Wikinews talk:Cite sources

What about reputability of sources? I seem to recall WpA having a big thing with that, is it the same here? 68.39.174.238 23:16, 25 July 2006 (UTC)

Source order detail

The page says currently "It doesn't matter what order your references appear in" - it seems like there is a strong preference to list sources in order from most recent to least recent (although I can't find that mentioned in the Style page). Should we maybe note that here or on the style page? JoshuaZ 01:26, 30 October 2007 (UTC)

I have often made edits to resort sources in this way. I guess the practices of those before me, resorting my sources, made me pass it on regardless of explicit policy. However, there should be a policy, and recent first makes sense, especially in 'breaking news' stories. But I have noted that the natural tendency is to list sources chronologically, top-down. --SVTCobra 01:37, 30 October 2007 (UTC)

How about "It does not matter what order your references appear in but chronological order with most recent first is a commonly used convention"? I'd almost favor just making it "Sources should be listed chronologically" since at least in theory policy reflects what the community consensus is about how to do things. JoshuaZ 01:43, 30 October 2007 (UTC)

It does matter according to the Archive Conventions. "Sources should be sorted newest -> oldest". Jcart1534 01:48, 30 October 2007 (UTC)

It figures that you would know the archiving conventions. Ok, we should have that somewhere more prominent. At minimum we need to deal with that that policy then contradicts this one. JoshuaZ 01:52, 30 October 2007 (UTC)

I suspected that the 'recent-first' policy came from somewhere, and not just hand-me-down practices. In light of this, I suggest making it policy for every news item. If for no other reason, it makes it easier on the archivers. --SVTCobra 01:54, 30 October 2007 (UTC)

Are we saying the same thing, JoshuaZ? ----SVTCobra 01:57, 30 October 2007 (UTC)

Actually I edit conflicted with you- I was going to clarify what I was saying to make clear that I was saying pretty much what you just said. Only detail is that I'd add it to the Style page to since people will likely look there for this sort of info. JoshuaZ 01:59, 30 October 2007 (UTC)

I agree, it should be on WN:SG as well. --SVTCobra 02:05, 30 October 2007 (UTC)

Ok, if there are no objections I'll go add it here and at WN:SG. JoshuaZ 23:58, 31 October 2007 (UTC)

Support I do not object. However, should this have been flagged first or brought up at Wikinews:Water cooler/policy? --SVTCobra 00:12, 1 November 2007 (UTC)

I've brought it up at the Water Cooler. JoshuaZ 00:17, 1 November 2007 (UTC)

Multiple Sources

Shouldn't we include somewhere that we require multiple sources —The preceding unsigned comment was added by Anonymous101 (talk • contribs)

"Only articles recent enough"

It took me a minute to figure out what "only articles recent enough that they have themselves passed through the system of independent peer review" may be used as sources meant. It referred to the reintroduction of the review system didn't it? "Recent enough" had me thinking in terms of weeks. If it's been ten years, maybe an update is in order.

1) Do we still want to permit Wikinews articles to be used as sources at all? 2) If so, should we change this line to something less confusing, like "only sources from after [2009], which were published under our current peer review system." —The preceding unsigned comment was added by Darkfrog24 (talk • contribs) 00:58, 9 March 2019 (UTC)

Re (1), of course we use Wikinews articles as sources. This is basic technique, on Wikinews and in conventional journalism.

(I'll take a look at the precise phrasing, re (2), when I'm less tired yet have a moment not reviewing. Probably not a good idea to build a specific date into it rigidly, although, yes, it'd be around 2009; for the somewhat less individually sensitive matter of deleting pre-publish redirects we used 2009.) --Pi zero (talk) 02:18, 9 March 2019 (UTC)

Well, okay, tried something. @Darkfrog24: Is this revised wording clearer to you? --Pi zero (talk) 02:29, 9 March 2019 (UTC)

I think that will tip off newer (as in since 2010) participants, yes. It was 2009, right? I didn't go back and check. Darkfrog24 (talk) 02:32, 9 March 2019 (UTC)

I hedged a bit, in the wording; one really ought after all to carefully examine an article to be sure it was reviewed properly if one is going to use it. But, yeah, that's pretty much it; the edit to Wikinews:Reviewing articles marking it as a policy was made on January 25, 2009, and the associated discussion indicates the practice was already in force. --Pi zero (talk) 02:48, 9 March 2019 (UTC)

Date URL accessed

There doesn't seem to be any documented guidance on how to note what date a given URL is accessed. For example when citing utility pages such as those on GovTrack.us, which don't have a publication date (they aren't proper articles) and are updated as new information surfaces, we are to somehow note the date we accessed the site, i.e., (source accessed).

