on WP:PUS? Thx, Humanengr (talk) 17:41, 14 August 2018 (UTC)
If I were in a dispute over a source I would use this page, and refer to the linked discussions. I would never use WP:PUS. If I were learning more about categories in general, I would find WP:PUS useful. Work permit (talk) 17:53, 14 August 2018 (UTC)
@Work permit: No objections for its deletion for now, but there are multiple talk page dicussions in Venezuelan articles that quickly die down after the government bias is brought up. Maybe another discussion to make a determination? I think the concensus would be similar to PressTV or RT, but that is for other users to decide.----ZiaLater (talk) 06:07, 14 August 2018 (UTC)
On the RFC, user:Humanengr felt " the name ‘Perennial sources’ provides a bit too much imprimatur". "Perennial" may have a negative connotation to some editors? How do people feel like changing it to something like "Recurringly discussed" or "Frequently discussed"? Work permit (talk) 02:52, 5 August 2018 (UTC)
"Recurring discussions" seems to fit alright. "Perennial" probably does have a somewhat negative connotation in the context of Wikipedia. In many circumstances here it strongly implies "lost hope", which is not the case at times with sources. I can't recall which one off the top of my head, but there was one "techy" source that recently fell into favor after getting its game together and doing serious reporting. GMGtalk04:34, 5 August 2018 (UTC)
Aaaah … Thx for that. What tripped me up is that ‘Perennial’ in the title modifies ‘sources’, but the intent is that this page is about “Perenially discussed sources”. That’s an alternative fix. I’m much less sensitive to choice of modifier than target. So perenially, frequently, repeatedly, recurringly, reoccuringly?, etc., all work well enough as long as they modify ‘discussed’ rather than ‘sources’. Humanengr (talk) 09:59, 5 August 2018 (UTC)
Good point. Would a page move while this is under discussion be too complicated? Should it be done after the Rfc is closed? Work permit (talk) 15:27, 5 August 2018 (UTC)
Not sure where to look for answer other than WP:TEA. I don’t think it would be a significant disruption. And for me (and at least some others), clarity rules. Humanengr (talk) 18:48, 17 August 2018 (UTC)
Oppose - The title was selected for consistency with WP:EL/P, a supplement that has been very helpful in providing guidance for the use of external links. "Perennial" in this context means that the same discussions keep coming back. That parlance is frequently used on noticeboards and article talk pages by experienced editors. While 'perennially discussed sources' would be slightly more grammatical, titles should be concise and recognizable. - MrX 🖋 15:51, 5 August 2018 (UTC)
@MrX, By being “for experienced editors”, ‘parlance’ misleads everybody else. The fix is not only grammatical in a structured sense — but extremely important to the general reader in terms of connotation and hidden bias. Do you have a policy reference for use of ‘perennial’ in such titles given that WP:EL/P is an 'Explanatory Supplement' that says "This page is not one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community." Humanengr (talk) 17:19, 5 August 2018 (UTC)
I expect the best way forward here is to determine in the RfC whether such a list of some type is in principle helpful. Once that is determined, we can decide what the best name would be through discussion here or an RM, and have what will probably be many follow up discussions about format and scope. GMGtalk16:39, 5 August 2018 (UTC)
Thx for that research. ‘Perennial proposals’ and ‘Perennial requests’ both implicitly embed a type of ‘discussion’ artifact in the target term, and are therefore not as misleading. The target of ‘Perennial website’ has no positive or negative connotation. The instant case ‘Perennial sources’ has as target a word meaningful — in a positive way — to the entire WP community. As a phrase, it conveys a stature similar to ‘Reliable sources’ and is thereby the most misleading of this set. None of those pages has been thoroughly vetted. For these reasons, this convention should be used only where the risks are of little consequence. ‘Perennially discussed sources’ both carries little risk of misinterpretation and can be viewed as an extended family member to the examples you provided, and so should be satisfactory to all. Humanengr (talk) 11:47, 6 August 2018 (UTC)
We are looking at when there is a general consensus for or against something being a reliable source, and also include perennial sources for which we have not come to a general consensus. In those, the wording said "there is no consensus that X is a reliable source" (emphasis mine). The implication of that wording is that we start from a default negative position (i.e. when something is generally considered unreliable it is also true that there is no consensus that it is a reliable source), when what is intended is more like "there is no consensus that X is a reliable source and no consensus that X is an unreliable source". As such I've changed instances of that language to "there is no consensus about the reliability of X". — Rhododendritestalk \\ 17:05, 24 August 2018 (UTC)
Designers & Dragons
Would someone consider adding the below to the perennial sources grid? I think this represents an accurate summary of the latest discussion, however, since I was a party to it I should probably not add it and it should be checked by an uninvolved person to determine whether it is an accurate representation of the discussion. Chetsford (talk) 18:35, 22 August 2018 (UTC)
Designers & Dragons is a book on the history of roleplaying games that has been published in several editions, both by Mongoose Games and Evil Hat Productions. Editors generally agree that Designers & Dragons is reliable for non-extraordinary claims regarding games and game companies. There is no consensus as to whether it is, or is not, a reliable source for biographical statements or WP:BLPs.
Before adding it I believe an uninvolved third party should review the summary. I believe there is also a moderate consensus that, for non-BLP, it is OK for facts but not notability. Jbh Talk19:56, 22 August 2018 (UTC)
There is certainly not a consensus, much less a policy-based consensus, that Designers & Dragons would not contribute to Notability. First of all, there is no policy basis for a category of sources regarded as reliable where SIGCOV would not contribute to Notability. Second, only a couple of editors have even articulated this position, so it is hardly a consensus even if it were policy-compliant (which it is not). The GNG is met by multiple significant mentions in RS. Newimpartial (talk) 21:20, 22 August 2018 (UTC)
Chetsford, I was responding to I believe there is also a moderate consensus that, for non-BLP, it is OK for facts but not notability[2] to which you replied Agree on both.[3] I am glad that we now have an understanding that no such "moderate consensus" exists.[4]
By the way, I agree in general with your formulation, Editors generally agree that Designers & Dragons is reliable for non-extraordinary claims regarding games and game companies. There is no consensus as to whether it is, or is not, a reliable source for biographical statements or WP:BLPs[5], though of course I do not agree with you or jbh about the underlying policy issues discussed there. Newimpartial (talk) 22:28, 22 August 2018 (UTC)
I'm not sure we're understanding each other here, but it may be an indenting issue. In any case, I'll leave it to a forthcoming third party to referee the wording of this addition. Thanks - Chetsford (talk) 22:30, 22 August 2018 (UTC)
Of what addition? There should be no added text re: a "moderate consensus" against WP:N, for which no significant number of contributors at RSN expressed support. Newimpartial (talk) 22:34, 22 August 2018 (UTC)
Both of you. Give it a rest! An uninvolved editor will sort stuff out. All you are accomplishing is spreading your BS all over the place. Keep it up and I'll take you both to ANI and ask for an interaction ban. Jbh Talk22:47, 22 August 2018 (UTC)
Seems to me one discussion diff does not a "perennial source" make, unless it's to a formally closed rfc that's predicated on previous discussions being unclear. Is this page also supposed to be an index of reasonably well attended rsn threads? — Rhododendritestalk \\ 23:30, 22 August 2018 (UTC)
That may be a valid point, Rhododendrites. This has recently been the topic of discussion in a number of AfDs; the link above was just to a formal, centralized discussion at RSN that arose out of those. That said, perhaps several discussions are needed before qualifying for this list? Pinging Mrx to see if he can clarify. Chetsford (talk) 23:41, 22 August 2018 (UTC)
A single discussion about a source would not really qualify as perennial. Have there been previous discussions about the reliability of this source?- MrX 🖋 01:42, 23 August 2018 (UTC)
It's been a central topic in these recent AfDs [6], [7], [8], has been added/subtracted here [9], and also figured tangentially in an ANI thread. I'm not sure if that's enough to pass the threshold, though. Chetsford (talk) 01:59, 23 August 2018 (UTC)
While not unrelated, I'm inclined to think that this page would best be served by inclusion only of well-attended RSN threads and/or RfCs. XfD wraps up all sorts of other issues, and many people wind up talking about reliability of a source when they're really talking about whether it's in-depth coverage, whether it's primary vs. secondary, whether it establishes notability, etc. — Rhododendritestalk \\ 16:54, 24 August 2018 (UTC)
I would agree with the above that at a minimum we would need two or more such discussions for a source to qualify for this list. Ideally that would be "more" rather than "two". The point of this page is to mostly avoid discussing the same source for a fifth or sixth time when every previous discussion has more-or-less reached a similar conclusion that is unlikely to change. So, for example, before someone opens the 7,912,607th thread on whether the New York Times is unreliable because liberal fake news, they would ideally come here first to ensure that they're not starting the exact same discussion we've already had 7,912,606 times.
Conversely, if a source has only really been evaluated at RSN once, then a second discussion is much more likely to actually be fruitful, to reach a slightly different conclusion, or to reach a similar conclusion that is more nuanced. So we don't want to give the impression that it's a "done deal" and discourage further debate, in the same way that we actually kindof do want to nip the liberal-fake-news-failing-New-York-Times discussion in the bud, because it's simply a consummate waste of time. GMGtalk17:12, 24 August 2018 (UTC)
"Perennial sources" vs. sources with one discussion link
As came up in the thread above about Designers & Dragons, I don't think it's appropriate to include a source on here, as a "perennial source," linking to only a single thread that discusses it. An exception could be like the Daily Mail, where we only really need to link one decisive RfC that's predicated on many other discussions.
Accordingly, I've removed Daily Wire, InfoWars, and Mediaite. I searched RSN for each of them before removing them (and in the case of Newsbusters added a couple threads to the one that was there). I was surprised we didn't have much pointed discussion about InfoWars, since that seems pretty widely regarded as unreliable. But it's not a list of sources we know are terrible so why bother discuss them -- it's a list of sources we have to repeatedly discuss. As such we also don't have Wikipedia, Perez Hilton, Stormfront, etc. despite all of them having come up somewhere in the past. If there are some clear discussions about InfoWars that have taken place elsewhere, I'd encourage the addition of those links to the discussion. At the moment, it's just a single link with some RS (this isn't the place to make a new argument that something is unreliable).