The guidance should recommend standard verbiage if we want standard verbiage. It might also be good to recommend that authors get an archived snapshot (from web.archive.org, for example) if possible when citing such sources.

Maybe a "date accessed" field could be added to template:source and the template could manage the verbiage. This could be useful even for citing proper articles, as they are often updated or even retracted.

I've received recommendation for two different versions of how to note the date accessed. That tells me two different people at least feel it should be standardized. Unfortunately the two recommendations were different from one another. ツ

I'm not trying to call anyone out or say I was mislead by anyone. In both cases the individuals were being helpful.

Michael.C.Wright (talk) 01:11, 21 March 2023 (UTC)

I was just thinking about this issue this morning. SVTCobra uses (source accessed), while Chaetodipus (cf. here) uses (date of access), also used by George Ho and the reviewing Cromium here; this article by WSS and reviewed by, of course, Pi zero, posits a third, (accessed). I myself have used (source accessed), but that was solely on the first's recommendation. Heavy Water (talk) 02:09, 21 March 2023 (UTC)

I think, I would tend to use (date of access) or at least something explaining the date. I don't know what I used to do in past decades. I probably used various forms. Cheers, SVTCobra 07:08, 25 March 2023 (UTC)

I had forgotten about this discussion and revisiting this page reminded me of it. Earlier this year I updated the style guide to implement the above (date of access) and after being reminded of this conversation, I just updated this policy with the same.Michael.C.Wright (^Talk/_Reviewer) 14:54, 16 February 2026 (UTC)

Wikipedia as a source

I have added a statement to the policy that Wikipedia is not a reliable source for en.WN articles.

I think (hope?) this is an uncontroversial addition of a long-standing convention that needs to be documented to combat "hidden" institutional knowledge. I think new users should be able to come here and easily find our policies and guidelines. This should also speed up the review process as we don't have to rehash the discussion each time in an article's talk page when one tries to use Wikipedia as a source.

I have also linked to an archived discussion about the topic, so that others can get a deeper understanding of the decision, should they choose to do so. —Michael.C.Wright (^Talk/_Published) 14:29, 24 September 2024 (UTC)

Statement added re: sourcing synthesis articles

I have added a statement clarifying that synthesis articles require two sources. This is stated in our WN:PILLARS, which is (surprisingly) only a guideline, yet we enforce the two-source rule as if it is a policy. —Michael.C.Wright (^Talk/_Reviewer) 16:49, 7 March 2025 (UTC)

Expanded section regarding redundant sources

I have expanded the section Use the sources to state that redundant sources are not desired. While this is implicit in the first and original statement, I think it is necessary to explicitly state and clarify it. I believe it is a habit of some Wikipedians to use a lot of sources to help improve legitimacy of edits but that isn't necessary or desired here, given our unique review process. —Michael.C.Wright (^Talk/_Reviewer) 14:14, 14 May 2025 (UTC)

Change request

This conversation has been marked for the community's attention. Please remove the {{flag}} when the discussion is complete or no longer important.

I suggest adding a paragraph at the top of the page:

"When a reviewer or a volunteer is involved with editing an article to ensure it is according to this policy, and some content is not sourced, the user has two options, either remove it or add a source. The removal may be recommended for news that is breaking, for news that is about to become stale, and also for content that may appear to be biased. When doing such a removal, the user is highly encouraged to leave a comment on the talk page of the article and on the talk page of its primary contributor, notifying of the change and advising that the content can be re-added if it is presented neutrally and is well sourced."

The reason why I am proposing this change is to reduce delays in reviewing, as in some occasions little pieces of content, that are not part of 5W nor highly relevant, get request for source and it delays publication.

Please let me know what you think. Gryllida 13:05, 15 August 2025 (UTC)

I Oppose this addition. It does not clarify or strengthen the core content policy, which states that "every piece of information in a Wikinews article must be referenced and verifiable."

The proposed paragraph is procedural in nature; describing what to do when material is unsourced, rather than defining the policy itself. As such, it would be more appropriate in a guide like Wikinews:Review checklist. The content policy should state what is required, not how to respond when those requirements aren't met.