Looks like while I was writing this MrX reverted the removal of InfoWars. While I don't disagree it's unreliable, I think this page needs a clear inclusion criteria. As "perennial sources," infowars does not look to make the cut (something which we could probably address via a Daily Mail type RfC to formalize it -- it's one of the few that I think would be successful in such a proposal). — Rhododendritestalk \\ 17:50, 24 August 2018 (UTC)
Hmm. Infowars is an interesting test case. It's actually discussed a heckuva lot, but it looks like the vast majority of the time, it's in the context of being considered so universally unreliable that it's used as the bar to gauge the unreliability of other sources. GMGtalk18:00, 24 August 2018 (UTC)
I think anything that keeps popping up one talk or WP pages should be on this list, especially if the Wikipedia article says that it's "a far-right American conspiracy theorist and fake news website". The purpose of this list is to provide a useful guide, a goal which is not furthered by focusing on strict adherence to an inclusion criterion. Remember WP:NOTBURO - MrX 🖋 18:04, 24 August 2018 (UTC)
It doesn't seem like a NOTBURO matter to say that if this page intends to document and summarize discussions of sources that frequently come up, it's not a great practice to include something with one linked discussion supplemented by external sources (which is the basis for the arguments in discussions we're indexing here, not for this page itself). I agree that InfoWars is a decent test case because nobody here is going to disagree that it's terrible, but this isn't the place for stating/deciding that something is terrible except insofar as it's well established by linked discussions. In the case of InfoWars, I would encourage anyone who has seen discussions of its reliability in other venues to just add those to the table. — Rhododendritestalk \\ 18:41, 24 August 2018 (UTC)
It is perennially discussed just say in a different way than say Find a Grave. I think we could take any number of the discussions on its talk page (or talk page archive) as evidence of its discussion as an unreliable source. I think we have no shortage of discussion based evidence that Infowars is not RS. Best, Barkeep49 (talk) 19:10, 24 August 2018 (UTC)
But that's not the point. The point is that this page is useful for people who want a record of past discussions. It's not a place where we should just put what's "obviously" an unreliable source, because the presumption is that anyone who would benefit from this page is either looking for the links to past threads to support an argument or doesn't know it's obvious despite what the article says. I'm not saying it hasn't been argued, but that it shouldn't be included here without finding those links. I would do so if I had any firsthand experience with them. We could also add Wikipedia itself, things written on bathroom walls, Perez Hilton, Stormfront, and YouTube comments, because it would be easy to find a place where someone says "Stormfront is not reliable," but this isn't the Wikipedia:Potentially unreliable sources, it's documentation of discussions. Glossing that because you and I know it's been discussed somewhere isn't the point. Anyway, that's all I'll say on it. — Rhododendritestalk \\ 19:19, 24 August 2018 (UTC)
I think I agree with your core point - one discussion does not make it a perennial source. What instead I'm suggesting is that the discussion about Info Wars has happened on its talk page (or perhaps other talk pages) rather than RSN and that is then OK for inclusion here. Best, Barkeep49 (talk) 19:31, 24 August 2018 (UTC)
Linking to previous discussions is one of the purposes of this page, but not the only one. It would easy enough to find multiple discussions where editors have discussed the reliability of Infowars. Best practice would be to link a couple, or a few, of the best of these discussions, but even one good discussion is better than leaving it off the list altogether.- MrX 🖋 22:40, 24 August 2018 (UTC)
My main concern is that if we get into linking to article talk page discussions, were going to get into linking to very local consensus, which may vary greatly compared to more broad consensus. GMGtalk23:02, 24 August 2018 (UTC)
If InfoWars is as unreliable as you describe, then a RfC should easily confirm this. I've posted a RfC on WP:RSN asking editors whether we should make the same conclusions on InfoWars as we did in the Daily Mail RfC. We can add this RfC to the table after it is closed. — Newslingertalk07:07, 30 August 2018 (UTC)
Youtube?
Surprised to notice Youtube on the list. Youtube is not a source, it's a video hosting site. Most of Youtube is unreliable for the same reason most of the web is unreliable, but it's entirely dependent on who published it. By default something you found on Youtube is not going to be reliable, just like by default a blog or other website you find won't be reliable, but there are plenty of organizations that publish content on Youtube that has as high or higher standards as their other media. Taking a look at the linked discussions, this is basically what I see. That does not sound to me that it is useful to say it is generally unreliable as a source (as opposed to most of the content it hosts being unreliable). — Rhododendritestalk \\ 16:01, 29 August 2018 (UTC)
I agree with your assessment. Hmm this is where the notes section becomes useful for the current model, it seems. Generally unreliable is still true "in general", yet specific channels can be. Maybe that an additional item that could be added along with "generally unreliable" (both for the same source, or one replacing both for that source) could be "selective content reliable" or similar? —PaleoNeonate – 17:55, 29 August 2018 (UTC)
Yes, this is going to be a problem. "This page says YouTube is unreliable! I'm reverting your addition of a news story posted at https://www.youtube.com/user/bbcnews!" is a predictable outcome.
If you're going to have this page, I think you need a fourth category for "it depends" – not good/bad/undecided, but for those platforms that we have clearly decided depend upon factors other than the domain name in the URL. WhatamIdoing (talk) 16:33, 30 August 2018 (UTC)
I'm hoping that the description is sufficient for this kind of thing. Any broad categorizations like those on this page will have issues like this, since there will always be particular circumstances where an otherwise unreliable source is acceptable or vice versa. I believe that's also part of what the earlier parts of the page are supposed to be getting at, although perhaps they could be improved by a discussion of specific cases (as in the common example "all sources are primary sources for their own opinion"). If misunderstandings occur anyways, or if there's consensus on some specific important type of selective reliability, one solution might be to split the entry into two parts as in the case of Forbes. Sunrise(talk)01:46, 31 August 2018 (UTC)
It is my experience that it's not safe to assume that anyone reads the directions.
There are things that could be done to improve this. If, for example, by "Amazon.com" we actually mean "User reviews at Amazon.com", then we could just say that in the first column. But overall I think there are significant limitations to this format. The design itself, with its strict demarcation between color-coded good/bad/undecided categories, encourages mindless rule-enforcing rather than thinking about exceptions and context. WhatamIdoing (talk) 16:36, 31 August 2018 (UTC)
Accessibility concerns
an example of a scrolling Nightmare and great sentence fragmentation.
We need to fix this graph chart.... almost impossible to read paragraphs with so much sentence fragment and the need to scroll so much just to read one paragraph. This is an example of what not to do.... great example of how to lose readership to a page.... no one's going to scroll forever hoping to find something..... need normal text with a table of contents. Page will simply be ignored if no one can find what they're looking for or be able to read anything. Should make it look like Wikipedia:External links/Perennial websites--Moxy (talk) 22:57, 24 August 2018 (UTC)
I'm not sure we can do very much about the view on mobile. The WMF hasn't really given us a lot of tools to make that work right. GMGtalk23:15, 24 August 2018 (UTC)
On what resolution? I'm 1366x768 on a fairly crappy laptop, and it looks nothing like that. The notes section takes up about half the screen width. I'm mostly just bewildered. GMGtalk01:16, 25 August 2018 (UTC)
Since TVs are usually viewed from a much longer distance than laptop screens, your display is scaled to enlarge the browser content (and system interface) by default. Your screenshot has an effective resolution of around 800x600, which is the same effective resolution as a smartphone. I'm not sure which version of Android TV you're using, but Android 7.0 and higher has screen zoom settings that allow you to scale down your screen to increase its effective resolution. (I'm also not sure if these settings are available in Android TV.) — Newslingertalk22:36, 25 August 2018 (UTC)
I also noticed that the "notes" column was squeezed at common resolutions (even on HD screens a fullscreen browser window is suboptimal for normal reading). —PaleoNeonate – 23:18, 24 August 2018 (UTC)
I just looked at it on a mobile device, and while it's not ideal, it is useable. In creating this page, I chose a table so that the columns could be sorted, so that the background colors would quickly identify the level of reliability, and so that that the information would have an underlying structure. That would not be possible in a prose format. I also don't see the value of a table of contents, although I guess it could be favorable on mobile devices. - MrX 🖋 02:02, 25 August 2018 (UTC)
With no table of contents you have to scroll for days to even know if the site your looking for is even here. Should never pick color over accessibility. Why does it need to be sorted. ....how does that help readers search any specific item on the list. If the page is to be take seriously it should be useful.--Moxy (talk) 02:10, 25 August 2018 (UTC)
We agree it should be useful, but I'm skeptical that your approach would be a better way of presenting this information. I assume that the vast majority of users who would access this guide would be serious editors using desktop or laptop devices, as evidenced by the page view stats. I would also mention that we use tables in many, many articles without significant complaint. By the way, we can add shortcuts now, as I have already done for WP:DAILYMAIL.- MrX 🖋 13:00, 25 August 2018 (UTC)
That still expects the software client and reader to fix any problems (while accessibility adapts to many scenarios). But that could be improved over time, especially if it gets mainstream enough to generate more end-user input; another idea could be an eventual index/toc at the top... —PaleoNeonate – 18:06, 29 August 2018 (UTC)
I agree. The "notes" section is very important. Yet very difficult to read. I've made two edits, each of which shrinks the other columns to give more space to the "notes" section. Curious to hear what people think. Work permit (talk) 18:28, 29 August 2018 (UTC)
The status column now seems squeezed for me but it's still shorter vertically than the text one, which now seems to breathe a little, so it's likely an improvement. Thanks, —PaleoNeonate – 18:37, 29 August 2018 (UTC)
I've now put in a third change, which has a main column header for "Discussions" with secondary columns for "list" "last" and "summary". This squeezes as much space as possible for the "Notes" (now "summary") section. Work permit (talk) 18:40, 29 August 2018 (UTC)
Last change. Label the discussions 1,2,3, etc to eliminate the brackets ("[]"). Keep numbers sequential to each source, to keep number from becoming double digit. Finally insert a break after two. This has minimized the width of the discussion list. I think, finally, the summary is readableWork permit (talk) 01:27, 30 August 2018 (UTC)
Additional Change. To reduce the clutter noted above in the status column, I moved the description to a footnote. In addition, I feel its a bit repetitive to keep including the text since the status is both color coded AND has a BIG symbol.Work permit (talk) 04:34, 30 August 2018 (UTC)
Considering that the footnotes don't provide any more information than the icons, I don't think the footnotes are necessary at all. The icons and the legend are sufficient. It would probably be better to place the legend before the chart to ensure that there is no confusion with the icons' meanings. — Newslingertalk05:24, 30 August 2018 (UTC)
Thanks. I’ve always used footnotes to clarify points in a table, or to convey repetitive information about entries. Could you mock up an example of how you would put the description of the icons in the legend/caption of the table? Maybe we could get more editors take on this as well. Work permit (talk) 05:39, 30 August 2018 (UTC)
How about this? My wording wasn't the best. I meant that the references to the footnotes could be removed. It might also be useful to expand the short labels ("Generally reliable") with a more detailed description clarifying that this is a general designation, not a universal one, and informing readers to use their independent judgment for every situation. — Newslingertalk06:36, 30 August 2018 (UTC)
I’m glad we both agree on getting the repetive text out of the table. That was my main desire. Your approach could work too, though I’m not used to seeing a legend on top. One nice thing about your approach is that the icons can be centered in the column, which would look better then my version. Also, wiki editing superscripts on images don’t format very well. Lets give it a day or two and hear from some other editors.Work permit (talk) 07:09, 30 August 2018 (UTC)
Yes, it would be nice to hear from other editors. The only reason I suggested placing the legend above the table was to help address Markbassett's concern (from #This seems a bad idea) that a reader who skims the page could misinterpret the icons to think that a source is universally endorsed or strictly prohibited. — Newslingertalk06:52, 1 September 2018 (UTC)
True. But as you see, another editor disagreed and put it back on top. Regarding Markbassets concern, having the text on top makes little difference. We have a big NO symbol and a red color background. That is a very strong visual. The text description will be lost on the reader, no matter where you put it. Work permit (talk) 07:07, 1 September 2018 (UTC)
Listing requirements
I think it would be useful to define specific listing requirements for this page. So far, we've generally been listing sources with:
The only entry on the list that doesn't meet these requirements is the one for Venezuelanalysis.
When I added the 32 non-RfC discussions for the Daily Mail, I encountered two issues: some discussions (like discussion #1 and discussion #2) are very short, and some discussions (like this RfC and this discussion) aren't particularly relevant for the purposes of assessing reliability. I've included the short discussions, but excluded the irrelevant ones.
Should the above requirements be included in the page? Also, should there be length or scope criteria for discussions to count toward these requirements? — Newslingertalk08:58, 3 September 2018 (UTC)
It's always good to explicitly state inclusion criteria.
I'd guess that this list could explode to the point of being difficult to manage and use without more restrictive criteria.
What's the feeling about switching to something like this for the non consensus sources? To my mind, something like this is basically a warning, that effectively communicates "don't use this source". But that's not really what no consensus means in this context. In this context is more often than not means that we use this source all the time, but some people have taken issue with it, or it may not be appropriate in some contexts. But as it stands, we've got two effectively warning symbols, and one "good-to-go" symbol. GMGtalk12:13, 3 September 2018 (UTC)
I think the caution symbol is appropriate because editors should always be cautious about using a source that has perennially failed to gain consensus as reliable. For example, The Daily Caller and TMZ are routinely rejected as reliable.- MrX 🖋 15:16, 3 September 2018 (UTC)
I think that the question mark would be better for "undecided".
Editors should be careful about how they use any source, even the ones that some editors might describe as being always good. Nothing is always good for every situation. WhatamIdoing (talk) 15:54, 3 September 2018 (UTC)
I mean, my thought is that if they are routinely rejected, then they should be marked as unreliable. If there is no consensus, if there is an absence of consensus, then that puts the source in the same category as every source that has never been discussed at all. Those sources also have no consensus. Going through these one-by-one, it seems that most results of no-con are fundamentally a finding of "nuance is required". GMGtalk16:19, 3 September 2018 (UTC)
Comment: no consensus = caution; while the question mark sign is more "has not been discussed yet", at least to me. I feel that the present icon () is an appropriate choice. K.e.coffman (talk) 22:38, 4 September 2018 (UTC)
That seems reasonable. I see the question mark as meaning "There's still an open question about this one", and the warning triangle as meaning "Warning, there are problems here". Maybe we need something else entirely. WhatamIdoing (talk) 03:46, 6 September 2018 (UTC)
I'm a little conflicted about this. I'm not sure if this is a positive side effect of this list, in that it helps us find gaps in codified consensus, or whether it's a negative side effect because it's going to encourage us to add layers of bureaucracy for the sole purpose of adding items to the list, when no specific content dispute is being addressed. I'm not sure at what WP:POINT that becomes problematic. GMGtalk10:52, 30 August 2018 (UTC)
I see the danger you mention. Infowars may be an outlier, since it is so obviously unreliable and yet there was only one discussion referenced. Lets hope this doesn’t, as you fear, get out of hand.Work permit (talk) 15:49, 30 August 2018 (UTC)
Version 4.0 of a popular chart (August 2018)[1] created by patent attorney Vanessa Otero is very carefully researched, and I highly recommend its use. Yes, it's not a RS here....yet, but it's the best I know of. Use sites in the yellow and green boxes as sources, and generally avoid the rest. She describes what's new in version 4.0 here.[2]
The latest version has individual charts for the programs shown on CNN,[3] MSNBC,[4] Fox,[5] and TYT Network.[6]
Some other articles about the chart:
2015 "How does your favorite news source rate on the 'truthiness' scale? Consult this chart"[7] is a short article about the chart.
2018 "How biased is your news source? You probably won't agree with this chart"[8]
Since then she has created an even better version, released September 4, 2018.[1] Her article about it is located here:
The chart shows the intricate relationship between opinions and facts, and how partisanship affects them. Although it's possible for a source to be fairly partisan and still factually accurate, there comes a point at which partisanship and spin begin to twist and distort facts.
The more a source allows opinions to dominate facts, the further away from the purely-factual center and the raw-news-reporting top it moves. That's why Fox News rates both low and extreme. Even sources like CNN, where analysis isn't as high as some others, still remains fairly close to center, but lower down than some other deep analysis sources. Note the colored boxes.
Otero's May 25, 2018 update:
"Why Measuring Political Bias is So Hard, and How We Can Do It Anyway: The Media Bias Chart Horizontal Axis. Post One of a Four-Part Series. May 25, 2018."[9]
Sources
^ abOtero, Vanessa (September 4, 2018). "Media Bias Chart 4.0". Ad Fontes Media. Retrieved September 4, 2018.
Hey BullRangifer. This page isn't really a venue for evaluating sources in and of itself. This just summarizes existing community discussions. If you feel that a source need to be vetted by the community, then the proper place for that is WP:RSN. GMGtalk22:26, 8 September 2018 (UTC)
The Mirror
It is interesting that of the British tabloids, The Sun, The Express, and the Daily Mail are included on this list but not The Mirror, which differs only in its political slant to the other tabs. Making a list of "untrustworthy" sources, and triggering alerts when they are linked to, but only including those that speak from a particular part of the political spectrum, is hardly NPOV. FOARP (talk) 13:25, 11 September 2018 (UTC)
Well, this is supposed to be a list of repeatedly discussed sources, and not a list of untrustworthy sources per se. I don't know if the Mirror has been discussed enough to be included or not. I don't know if there is a bias in this list or not. If there is, it ought to be a reflection of whatever biases are present in community discussions at WP:RSN, and the solution would be to increase the diversity of participants in those discussions. GMGtalk13:39, 11 September 2018 (UTC)
Hey IJBall. This list has only been under construction for a couple of months now, and I expect it will grow substantially as time goes on. If it is overly focused on US sources, that's probably an artifact of the comparatively small number of editors who have worked on it so far. If these sources have been repeatedly discussed, then feel free to add them to the list. If you're not comfortable with that, you can start a discussion for a source here on the talk page and we can work to incorporate it in the list that way. GMGtalk12:44, 18 September 2018 (UTC)
OK, understood... However, I would advise adding IMDb to the list now, as it has been discussed multiple times before, already has a couple of specific pages discussing it (I linked to those above; another one that does is WP:ELPEREN...), and newbie editors definitely make the mistake of using IMDb as an "inline source" in their attempts at new articles. --IJBall (contribs • talk) 12:47, 18 September 2018 (UTC)
@IJBall: Done - By my reading of the large number of previous discussions, the overall feeling is that it is unreliable. That's not to say there aren't dissenters, but those don't seem to rise to the level of no-consensus. GMGtalk13:58, 18 September 2018 (UTC)
Threads closed
RSN threads have been closed for Breitbart and Occupy Democrats and need to be added. I haven't had enough coffee yet for that much formatting at 6:30 local time. GMGtalk10:24, 25 September 2018 (UTC)
@Sunrise: could you elaborate your assessment when adding Quackwatch as a reliable source to list? [11] I just went through all of the four discussions and they were all rather bitterly divided, with strongest criticism of the source in the two most recent discussions. There were also serious WP:COI issues in pushing it to be a reliable source (see query about a conflict-of-interest here). The statement of "Quackwatch is generally considered a reliable source" would be false based on those four threads. My own assessment would be that it can be used case-by-case but it's not greenlit as reliable. Therefore, I would change it from "reliable" to "no consensus" and adjust the text accordingly. --Pudeo (talk) 09:35, 19 October 2018 (UTC)
While I certainly wouldn't consider it the pinnacle of reliability, in many cases I would consider it a valuable source - the website is maintained by a subject-matter expert but not peer-reviewed, and therefore is appropriate in some circumstances but not in others. I would say that whether this qualifies as "green" or "yellow" in the context of this list is mostly a matter of taste, and is part of the challenge of creating a simplified classification scheme like this one. (I don't think an individual editor's COI is relevant though, since it doesn't affect the merits or demerits of the source itself.)
I would also note that some of the "division" you're referring to comes from a couple of editors (later indeffed or topic-banned) who were attempting at great length to discredit it, with the majority of editors refuting or ignoring their comments (while acknowledging the source's limitations). Some context may be required for the full analysis, but I would normally classify the circumstances as civil POV pushing; out of the last two links I think this is clearest in the first. The latter is indeed quite divided, but not over reliability as much as policy recommendations; I would also keep in mind that as a MEDRS discussion it's being held to a much higher standard. Sunrise(talk)23:03, 21 October 2018 (UTC)
@Sunrise: I agree that the three-step classification isn't always easy, but I would disagree to what extent the topic-banned or blocked users played in the discussion. I saw generally well-respected editors in the two newest ones criticizing the source. I will refrain from new comments about the source itself, because this page isn't for discussing the source, but for summarizing the previous discussions. Needless to say, this topic has a history of disputes because the source was already the core of a 2007 ArbCom case. But I don't object to your recent edit. I would also thank @Newslinger: for spending a lot of effort (more than I did) for finding several old RSN discussions and summarizing them well. --Pudeo (talk) 23:44, 21 October 2018 (UTC)
Pudeo, the ArbCom case was spurious, and two people were community banned as a result of their instigation of the case. Unfortunately some ArbCom members didn't perform due diligence (one, a disbarred lawyer in RL, wrote a "finding" before any evidence was presented! He supports quackery.) and allowed part of the false charges to end up as part of the "findings", although it was never examined, much less "found". That was the wording that QW was an "unreliable source". In 2008, the matter was fixed, and both myself and QW were vindicated. You can read about it here. So the ArbCom case, far from being points against QW, actually is evidence favorable to QW. Those who falsely accuse QW of being unreliable and attack its defender get community banned. Literal quacks started that case. That says a lot in favor of QW.
I strongly recommend that you read this: Wikipedia_talk:Requests_for_arbitration/Barrett_v._Rosenthal#Request_to_amend_prior_case:_Wikipedia:Requests_for_arbitration.2FBarrett_v._Rosenthal. For some odd reason a wording that I "wrote for the site" appears in a comment from my defender, Shoemaker's Holiday, but that's not true. I never did that. That was a false accusation that just got repeated. ArbCom affairs are very disruptive, messy, and nasty. I nearly offed myself during that case, and I have never fully recovered. It left deep scars. Being falsely accused is a terrible thing for someone raised with a partial Asian culture. (Born in Japan, and lived in Korea and the Philippines. Father born in Korea.) Saving face means a lot in our family. -- BullRangifer (talk) PingMe05:22, 22 October 2018 (UTC)
When I wrote it I tried to include only the most representative examples, which was how this page was set up - no entry had more than 5 links at the time. Since it looks like things have changed in the last couple months, I have no objection to adding the rest.
I think the newer version is mostly fine, and in fact I don't see it as being all that different. I've made one change, which is to replace the initial sentence (the statement of no consensus) since in my view, it is more that there is a consensus for reliability in some cases and unreliability in others. I described it as an expert SPS instead, since that's the basis for justifying its use in the cases when it is appropriate. Based on the WP page it may actually have grown beyond that by now, so this may need to be tweaked, but at the least that's how it started out.