If the concern is review delays, then the solution is to encourage contributors to include only verifiable content up front. Shifting that burden downstream to reviewers undermines the clarity of the policy and risks normalizing unsourced additions.Michael.C.Wright (^Talk/_Reviewer) 15:08, 15 August 2025 (UTC)

I have incorporated[1] this into Wikinews:Review checklist.Michael.C.Wright (^Talk/_Reviewer) 16:41, 22 August 2025 (UTC)

To clarify your comment, if information is not essential to 5Ws, it may be removed. Of course essential information would still need to be verified. Proposed wording:

"When a reviewer or a volunteer is involved with editing an article to ensure it is according to this policy, and some content is not sourced, the user has two options, either remove it or add a source. The removal may be recommended only in cases when the information being removed is not essential for meeting the 5W and NPOV requirement, particularly for news that is breaking, for news that is about to become stale, and also for content that may appear to be biased. When doing such a removal, the user is highly encouraged to leave a comment on the talk page of the article and on the talk page of its primary contributor, notifying of the change and advising that the content can be re-added if it is presented neutrally and is well sourced." Gryllida 03:19, 23 November 2025 (UTC)

> “To clarify your comment, if information is not essential to 5Ws, it may be removed.”

That is an incorrect reframing of my comment.

To clarify my comment: This proposal blurs two things that should remain distinct:

• Core policy (what is required)
• Procedure (how reviewers handle problems)

Also, adding language to our sourcing policy that encourages reviewers to remove unsourced material (even conditionally) signals to contributors that:

1. It is acceptable to submit drafts with unsourced claims. (it should remain unacceptable)
2. Reviewers will handle cleanup.

This cuts directly against the longstanding principle that contributors must do the heavy lifting; reviewers verify.Michael.C.Wright (^Talk/_Reviewer) 14:14, 23 November 2025 (UTC)

WN:PAYWALL

@Michael.C.Wright: you brought this up at Talk:The 62nd Munich Security Conference begins in Germany, so after reading the policy, am I to take that even The New York Times is not allowed, as in 'Drill, baby, drill': Trump cuts down Environmental Protection Agency mandate? HKLionel TALK 06:49, 15 February 2026 (UTC)

That is unfortunately correct.

The intent is for any of our readers to be able to verify content in our articles. If our readers don't have full access to our sources because they haven't subscribed to one, they can't verify our article. So we require sources to be freely available.

Even though I seem to have full access to that individual article without a subscription, I'm not sure why or for how long. Therefore I recommend removing that source and ensuring any facts supported only by it are either removed, or supported by a newly added source.

It is another of our oft-debated policies. See this thread as an example of past discussion on paywalled sources.Michael.C.Wright (^Talk/_Reviewer) 14:26, 15 February 2026 (UTC)

Astounding, but I see the rationale. Yes, NYT seems to be soft-paywalled. The thread you linked to actually links to my discussion with Gryllida (t · c · b) at Talk:Sanae Takaichi leads LDP to victory in 2026 Japanese election#When and How missing: I used an article from The Japan Times, which is apparently also soft-paywalled, and she requested that I replace the source. I believe this must be clarified as the 2022 thread apparently lead to nowhere. Both you and Gryllida's inputs would be much appreciated. HKLionel TALK 12:14, 16 February 2026 (UTC)

Personally, I believe in the following: "The public is entitled to as much information as possible to judge the reliability and motivations of sources."[2]

Paywalled sources prevent non-subscribers (to include non-subscribed reviewers) from judging the reliability and motivations of those sources.

If our WN:PAYWALL policy was changed, WN:REV would also need to be changed to reduce the level of verification required. That has been proposed for some time without support and I oppose the two latest proposals as-written.

If we want to be considered a citizen journalism project, we should conduct ourselves as such. If we want to be more like Wikipedia; a wiki first, then we should close up shop and edit there, because unverified "news" is not news. It is fake news.Michael.C.Wright (^Talk/_Reviewer) 14:31, 16 February 2026 (UTC)

Yes, so the point at which a source is considered paywalled needs to be clarified. HKLionel TALK 14:34, 16 February 2026 (UTC)

In the 2022 thread, Acagastya had a good point here, where they said: "There are a number of paywalled sources which will first have the article free for view and once it gets traction, it brings in the paywall, only adding to the frustration of the reviewer."