For the rest, I think the third sentence is the most likely to be problematic - specifically the link to WP:BIASED, since those who make such claims are most likely editing from a fringe point of view. I would rather include only "some say its statements should be attributed" (which is indisputable) and leave out the rest, but I'd like to know your reasoning before making that a definite conclusion. Sunrise(talk)23:03, 21 October 2018 (UTC)
Hi Sunrise, in a 2007 finding (whose header was changed in 2009) from the Barrett v. Rosenthal ArbCom case, which was referenced in several of the discussions, Quackwatch was described as a "partisan site":
3.2) Fyslee has repeatedly used Quackwatch and similar partisan sites as references
@Newslinger: Thanks for the reply! Sorry for the delay in getting back to you, as I've been far too busy this week. I suppose that it's fair to directly cite the Arbcom statement. The site does seem to qualify for the very broad definition of "partisan" implied by the proposed 2009 motion, although I'm not sure how relevant that is given that it didn't pass (I also don't agree that having the goal of combat[ing] health-related frauds, myths, fads, fallacies, and misconduct, which sounds like a description for a fact-checking website, could inherently make a source biased rather than being a point in favor of neutrality). I suppose this state of affairs could be challenged, but I would want to avoid setting a precedent that could lead to other challenges for related Arbcom rulings of the same era, as some of them save a great amount of time and effort on the part of editors. Perhaps an argument that the original ruling is simply out of date due to changes in the source over time would be a viable approach.
@BullRangifer: With regard to the end of your previous comment, I suppose it's quite late by now, but please feel free to email me if you ever need someone to talk to. Just please make sure to let me know on my talk page, since I don't check my WP email account regularly, and I'll get back to you as soon as I see the message. Sunrise(talk)04:41, 28 October 2018 (UTC)
Not so fast. The COI accusation was never proven. It's a seriously false allegation, and the two fringe editors who appeared here at Wikipedia to attack me received the harshest punishment Wikipedia can hand out. They were both community banned. That one editor (not Pudeo) could then use their false accusations of me having a COI and repeat them is really nasty. It poisons the well against me. Since when do editors side with fringe, community banned, editors who were personally involved in legal cases involving Barrett? They were the ones with the real COI, not me.
The website and article here should not be judged by the fact that I, and lots of other health professionals, like it. Since when is that consideration used to judge sources negatively? The rest of the discussions show a consensus to treat QW as a generally RS, with the usual caveats for attribution when necessary. We value the viewpoints of subject experts. -- BullRangifer (talk) PingMe14:43, 19 October 2018 (UTC)
@BullRangifer: As pointed out by SlimVirgin in the 3rd discussion in 2015, you have previously stated that you have acted as the "assistant listmaster" to Healthfraud discussion list hosted by the Quackwatch[12] and as "Stephen Barrett's assistant listmater"[13], although you have clarified that it's been years since you have last done it. So the conflict of interest question is rather legitimate if we want to be on the safe side. You were asked to step down from two of the RS discussions by SandyGeorgia because of that.
But understand that I'm not challenging the status quo of using the source itself. It can very well be used as a source especially because of WP:PARITY, but it shouldn't be greenlit as a RS for every article (extra care should be with BLPs). I believe that is the result of the past discussions as well. If you check the May 2015 MEDRS discussion, it was very divisive and there were proposals against and in favour of Quackwatch, and all failed to gain consensus.--Pudeo (talk) 15:15, 19 October 2018 (UTC)
Hi BullRangifer, I didn't take Pudeo's COI accusation into account when I revised the Quackwatch entry. In the 11 discussions linked from the entry, editors consider Quackwatch a self-published source (by an expert) with little to no peer review, and I didn't see consensus for classifying the site as "generally reliable" or "generally unreliable".
While you argued that Quackwatch is a reliable source, you also diluted your statements by repeatedly stating that the suitability of Quackwatch should be "judged on a case by case basis" as a "compromise" in discussion #6 (2009-2010) under the username Brangifer, discussion #10 (2015), and discussion #11 (2015). When "generally reliable" sources are discussed at WP:RSN, most editors don't describe them as sources that need to be judged on a case-by-case basis. It's true that every use of any source should be scrutinized regardless of the source's reputation, but editors need to display more confidence in a source for it to be labeled "generally reliable".
Since the most recent discussion was from 2015, anyone is welcome to start a new discussion on Quackwatch at WP:RSN (the preferred venue) to gauge current consensus. I recommend making this discussion a RfC to maximize your chance of establishing a decisive result. — Newslingertalk15:56, 19 October 2018 (UTC)
I have no great desire to see the issue relitigated. My participation has usually been a response to unfair attacks by the supporters of quacks and unreliable sources who came here to diss QW, advocate nonsense, and bring their real world legal attacks on Barrett here. Jimbo Wales calls them "lunatic charlatans". Attacks on QW are not taken lightly here!
When defending QW, I have been only one of a large number of science based editors who did the same. I'm only notable to you, and no one else, because I was maliciously slandered here. I'm nobody special, never wrote a thing for QW, and only got criticism from Barrett the few times I emailed him with questions.
As far as the COI allegations, they have zero bearing on how we should judge the reliability of the site.
Regarding "case by case basis", that's the same cautionary approach we apply to EVERY source. Period. Every website has all kinds of content, and even though QW is generally accurate and reliable because of editorial oversight, fact checking, subject expertise, and its blatantly obvious main mission to fight inaccurate information (even NYTimes and WaPo don't have that as a primary mission), there is content on QW that obviously should not be used blindly. We're talking about common sense sourcing requirements. My saying they apply to QW is not lack of confidence in the site. Far from it. -- BullRangifer (talk) PingMe19:14, 19 October 2018 (UTC)
While it's clear that you were in favor of Quackwatch's reliability, other editors who mentioned "case by case basis" were less supportive. As I mentioned before, I think a RfC would offer everyone a good chance to finally resolve the status of this site. Among the 11 previous discussions that were scattered across WP:RSN and WT:MEDRS, none of them were RfCs. A new RfC would take priority over all of the previous discussions and establish the current consensus. Any editor can start the RfC. If no editors volunteer, then this issue will be relitigated every single time Quackwatch is brought up in a content dispute. — Newslingertalk20:06, 19 October 2018 (UTC)
Just a note that I updated the no consensus classification to "no consensus, unclear, or additional considerations apply." This first occurred to me during the previous discussion, but there seem to be quite a few cases that are not truly no consensus but are rather things like "not discussed in several years" or "check for these additional factors". Please feel free to adjust the definition further as I'm sure it can be improved. :-) Sunrise(talk)05:18, 28 October 2018 (UTC)
Looks great. All of the entries in the legend could benefit from a short explanation, as the icons in the "Status" column are now anchor links to the legend. — Newslingertalk10:56, 29 October 2018 (UTC)
I wonder if it wouldn't be helpful to add something like: Stale, something something explanation, for entries where the most recent discussions are say...five years old or more and the reliability issues aren't tied to some core aspect of the source. So for example, Wikinews is unlikely to change, because the problem is a fundamental part of the source. But something like WorldNetDaily may somehow get their act together and be more reliable at some point. GMGtalk12:19, 29 October 2018 (UTC)
This sounds reasonable. Would this replace stale entries' current icons, or would both icons be used? Also, what should the background color of stale entries be? — Newslingertalk14:40, 29 October 2018 (UTC)
Well it seems there'd be two obvious ways to do it:
Looks very nice. Between these two, I personally prefer the icon next to the year. Since most of the discussion summaries span more than one line, I would probably put the icon and the year on separate lines, but this is a minor detail. Example:
Discussions regarding Newsmax are dated, with the most recent occurring in 2013. Circumstances may have changed. Discussions are also lacking in depth, and in focus on evaluating this source specifically. Newsmax has been cited in discussions of other sources as a low benchmark for a partisan outlet with regard to US politics, and for a propensity for comparatively fringe viewpoints.
Works for me. At the very least it draws visual attention to the year as an indication that this information has an elevated importance in these instances, while it can for the most part be ignored in the majority of entries. GMGtalk15:20, 29 October 2018 (UTC)
Ummm....I'm not so sure. I mean, I could imagine that there would be sources where the result was so overwhelmingly clear, that no one ever felt the need to bring it up seriously again. For example, now that we've had a site wide RfC on the Daily Mail, I would imagine it will be quite some time until we have another broad community review of the source, simply because the existing consensus is so strong.
Alternatively, we would continue to provide the latest consensus, but if a reader disagrees with it, this can be a big red flag that it would be perfectly appropriate for them to open a new discussion if they have reason to believe the situation has changed, and not hesitate because they might get hit with a big load of "not this again" from the community. GMGtalk15:39, 29 October 2018 (UTC)
Done. Since nobody has objected to the stale icon, it has been implemented and added to the legend. Whether the sources should retain their original color or be colored yellow is something we can continue to discuss. — Newslingertalk14:38, 16 November 2018 (UTC)
This reassessment was made after assigning a higher weight to the most recent discussion (#8, Wikipedia:Reliable sources/Noticeboard/Archive 250#Electronic Intifada (Again)), which took place in October 2018. This discussion showed consensus that EI is a generally unreliable source with a poor reputation for accuracy, fact-checking, and error-correction.
The discussion before this one (#7) was from July 2014. As this page favors recent consensus ("Consensus can change, and if more recent discussions considering new evidence or argument reaches a different consensus, this list should be updated to reflect those changes."), discussion #8 should be weighted very highly as it is the only recent discussion. None of discussions #1-7 (from 2008 to 2014) show consensus that EI is a generally reliable source; they showed no consensus at best and a "generally unreliable" consensus at worst.
The original entry for EI was added by me on October 17. As the list of discussions hasn't changed, this is just a reconsideration.
If anyone disagrees with the reassessment, please feel free to edit the list again and state your case here. Please note that although discussion #8 is very recent and quite thorough, there has never been a RfC on EI's reliability, and a new RfC would take priority over all of the past discussions. — Newslingertalk00:03, 21 November 2018 (UTC)
After reviewing the discussions for the entry on HuffPost again, I've changed the classification from "generally reliable" to "no consensus, unclear, or additional considerations apply" because the discussions don't show a strong enough consensus in favor of HuffPost's reliability. Here is a list of the 15 discussions on HuffPost:
As is the standard, these discussions were compiled from the "Search this noticeboard & archives" box on WP:RSN. All WP:RSN discussions with the term "HuffPost" or "Huffington Post" in the section title, excluding discussions that focus solely on articles written by HuffPost contributors, are listed above.
The entry for HuffPost was originally added on July 29, referencing 4 discussions with a "generally reliable" classification. On October 18, I expanded the HuffPost entry and split it into staff and contributor entries, retaining the "generally reliable" classification for the staff entry, and classifying articles written by HuffPost contributors as "generally unreliable".
If anyone disagrees with the reassessment, please feel free to edit the list again and state your case here. Alternatively, since there has never been a RfC on WP:RSN for HuffPost, a RfC would clarify the current consensus on HuffPost's reliability. — Newslingertalk19:53, 20 November 2018 (UTC)
I don't necessarily dispute changing it to 'no consensus', but not all of those linked discussions are particularly useful. Some are from when the website was only a few years old. Also, some detractors call HuffPost fake news, which is absurd. Comments like "the Huffington Post will happily publish unverified, borderline fake, stories to please their leftist followers." and "Huffington Post puts out fake news, but lefties don't notice because it is part of the Team." The website has won some pretty prestigious awards, although it's hard to deny that their headlines are sometimes sensationalistic and that they are left-biased. They are not a publisher of fake news though. - MrX 🖋 20:56, 20 November 2018 (UTC)
I try to include all discussions on a source using the same search methodology to avoid accusations of cherrypicking the discussions. Maybe it would be better to bold significant discussions in the chart to make it easier for other editors to review? I agree that some of the comments are hyperbolic, but I generally don't discount the opinions of editors unless they cite obvious factual mistakes. It's also true that HuffPost has won a number of awards, but only the 2012 Pulitzer was mentioned in the discussions, and I didn't consider the others because they weren't mentioned. Any editor is welcome to start a RfC on WP:RSN at any time, though there are currently 2 active ones there and I'm not sure if 3 would be too much. — Newslingertalk00:25, 21 November 2018 (UTC)
That's fine, since we haven't really settled on any criteria for what type of discussions count toward establishing consensus. As far as comments about reliability, I think we can safely discard any comment that refers to Huff Post as "fake news", per WP:CIR.- MrX 🖋 01:51, 21 November 2018 (UTC)
Sources by usage in articles
Not sure if you guys have seen this before, but SashiRolls has compiled a list of commonly used sources based on the number of links to each source. It's not perfect, but it may be useful when deciding which sources to add to the list first. ListCategorizedfeminist (talk) 11:34, 22 November 2018 (UTC)
If a source is commonly used, I think it would not even need to be on this list. In fact, some like the Los Angeles Times should be removed for that reason. - MrX 🖋 17:19, 23 November 2018 (UTC)
Which isn't surprising for one of the UK's best-selling newspapers. I think it would be best to prioritize replacement of Daily Mail citations supporting more controversial statements first. feminist (talk) 03:46, 24 November 2018 (UTC)
This may be true in some cases, but there are reasons an editor may suspect the reliability of a source. I recently added The A.V. Club and Deadline to the list. The A.V. Club is owned by the Onion, and Deadline started as a one-woman blog, so it's reasonable for an editor suspect their reliability. feminist (talk) 03:55, 24 November 2018 (UTC)
Well, it's arguable, but at the very least The A.V. Club has the same parent company as the Onion, and has historically been known as a sister publication to the Onion. Doesn't change the facts that 1. it's widely used as a source and 2. it's reasonable for an editor to suspect its reliability. feminist (talk) 12:43, 26 November 2018 (UTC)
Based on a review of previous RSN discussions, I see that Deseret News is considered generally reliable as a local newspaper, but opinions are divided on whether it can be used as an RS for topics relating to the Mormon Church. Based on this, I added it to the list as "no consensus". Would "generally reliable" or "no consensus" be a better classification in this case? feminist (talk) 12:34, 26 November 2018 (UTC)
I've also examined the discussions and the AfD discussion for Octaviano Tenorio that was mentioned in the second and third discussions. From I can see, editors generally believe that the Deseret News is reliable. There's no consensus on whether it's independent of the LDS Church, but a source's independence is separate from its reliability. Even if editors decide that the Deseret News were non-independent for a specific topic related to the LDS Church, it should still be usable if its statements were properly attributed. Whether the Deseret News is independent is an important concern for AfD, where editors have to decide whether a subject meets WP:GNG, but I think independence is outside the scope of the "generally reliable", "no consensus", and "generally unreliable" classifications on WP:RSP.
So, my take is that the Deseret News should be classified as "generally reliable", and I've changed the entry to this classification. I've also changed a sentence from "There is no consensus on whether the Deseret News can be used as a source for topics regarding the Mormon Church." to "...there is no consensus on whether the Deseret News is independent of the LDS Church. The publication's statements on topics regarding the LDS Church should be properly attributed.", since even if the newspaper didn't confer notability on a subject, it could still be usable in the article. If you can improve this further, please do. — Newslingertalk01:23, 27 November 2018 (UTC)
I'm surprised by no consensus on whether it's independent of the LDS Church... it's literally owned by them, isn't it (parent of a parent company or somesuch)? — Rhododendritestalk \\ 04:30, 27 November 2018 (UTC)
Reliable for what? Snopes is reliable for a claim that something's been passed around the internet. It is not reliable for a claim about whether a given medication has an acceptable safety profile.
That's one of the fundamental problems with this kind of list. Even garbage sources can (very rarely) be "reliable" for a carefully written claim, while the best, most obviously gold-plated academic source is only reliable for a tiny, tiny fraction of the claims that we would want to include in Wikipedia. "Yes, reliable" and "No, unreliable" doesn't address the main problem with reliability – which is not "Is this publisher always good?", but "Is this specific page, which contains these specific statements, reliable for this specific statement, in this specific Wikipedia article?" WhatamIdoing (talk) 15:52, 3 September 2018 (UTC)
I keep meaning to dig up the info on things like the Journal of the American Medical Association and then never find the time to do it so I'm posting it here in case someone else has the interest to pull the info to add sources like that to the table. Best, Barkeep49 (talk) 18:48, 29 November 2018 (UTC)
Should such high-impact journals be added though? I doubt many editors would question their reliability. Journals considered unreliable can be added. feminist (talk) 08:06, 3 December 2018 (UTC)
Since this page doesn't have defined inclusion guidelines, you could argue this either way. Personally, I see two main ways to use this list. Editors who have a specific source they'd like to evaluate can search within this list to quickly find the entry for the source. Also, editors who don't have a specific source in mind can browse this list to find additional sources for their research. In either case, I don't see any harm in having JAMA and other high-impact journals in the list. If there's consensus that the list is too long, then we can establish inclusion guidelines to shorten it. — Newslingertalk08:15, 4 December 2018 (UTC)
I'm concerned about this list becoming behemothic. Does a highly-respected journal like JAMA actually need to be on this list? What about the other hundreds of highly-respected journals?- MrX 🖋 01:40, 5 December 2018 (UTC)
I suggested JAMA because I was aware despite it being high quality that there had been 2018 conversations around it at RSN. I will say that I used this list today in conversation with a longtime editor about whether a given source was likely to be RS. A too long of an article is some cause of eventual concern. Best, Barkeep49 (talk) 02:08, 5 December 2018 (UTC)
@Barkeep49: But are editors seriously questioning the general reliability of the publication? That's the purpose of this list.- MrX 🖋 13:04, 5 December 2018 (UTC)
@MrX: Fair point certainly. At the risk of repeating myself, I suggested it given a 2018 conversation and think it has value here. I don't feel strongly and if there was agreement to remove it I would support that decision. Best, Barkeep49 (talk) 03:19, 6 December 2018 (UTC)
@Barkeep49: I think it should be removed and we should solidify some inclusion criteria (see below). Of the two JAMA discussions, neither pertain to the overall reliability of the publication. One is a 10 year old discussion about the a statement from the AMA Council on Scientific Affairs dealing with Alternative medicine and the other is about a specific opinion article in JAMA. Opinion articles will always be outside of the scope of the question of whether a source is reliable. In cases of long-running, widely-respected medical journals, WP:MEDRS is instructive also.- MrX 🖋 13:19, 6 December 2018 (UTC)
Perhaps split this list into different sections based on the type of source (e.g. newspaper/website/magazine/journal/organization) or subject area (news/entertainment/tech/politics/etc.)? feminist (talk) 05:03, 5 December 2018 (UTC)
I don't think that would help. I think we should start defining some inclusion criteria so that this list remain manageable. I'll start: It should not include sources that are widely regarded as reliable. - MrX 🖋 13:04, 5 December 2018 (UTC)
@Newslinger: I will need to think about this some more, but to start with, we can look at links. If there are at least n links to a source (where, let's say, n=10,000) then the source is presumed to be widely regarded as reliable.for example, the New York Times or the BBC the BBC contrasted with Town Hall. Similarly, we could include publications that have received major awards for journalism. We could set a threshold for the number of (serious, quality) RSN discussions about a source's general reliability. I don't think it would be that hard to come up with inclusion guidelines, allowing for rare exceptions.- MrX 🖋 13:09, 6 December 2018 (UTC)
This seems a bad idea
This seems a fundamentally flawed item as open essay.
In execution, notes of 'what MrX thinks' might be helpful for him but is not policy, consensus, nor fully complete. Ultimately a source is only judged RS in some context and is subject to RSN. This list is moot for that.
This essay is also confusing RS, some source for POV info, as if 'reliable' means 'true info'. This list is invalid in generalizing too far. While it might do well to say seek WP:BESTSOURCES, it does not do this -- this is simply looking like an invalid balcklist of some and whitelist of others.
@Markbassett: I would be interested to hear about what you specifically disagree with. The list is neither a blacklist or a whitelist. It's merely an index to discussions and a summary of all of the discussions for any particular source. Can you point out where in this guide is written that "reliable" means "true info"? - MrX 🖋 18:23, 25 August 2018 (UTC)
* Wrong headed : A source is only judged RS in some context of a specific item for a specific article edit and is subject to RSN. This list is fundamentally wrong in labeling a originating entity entirely 'reliable' or entirely 'unreliable', and in using the term as if it means 'right'. The inherent issues of using a 'no' symbol and Redlining smells a bit too much of a censorship board and wiping out POVs on one side, while a green and 'OK' universal blessing is equally as bad for opening things to bad info or POV on the other. If nothing else, any source has limits of scope - its areas of knowledge and varied nature of items -- and a few bloopers in any mix.
* Conflicts with policy the policy approach has no opening for a blank generalization, and this article does not incorporate any of the items of RS such as categories for news or biased sources. This does not provide for existing policy, does not contain past consensus, nor is it fully complete in the rows shown for discussions involving past discussions on the mentioned items. If this has no functional utility to RS criteria -- then
* Immature and rush to fire, ready, oops There has been no observable discussion about what such an item should have or why not incorporate into the pre-existing items such as WP:PUS. That two days after started it's being promoted for this and that and inserted into WP policy as if discussion is complete ... does not speak well of it as a reputable analysis. That there is no stated criteria or transparency as to evolving the entries speaks ill of it as a WP normal entry, let alone as a guidance. Every single entry seems an editor trying to make their POV into WP policy without exploration or transparency. Even article pages have a visible TALK history over content and editorial guidelines, but this seems just trying to put POVs in instead of Policy. Why many multiple discussed entities are not (or are) present is not clear; why only some discussions for them are shown, why e.g. MrX said the comment he did or what comments are for ... too much is simply missing.