How would we work around that? If a source that we've previously verified turns 'paywalled' later, do we remove that source and any supported facts? If so, how do we know which statements it supported?Michael.C.Wright (^Talk/_Reviewer) 14:40, 16 February 2026 (UTC)

Perhaps editors who use sources with a history of paywalling could be required to provide a list of quotes that support specific claims in the article, on the talk page or some other place that the reader can access. However, this may raise copyright concerns (which I'm no expert on). Another way is to compile an information page about ways to circumvent paywalls (hopefully legally, such as resetting data collected by the website on the reader in some way such that the reader appears to have never accessed the website, something like that) and link to it for readers that want to verify claims. I believe this could actually help allow paywalled articles to be used, but that's another step to be taken at another time. HKLionel TALK 15:00, 16 February 2026 (UTC)

> Perhaps editors who use sources with a history of paywalling could be required to provide a list of quotes that support specific claims in the article,

I think if we implemented in-line citations for every fact, we could do that similar to how Wikipedia does it; which allows for footnotes where we could include the quotes without copyright concerns (proper quoting, with attribution does not violate copyright).

So if we already required in-line citations as a policy, I do see relaxing paywalled policy to allow individual articles that are open at the time of verification, as long as quotes were provided and verified by the reviewer during pre-publication review.

> Another way is to compile an information page about ways to circumvent paywalls
Circumvention has been discussed and was also briefly touched on in 2022 by @JJLiu112 here, where they said

"...archive has limited utility. For instance, this paywalled source from FT (https://web.archive.org/web/20220302161931/https://www.ft.com/content/b3d2bce8-11d7-4808-82fd-bd0417bd2ded) remains so."

Paywalled sources are becoming wiser to how people circumvent them.Michael.C.Wright (^Talk/_Reviewer) 15:09, 16 February 2026 (UTC)

Yes, I think that would be a good idea. For the latter, I wouldn't know much about the area, so no further comment. HKLionel TALK 15:19, 16 February 2026 (UTC)

Fyi HKLionel i did not like inline citations, for following reasons

• makes job harder for article writer
• reduces quality of a report as "<foo bar>(bbc citation)" is rather useless to a reader. because if it was a police speaker who said <foo bar> and bbc only reported that, then the reader should see "police speaker said <foo bar>" and not care whether it was bbc or cnn who reported it, unless the credibility of bbc or cnn was questioned. If credibility was questioned, write "bbc reported police speaker said <foo bar>".
• inline citations are used in wikipedia because they do not have an attribution policy. They just write "<foo bar>[1]" and reader does not care where it came from. This is because if that is incorrect, it can be corrected any time. In news, as it is reported once only, it is essential to attribute, so if <foo bar> was false, then at least "police speaker said <foo bar>" is accurate. There is no chance to come back and fix it, after a news was published.
• even if forgetting all of the above, inline citation is useless ro a reviewer because finding "foo bar" in a bbc source does not deprive a reviewer of a necessity to check all sources about "foo bar".

If that is implemented i think there should be a clear note that it is only a guidelibe, not all reviewers required it, should not be published with the inline citations only for draft work, and a software should allow a reviewer to delete all inline citations. I would strongly oppose that too. Gryllida 18:48, 16 February 2026 (UTC)

I did not mean to suggest the use of inline citations, just a list of quotes that people can use to verify. HKLionel TALK 19:01, 16 February 2026 (UTC)

Attribution in prose (“Police spokesperson said…”) and in-line citation serve different functions. Attribution identifies who made a statement. Citation identifies where the claim was verified. One does not replace the other.

Since news articles are intended as snapshots in time, that supports stronger traceability at publication, not weaker. Limited post-publication correction increases the importance of clearly mapping claims to sources during review.

In-line citations do not remove the need to review all sources, but they make explicit which source supports which claim. That can reduce ambiguity and expose unsupported statements more quickly.

If concerns are about clutter or workflow burden, that is a practical discussion. But describing in-line citations as "useless" overlooks their role in claim-level verification (and their use in university, scientific publications, etc).

It may help to separate the questions:

• Do they improve verifiability?
• Are the workflow costs acceptable?
• Should they remain in published articles or only in drafts?

Clarifying those points keeps the discussion focused on principle rather than preference.Michael.C.Wright (^Talk/_Reviewer) 19:23, 16 February 2026 (UTC)