User:NewslingerThis page conflicts with policy and otherwise ignores it, so no it's not at all explanatory supplement even of how it was arrived at except by apparently expressing ad hoc personal opinion not processed thru a consensus discussion. Fine for his personal notes, not usable for anything RSN. Cheers Markbassett (talk) 04:29, 30 August 2018 (UTC)
Which policy does this page conflict with? I don't see any conflicts. Also, I didn't say that this page is an explanatory supplement. I said that this page may become an explanatory supplement pending the outcome of this RfC, which both of us have already participated in. If you object to any of the assessments, you can dispute them on this talk page. — Newslingertalk06:01, 30 August 2018 (UTC)
User:NewslingerWrong headed in labeling a originating entity entirely 'reliable' or entirely 'unreliable'. The entire source entity - a mix of articles and opinion pieces, blogs and self-published stuff, old and new -- is all put under some color and stop or go sign and the wikilink to RELIABLE or to UNRELIABLE. No big differentiation among bits and usually no indication that there even are varying bits within a source entity, nothing showing the type of source that guides a RS discussion, nor qualities of checking the source has at it -- this is just giving one single label of the source as entirely one way or another. That patently is contrary to prior discussions -- by definition and content one can see that all of these have "uses" and all of these have been repeatedly challenged for issues. So no, the rating of a source as entirely one or the other is both wrong headed and by individual cases demonstrably false to fact. Cheers Markbassett (talk) 04:41, 30 August 2018 (UTC)
This page uses the phrase "Generally reliable" instead of "entirely 'reliable'" and the phrase "Generally unreliable" instead of "entirely 'unreliable'" to allow for editors to make reliability judgments independent of the general classification. The wikilinks go to Wikipedia:Identifying reliable sources and Wikipedia:Identifying reliable sources#Questionable sources, which are content from the guideline that helps editors make these reliability judgments. Anyone can challenge the general classification for any situation. Additionally, if past WP:RSN discussions mention certain parts of the source that should be excluded from the general classification (e.g. opinion pieces, contributor articles), they should be noted in the "Summary" column of the table. — Newslingertalk06:01, 30 August 2018 (UTC)
This page says "generally" meaning entirely -- for the entirety of a site it is either "WP:RELIABLE" or "WP:UNRELIABLE" - without notice that sites are composed of mixed value parts, and with little regard for the scope or context that applies. By RS guidance, "Proper sourcing depends on context", and WP:CONTEXTMATTERS - "Each source must be carefully weighed to judge whether it is reliable for the statement being made in the Wikipedia article and is an appropriate source for that content." Just being delivered by XYZ does not mean the source is XYZ -- and when the piece is from an XYZ staff writer the writer and XYZ qualifications about the statement should be judged. There is no such thing as 'generally (entire site) reliable so the entire table of that nature is fairly bad idea. Markbassett (talk) 01:10, 5 December 2018 (UTC)
In the context of this page, the word "generally" does not mean "entirely". It means "usually". The remainder of WP:GENERALLYRELIABLE and WP:GENERALLYUNRELIABLE, as well as the summary in the entry of each source, explain the types of situations where a source may be considered more or less reliable than usual. — Newslingertalk09:07, 6 December 2018 (UTC)
User:Newslinger - But we are not talking about that -- we are where my objection saying "entirely 'reliable' or entirely 'unreliable'" in my expression of a problem got diverted into some conflusion back that the article was not literally saying "entirely". No -- but my complaint was it acts that way, not whether it literally said the word. That there may be a caveat 5 screens lower that points to the actual guideline does not change that the message conveyed in this table format is 'trust everything at this website/publisher to be correct'. The WP:RS however, has it that such would be determined specific to a case of context for a specific article content and whether the specific cited author and article/book and publisher are suitable. Listing the publisher or website as a single entirety and treating it as an entirely reliable or unreliable is what's in complaint here, not whether this table literally does or doesn't say 'entire'. A decent publisher has better articles and worse articles and op-eds and blogs and articles which were bad remain available and a separate retraction notice is given. It is which specific item and specific part of the publisher and which specific author that is judged RS, for that one specific context usage. It is not the canonization for the entirety of a producer works. Even within a single decent newspaper article, one needs to distinguish where the author is stating factual information (however skewed) from where the author is making any judgemental statements or any quotes from third parties. There is no "entirely 'reliable'", and there never can be the way WP:RS is stated. WP:RS has no "Generally reliable". As a sidenote, I'd maybe point people to the higher-level policy WP:SOURCE that is above the guideline. Cheers Markbassett (talk) 17:58, 6 December 2018 (UTC)
About your recommendation to use WP:PUS: That page is also an essay, and none of the entries on that page are supported by community discussion. In contrast, entries on WP:RS/P are supported by past discussions and RfCs on WP:RSN. Unlike WP:PUS, nobody can directly add an entry to WP:RS/P. They must first go through WP:RSN, and the consensus established from the discussion would then be noted in WP:RS/P. If you are concerned about editors "trying to make their POV into WP policy without exploration or transparency", WP:RS/P would be a strict improvement over WP:PUS. (And even then, WP:RS/P is also an essay and will not become anything more than an explanatory supplement at best.) — Newslingertalk15:14, 29 August 2018 (UTC)
User:NewslingerImmature and rush to fire, ready, oops says this failed to do any prior discussion or discuss alternative with prior guides such as PUS -- if you want me to list more of the considerations it might have thought about or of the other pre-existing places it could have or should have made discussions and considered, I can propose some, but that really should have been a question this article asked instead of ignored. Cheers Markbassett (talk) 04:45, 30 August 2018 (UTC)
User:Newslinger This page is also arbitrary in listing and content but is seeking to claim it's a supplement -seems worse. Another problem is we already have this WP:PUS plus WP:ELPEREN plus another article or three about sites... and this one does not match the judgement of prior articles. Just seems inevitable that will lead to problems. Markbassett (talk) 03:54, 31 August 2018 (UTC)
On a final note, if there are missing WP:RSN discussions for an entry, please add them yourself and revise the summary to take the addition into account. (The WP:RSN search function is not very good, and it's possible to miss discussions from a search.) If there is a disagreement, it can be settled on this talk page. — Newslingertalk15:19, 29 August 2018 (UTC)
User:Newslinger Final note back is also the first note -- Immature and rush to fire, ready, oops says this failed to do any prior discussion, including those that would prepare it to even get edits on what kinds of content would go in here. Yes, some of the obvious oops include missing RSN discussions -- so far include listing just one RSN discussion (so not really very 'Perennial'), few RSN items shown out of actual dozens (so not complete), the comments not being associated to RSN pings shown, and no criteria/explanation given (so no clear relationship). So no, it's not so simple as just adding 10 times more RSN items to each and completely rewriting the table -- flipping red to green and green to yellow and yellow to blue -- even when adding in the inevitable edit warring ... it's also still just Wrong headed in concept and immature to start work on a table. Until and unless the discussions and documenting is done first over what this is for, how it would be properly used, what the contents should be, and what the contents should be. Otherwise one has here a good example of poorly-defined data of no demonstrable meaning or worth, and no way to identify the not-recorded thoughts for any of the already done (in varying ways) stuff currently here. Usually the fastest and surest approach for data quality is throw it all out and start fresh, but that's also usually the one thing folks are unwilling to admit -- that they're just wasting time adding to a muddle. Cheers Markbassett (talk) 05:09, 30 August 2018 (UTC)
If there are missing discussions or missing sources, you can add them here and edit the summary to reflect the additions. If those sources aren't discussed enough, you can start a discussion or RfC on WP:RSN, and then list them here. If there is edit warring on this page, editors can submit a request for page protection and require a discussion on this talk page before each controversial change to the page. The problem before this page was created is that the poor search feature in WP:RSN makes it difficult to track down all of the past discussions for a source. This page indexes them to save editors time and effort that can be used in article writing, while using language ("Generally") that preserves the ability of editors to contest any of the summaries. — Newslingertalk06:01, 30 August 2018 (UTC)
I suppose that congratulations are in order for Markbassett, for predicting that this page would cause problems (at least in its current form).
See also WP:ELN this week, where a new-ish editor is claiming that links to Amazon.com and iTunes are banned by this page. He has been removing such sources when he finds them cited at the end of sentences such as "The album was released on <date> by <label>" – statements for which these sources are acceptable.
I think that this page needs re-designed, to keep in mind the extraordinary breadth of legitimate uses for some of these platforms. The product details on Amazon.com should not be treated the same as a conspiracy site. WhatamIdoing (talk) 16:42, 30 August 2018 (UTC)
I'm not sure that misuse of this list by an editor who doesn't get the difference between external links and references is really evidence that the problem there is really with this page, especially when that user is explicitly ignoring the advice given on this very page. GMGtalk16:57, 30 August 2018 (UTC)
Bad idea? It's a TERRIBLE idea. It reeks of "officially approved thought", and to say that a blatantly partisan and opinion driven media outlets such as the Huffington Post is rated green for most reliable only reinforces the notion that "approved sources" is basically a cabal of editors imposing their politics on Wikipedia readers. The SPLC just went through a major scandal involving Maajid Nawaz... basically, they slandered him and only apologized when forced to... and yet the SPLC is rated green. "Truthful and reliable".
Bottom line: this whole list is blatant politicization, the kind of thing that can't possibly be agreed upon by a large pool of people. If Wikipedia is going to take an official position on "reliable sources" in such a manner that makes it look politically partisan... and Wikipedia IS DOING THAT now... then it's going to rightfully be labeled a politically partisan source itself.
I've been an enthusiastic Wikipedia user for almost two decades now, and have authored and edited as well. If Wikipedia continues to tack in this direction, then its editors are going to destroy any notion of Wikipedia being a non-partisan, neutral source of information. To put it bluntly, this list and the mindset behind is somewhat Orwellian and goes against the very the spirit in which Wikipedia was first created. It's well and good to tell readers "let the buyer beware" with sources... ANY AND ALL sources... rather than take an official view of "this source good and truthful, that source bad and lying". I'm old enough to remember a long list of "reliable" news outlets blatantly lying for one reason or another. From NBC News putting explosives into fuel tanks to get "test" results they want, to Rolling Stone perpetrating the most malicious rape hoax of modern memory, lots of these "green" sources have blatantly lied, and only admitted it when caught.
Put whatever links authors want into pages, without an official Wikipedia seal of approval, and tell readers Caveat Emptor in ALL sources. It's the only way to honestly do things and respect the readers.
Hi DesScorp, if you don't think the Southern Poverty Law Center is a generally reliable source, you're welcome to start a new discussion or RfC at WP:RSN and discuss this with other interested editors. After the discussion finishes, the current consensus will be documented in the table. Alternatively, if you think community consensus on the SPLC is categorized or summarized incorrectly, you can make changes to the table yourself or discuss the entry on this talk page (preferably in a new section). — Newslingertalk07:45, 6 October 2018 (UTC)
I'm just noting for the record that MShabazz recently changed the entry for Fox News from "generally reliable" to "no consensus, unclear, or additional considerations apply".
Originally, I added Fox News as "generally reliable" on September 2. This was because the 2010 RfC stated: "Consensus is that while Fox may not always be reliable it is a Reliable Source", which meets the definition of "generally reliable".
However, aside from the 2010 RfC, the other discussions were not very conclusive. I still retained the "generally reliable" classification because RfCs take priority over other discussions, and both of the 2018 RfCs (here and here) ended prematurely.
Altogether, the discussions and RfCs show that Fox News is somewhere between "generally reliable" and "no consensus". If there's consensus to keep it in the latter category, then that's exactly what we should do. — Newslingertalk00:33, 26 November 2018 (UTC)
I think it's a stretch to categorize it as no consensus. MShabazz probably needs to make a convincing argument that is based on past discussions.- MrX 🖋 01:08, 26 November 2018 (UTC)
When a source is described as "not always reliable" and "a partisan news organization", and "[e]ditors are advised to exercise caution when using [it] as a source", by definition there is no consensus that it is a reliable source. I think it's a stretch to consider that a consensus for viewing it as a reliable source for facts. — MShabazzTalk/Stalk01:16, 26 November 2018 (UTC)
It is also described as "it is a Reliable Source". In practice, we routinely cite their news coverage without much objection. I think if you want to change this entry to "no consensus", you're going to have to start an RfC to determine if that is actually the case.- MrX 🖋 12:21, 26 November 2018 (UTC)
For anyone reading this, the best way to settle an edit war is to post a new RfC on WP:RSN, which would supersede the one from 2010. If anyone is willing to do this, I would recommend that you phrase it along the lines of "Should Fox News be considered...?" and then offer the three options from the legend. — Newslingertalk03:20, 26 November 2018 (UTC)
I added a link to this discussion on the entry for Fox News, while keeping the classification as "generally reliable" for now. It is clear that at least some editors are unhappy with the change and would prefer some discussion first. I hope my edit is a compromise for this dispute. feminist (talk) 12:19, 26 November 2018 (UTC)
In the discussions I've seen, it's generally regarded as a reliable source for its main news content, but not for much of its other programming, which can indeed become unreliable, especially in its blurring of the line between fact and opinion. It's hard to classify "some parts but not others" with the current scheme, as "no consensus" doesn't quite cover it (regardless of all that's lumped under that heading in the footnotes). Starting a new RfC on whether it's reliable will almost certainly have the same result. It's how to represent that result that's in question. Perhaps there should be an option for "mixed" or "complicated" or somesuch, with an explanation. Simply saying "generally reliable" for all of Fox News makes it seem as though the channel's content is generally reliable, regardless of qualifiers, and I think there would be pretty strong consensus that, say, Hannity, is not "generally reliable" for statements of fact. This "mixed" could also apply to e.g. Forbes vs. Forbes contributors, though for somewhat different reasons. — Rhododendritestalk \\ 04:18, 27 November 2018 (UTC)
I've added two sentences to the beginning of the Fox News summary: "Editors show consensus that news reports from Fox News are generally reliable. Content from the channel's talk shows, including Hannity and Tucker Carlson Tonight, are equivalent to opinion pieces and should be handled with the appropriate guideline." This should be clearer than simply pointing to WP:NEWSORG.
I like the idea of something between reliable and mixed for things like Fox News and Desert News below. Limited circumstances isn't quite right (since all RS are not RS in some circumstances) but something that doesn't give the proverbial green light to its use. Best, Barkeep49 (talk) 06:50, 27 November 2018 (UTC)
Agreed. I think a good solution would be to split this into two entries: something like "Fox News" (generally reliable) and "Fox News content" (additional considerations apply). feminist (talk) 10:43, 27 November 2018 (UTC)
Since a lot of editors are confused about what deprecation means, I started a new essay at Wikipedia:Deprecated sources to help explain. The essay is is based on an interpretation of the 2017 Daily Mail RfC. It might be useful to add guidance on editor conduct (when the inclusion of a deprecated source is disputed) to help prevent the problems Andy Dingley described in the current RfC for The Sun. Please feel free to make any improvements as you see fit.
Breitbart and Occupy Democrats seem to have contradictory descriptions. Each states "It can still be used as a source when attributing opinion/viewpoint/commentary", yet they are "deprecated in the same way as the Daily Mail" (generally prohibiting use as a source, with no caveat allowing attributed opinons). Breitbart is also blacklisted, effectively blocking its use altogether. –dlthewave☎04:27, 7 December 2018 (UTC)
Hopefully, the newly written WP:DEPRECATED helps clarify this. The 2017 Daily Mail RfC made a couple of exceptions, and WP:ABOUTSELF applies on top of that. Both Breitbart News and InfoWars were blacklisted due to spam. Generally, sources get proposed for deprecation when they have serious reliability issues, and for blacklisting when they have serious spam issues. These two restrictions are independent from one another, since they are proposed at different venues (WP:RSN and WP:SBL). — Newslingertalk05:02, 7 December 2018 (UTC)
Ok, the wording came from the respective RfCs. I've seen editors use "opinions and viewpoints" as an excuse to include any and all opinions published by the source, including those of authors, interviewees and guest columnists regarding topics unrelated to the source. Hopefully WP:DEPRECATED will help establish some consistency and limit these opinions to WP:ABOUTSELF. –dlthewave☎00:52, 8 December 2018 (UTC)
"Uses" links
Currently the "uses" links use the {{linksearch}} template. The problem is that this template turns up links from all namespaces, when links in articles are what matters. I propose replacing the template with a search link formatted [[Special:Search/insource:"website.com"]]. Thoughts? feminist (talk) 16:37, 23 November 2018 (UTC)
I'm not in favor of that. The current method shows a numbered list that is easy to digest. What you're proposing, shows snippets that I find unhelpful. Also, it finds domain names (and not exclusively URLs), for example in the text of an article or in the publisher field of the cite template. - MrX 🖋 17:16, 23 November 2018 (UTC)
I hesitated before switching from Special:Search to {{linksearch}} for this reason. Special:Search's article space filter is very useful, but {{linksearch}} is more precise. Either way is fine with me. If we switch back to Special:Search, we should probably use a search link for each domain instead of multi-domain searches, since regex searches are too slow and time out. (On a side note, I noticed that I forgot to escape the periods in the old multi-domain searches, but even this query times out as well.) — Newslingertalk00:27, 24 November 2018 (UTC)
There is no consensus about the reliability of Media Research Center publications, including Newsbusters. As a biased or opinionated source, their statements should be properly attributed.
This lets the reader quickly see how many domains there are, and gives each link a unique identifier. (Note that the width of the "Uses" column was set with style="width: 2.7em;" in this example.) — Newslingertalk03:33, 2 December 2018 (UTC)
I've replaced {{sl}} and {{linksearch}} with the new {{Domain uses}} ({{duses}}) template to reduce the page size. Since each domain only has to be typed one time, this also makes the list easier to edit and decreases the likelihood of errors being introduced. — Newslingertalk11:29, 8 December 2018 (UTC)
That's great and I'm delighted that the page is being improved. I wonder if the padlock symbol should be replaced by a more intuitive symbol, like a link in chain.- MrX 🖋 13:07, 8 December 2018 (UTC)
Of course. Feel free to make changes to Template:Domain uses, since it's only used on this page and WP:DEPRECATED at the moment. I chose the lock icons to indicate the separate Special:Linksearch searches for HTTPS and HTTP links. If this is unintuitive, then the chain link icon would make more sense. Alternatively, we could also use the lock for HTTPS links and the chain for HTTP links. If you tweak this, please be sure to change the width of the "Uses" column to ensure that the table is formatted properly. — Newslingertalk14:22, 8 December 2018 (UTC)
I've changed the open lock icon (for HTTP links) to the chain link icon. Is this okay, or would it be better to have two chain link icons? — Newslingertalk21:37, 8 December 2018 (UTC)
I think we need to add Science-based Medicine (sciencebasedmedicine.org) which has surfaced in quite a few discussions.[14][15][16][17][18][19][20] I'm parking this here until I can get around to adding it, but if someone beats me to it, that's good too. - MrX 🖋 21:50, 9 December 2018 (UTC)
That seems to be a disputed view. Quite a few editors consider SBM to be not be self-published because it's under editorial control and has a staff of writers.- MrX 🖋 12:23, 11 December 2018 (UTC)
I agree with your assessment, there doesn't seem to be consensus that SMB is not self published, but their is no consensus to say it is self published either. There is consensus that SBM is RS, but is is not clear if it can be used for statements of fact about living persons(SPS cannot be used for most BLP content). I have been involved in several discussions related to SBM and 2 problems always seem to come up in those discussions.
1: A surprising number of editors seem to harbor the misconception that SPS and RS are mutually exclusive (or they misinterpret suggestions that SBM is an SPS as saying it is not reliable). In truth, expert SPS are reliable sources in their subject area, BUT still may not be used as sources for statments of fact about living persons (see WP:BLPSPS)
2: Some editors assume that anyone who says SBM is an SPS must be a FRINGE-pusher, based on the logicfallacy that because FRINGE proponents don't like SBM, anyone who wants its use as a source limited in any way must be a FRINGE supporter, this bogus argument(and all variations on it) leads to heated discussions that lack in substance, and accomplish nothing (other than a slightly increased risk of an unnecessary ANI case).
You have some good points. I would suggest starting a well-worded RfC at WP:RS/N so that a broader set of editors can opine on whether SBM is a reliable source, a self-published source, and if it can be used in WP:BLPs.- MrX 🖋 12:54, 12 December 2018 (UTC)
As mentioned in a discussion above, links in the "Uses" column are now represented with the {{duses}} template, which is also fully documented.
These changes reduced the size of the page by just over 31,000 bytes. If anyone has more ideas for making this page easier to edit, we would all love to hear them. Thanks. — Newslingertalk14:32, 8 December 2018 (UTC)
The "Last" discussion and "Status" columns now use the {{/Last}} and {{/Status}} templates, respectively. Sources with last discussions that are at least 5 years old are now marked as stale automatically, and this can also be overridden. The size of the page decreased by another 14kb, down to 130kb. — Newslingertalk04:26, 9 December 2018 (UTC)
Moxy, good idea! I, myself, would prefer for the list to be reorganised into sections based on country, genre, language or something to that effect. Regards, SshibumXZ (talk·contribs). 23:41, 9 December 2018 (UTC)
Done. There is now a table of contents. SshibumXZ, I'm not sure if there's a good way to reorganize the table without breaking its sorting feature. If there's an important aspect we want to classify the sources by, we could add an additional (very narrow) column and let readers sort by that column. For example, we could add a "Country" column and fill the cells with the flag of each source's country, which would then be sortable. I'm still concerned this would add too much to the table, but I can't think of any other solution besides creating a new page. — Newslingertalk00:36, 10 December 2018 (UTC)
I would object to adding any more columns unless we have a very well-articulated reason. If someone is consulting this list, it should be because they are trying to evaluate a specific source, not the sources of an entire country or language. Such a list might serve a purpose, but that's not the purpose of this list.- MrX 🖋 12:05, 10 December 2018 (UTC)
If we are to add columns, a column that lists a source's subject area would be more useful. I can imagine editors coming to this list to see what sources are considered reliable for a certain topic (e.g. entertainment, tech). feminist (talk) 04:11, 13 December 2018 (UTC)
No problem. {{/Uses}}, which wraps around the {{duses}} template, has also been implemented to display the "Uses" column. This is probably as far as I go with templatizing the table. I considered turning the rows into templates, but didn't see any real benefit. — Newslingertalk04:35, 10 December 2018 (UTC)
I just added The Hill to the list as a generally reliable source. Just noting here for the record since I participated at the latest discussion which means I may be biased. My review of the current discussion and past mentions shows that most editors consider The Hill to be reliable. feminist (talk) 10:16, 6 January 2019 (UTC)
Galobtter recently changed the ordering of the plus sign and the discussion count in the Discussions List column of the table. It is now after the count (e.g. 32+) rather than before it (e.g. +32). I don't have a preference on whether the plus sign is before or after the count.
When I added the discussion count last October, the idea was to indicate that there were "32 more" discussions of the Daily Mail. 32 more didn't fit, so I settled for +32. It doesn't really matter whether the plus sign is before or after the number, as long as it's consistent.
Noting for the record: Currently, for sources without RfCs, we wrap the list of discussions in an {{efn}} explanatory footnote tag if there are at least 14 discussions in the list. This is because most desktop web browsers can fit numbers "1" to "13" into three rows, but number "14" would create a fourth row and make the table cell too crowded. For sources with at least one RfC, we use {{efn}} if there are at least 6 discussions (2 rows) in the list. — Newslingertalk10:40, 12 January 2019 (UTC)
32 more is indeed best indicated by "+32", however I think plus signs aren't needed when no discussions are linked directly (when the only links are in the footnotes) - there is nothing to be "more" to, and I've done [21] - I think that makes the most sense anyhow. Galobtter (pingó mió)
I was surprised that there seems to have been relatively little discussion of the reliability (or lack thereof) of Vox. But the discussions that have happened seem to suggest that it is considered reliable (e.g. [22][23]). I am curious as to what other editors think regarding how reliable Vox should be considered, though I am inclined to think it is generally reliable. IntoThinAir (talk) 20:39, 17 January 2019 (UTC)
The only reason I really brought it up here was that I expected it to be included on this page (i.e. the perennial sources list), and was surprised by its absence. IDK if this is a reliable source, but it also says that Vox is pretty factually accurate but also has a moderate/strong left-wing bias. IntoThinAir (talk) 21:32, 17 January 2019 (UTC)
Vox is a popular website, with an Alexa rank of 1,069 and over 1,900 uses in articles. If you want to discuss how the source is used or have concerns about the source (e.g. potential bias), you can start a discussion at the reliable sources noticeboard, which will cause Vox to be indexed in the list. Otherwise, Vox appears to be uncontroversial despite being used frequently, and that's a positive indicator of the source's reliability. — Newslingertalk22:17, 17 January 2019 (UTC)
Added as generally reliable, but partisan/biased. My reading of the discussions is that, taken together, they show a rough consensus for treating Vox as generally reliable. I no longer read Vox (and dislike it) because of their treatment of Sam Harris, so I may personally be biased. feminist (talk) 16:48, 18 January 2019 (UTC)
I've added a few lines of CSS to {{Reliable sources/styles.css}} to allow the small tags to be replaced. However, this also involved adding line spacing; 1.6 was close to the original (it's not the exact value because it probably differs depending on the skin being used). Would it be useful to change it?
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
Al Jazeera is considered a generally reliable news organization. Editors perceive Al Jazeera English (and Aljazeera.com) to be more reliable than Al Jazeera's Arabic-language news reporting. Some editors assert that Al Jazeera, particularly its Arabic-language media, is a partisan source with respect to the Arab–Israeli conflict. Al Jazeera's news blogs should be handled with the corresponding policy.
I've changed it to 1.75 for now, since most users use Vector and might find 1.6 more difficult to read than Timeless and mobile users might. Jc86035 (talk) 18:20, 24 January 2019 (UTC)
This idea complicates the list and would probably open a can of worms, so I'm not even sure if it would be desirable, but I'm throwing it out nonetheless:
Increase the levels of reliability from the current 3.
A hypothetical 5-tier system may look like:
Tier 1: Top-quality sources that are widely recognised in their field as consistently reliable or high-quality. This does not automatically mean that the source is reliable outside their field of expertise, or that the source isn't biased, though bias can easily affect a publication's reputation for reliability.
Tier 2: Generally reliable sources with a weaker reputation than Tier 1 sources. Can generally be cited, though minor considerations may apply. Attribution may be necessary in some cases.
Tier 3: No consensus.
Tier 4: Questionable, though may be cited in limited cases. Can still be used for opinion, with attribution.
Tier 2 - Al Jazeera, The A.V. Club, Business Insider, BuzzFeed News, Daily Dot, Entertainment Weekly, Forbes, Fox News (straight news reports), Mother Jones, National Review, Newsweek, People, Playboy, PolitiFact, The Register, Rolling Stone, Snopes, SPLC, VentureBeat, Vox
Tier 3 - Ballotpedia, CoinDesk, Daily Beast, Dotdash, Epoch Times, The Mary Sue, Page Six, PinkNews, Salon, TechCrunch, TMZ, Us Weekly, Vice, Washington Examiner
Tier 5 - deprecated sources, and currently "generally unreliable" sources with no special provisions for usability
Such a system may potentially be more informative to users, and help reduce debates on sources that seem to teeter between "generally reliable" and "no consensus". The main problem with a 5-tier system, besides being more complicated, is that it may not actually be effective at reducing debates, as instead of debates between "generally reliable" and "no consensus", we may now have debates on whether a source is tier 1 or tier 2. So... I'm not sure if I want this. feminist (talk) 17:56, 18 January 2019 (UTC)
This would be a huge change to Wikipedia in my view. If you are seriously proposing this in its current form then do it somewhere more visible. Emir of Wikipedia (talk) 18:42, 18 January 2019 (UTC)
I'm not formally proposing it at this point, just throwing this idea out to see if it would be seriously considered. feminist (talk) 03:19, 19 January 2019 (UTC)
My first impression of this proposal is positive. I would appreciate having more granularity in the levels of reliability, because this would allow the list to convey more information to readers at a glance. The purpose of this list is to save editors time and effort: there's nothing in the list that can't be obtained from manually and repeatedly sifting through past noticeboard discussions, but the existence of the list prevents readers from having to do so in order to assess a source's reliability. By making the contents of the "Summary" column more accessible, this proposal appears to further that purpose.
Currently, we actually have 4 tiers, and here's how they would correspond to the 5 new tiers in the proposal:
The most contentious aspect of the current 4-tier system is deciding whether to classify a source as generally unreliable or deprecated. This is because implementing an edit filter requires widespread consensus, which usually needs to be shown through an RfC on WP:RSN. (The only concrete difference between the "generally unreliable" and "deprecated" classifications is the edit filter.) I don't think the proposed 5-tier system would be as contentious, since there's no technical measure (like an edit filter) involved, and editors will continue to use their own discretion to decide whether the use of a source is appropriate for a given situation.
If this proposal were implemented, the new tiers would need to have very precise criteria to make it easy to differentiate between Tiers 1 and 2, and between Tiers 4 and 5, using only the information we have in the "Summary" column of the list. This proposal is only plausible if we can define criteria that are clear enough to prevent borderline cases where a source could arguably be in two different tiers.
We'll probably want to hold an RfC on this page to determine whether there is consensus for switching to the 5-tier system (addressing Emir of Wikipedia's concern). Before the RfC, it would be a good idea to create a draft subpage showing what the list would look like with 5 tiers, and containing the criteria for those tiers. — Newslingertalk12:19, 19 January 2019 (UTC)
Wikipedia:Reliable_sources#OverviewSource reliability falls on a spectrum: highly reliable sources, clearly unreliable sources, and many in the middle. Editors must use their judgment to draw the line between usable and unreliable sources. Similar things are scattered around Wikipedia but there has never really been a proper consensus as far as I can tell. --Emir of Wikipedia (talk) 16:40, 19 January 2019 (UTC)
Yes, a more granular classification would be more useful overall. However, the purpose of this page is to summarize past discussions, and not to reach novel conclusions about the reliability of sources. Unless these discussions are closed using this more granular system, I don't know that we can reliably reach that conclusion on our own in populating this list. GMGtalk16:28, 19 January 2019 (UTC)
No, I don't think we should remove it. We do have as fairly objective classifications: discussions which reach a clear consensus, and discussions which don't. I have suggested before replacing the symbol for "no consensus" with a question mark, rather than a cautionary symbol, since these discussions basically mean "go do some reading and figure out what the issues were", but that suggestion never reached a strong consensus itself. GMGtalk16:57, 19 January 2019 (UTC)
Re: your classification of how the current tiers would correspond to a 5 tier system. As you may see from my proposed examples, one of the purposes for a 5-tier system is to allow a small number of sources currently classified as "no consensus" to be moved up or down. So some of them which are closer to reliability may be placed at level 2, while some of them which are to be avoided in more circumstances, may be at level 4. feminist (talk) 08:03, 20 January 2019 (UTC)
My concern about this proposal is that making the categories more granular will make it more difficult to determine (and reach consensus on) which sources go in each bucket. The main purpose of this guide is to provide an index to discussions. Categorizing the discussions is useful, but more specificity would likely spawn disputes. I would dispute a few of the entries in the example posited by the OP. By have only three relatively clear categories, the chances of such disputes is minimized. I also don't think more categories would especially benefit editors consulting this list.- MrX 🖋 13:22, 24 January 2019 (UTC)
Basically, if we construct the criteria for this page correctly, we should have essentially no substantive content disputes on this talk page, and this should mostly be comparatively cosmetic discussions about things like presentation and formatting. If we screw that up, then all we've really done is create a higher tier of RSN with much less participation, which more than being unhelpful, may be actively harmful to reliable consensus building. GMGtalk13:09, 26 January 2019 (UTC)
RSN summaries
Would it be useful to have a supplementary page summarizing consensus, and maybe passing mentions of sources, in recently archived RSN discussions? (Presumably the newest discussions would be summarized first, being somewhat more useful due to reflecting newer consensus.) I think it would be useful for those sources which haven't been discussed enough to warrant an entry on this page. Jc86035 (talk) 16:46, 27 January 2019 (UTC)
Another way to handle this is to formally close recently archived discussions with {{archive top}} or a similar template, even if the discussions are not RfCs. This is optional, of course, but it could be useful for people who are reviewing the archives. I don't think passing mentions would be very useful, since they're less accurate (because they're less likely to be challenged). — Newslingertalk03:14, 28 January 2019 (UTC)
@Newslinger: I think this would also be useful, although I'm not sure how acceptable it would be to edit non-recent archived discussions. It might also seem unnecessarily bureaucratic, although this would also be a concern for the table-of-discussions approach. Jc86035 (talk) 05:25, 28 January 2019 (UTC)
Done Incidentally, I'm not really sure we're providing much value added here if we simply say in the notes section "there is no consensus", and we should instead be trying as much as possible to summarize the nuances of the particular discussions. I may try to go back over time and provide more in depth summaries. Of course, anyone is welcome to help with that. GMGtalk 12:43, 4 August 2018 (UTC)
for the non consensus sources? To my mind, something like this 
) is an appropriate choice. K.e.coffman (talk) 22:38, 4 September 2018 (UTC)
Stale, something something explanation, for entries where the most recent discussions are say...five years old or more and the reliability issues aren't tied to some core aspect of the source. So for example, Wikinews is unlikely to change, because the problem is a fundamental part of the source. But something like WorldNetDaily may somehow get their act together and be more reliable at some point. GMGtalk 12:19, 29 October 2018 (UTC)
Generally reliable: Editors show consensus that the source is reliable in most cases.
Generally unreliable: Editors show consensus that the source is questionable in most cases.
2010
in articles. If you want to discuss how the source is used or have concerns about the source (e.g. potential bias), you can start a discussion at the reliable sources noticeboard, which will cause Vox to be indexed in the list. Otherwise, Vox appears to be uncontroversial despite being used frequently, and that's a positive indicator of the source's reliability. — Newslinger talk 22:17, 17 January 2019 (UTC)
