Wikipedia:Village pump (policy)

"WP:VPP" redirects here. For proposals, see Wikipedia:Village pump (proposals).

 Policy Technical Proposals Idea lab WMF Miscellaneous 

• Table of contents
• First discussion
• End of page
• New post

The policy section of the village pump is used to discuss already proposed policies and guidelines and to discuss changes to existing policies and guidelines.

• If you want to propose something new that is not a policy or guideline, use Village pump (proposals).
• If you have a question about how to apply an existing policy or guideline, try one of the many Wikipedia:Noticeboards.
• If you want to ask what the policy is on something, try the Help desk or the Teahouse.
• This is not the place to resolve disputes over how a policy should be implemented. Please see Wikipedia:Dispute resolution for how to proceed in such cases.
• If you want to propose a new or amended speedy deletion criterion, use Wikipedia talk:Criteria for speedy deletion.

Please see this FAQ page for a list of frequently rejected or ignored proposals. Discussions are automatically archived after remaining inactive for two weeks.

« Archives, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191

Petition to amend ARBPOL to add options for U4C

Should ARBPOL be amended to add appealability and submission of questions to U4C? _signed, SpringProof ^talk 04:31, 23 March 2024 (UTC)

I am hereby petitioning the following two changes to the Arbitration Policy:

A: The following sentence shall be added to WP:ARBPOL#Appeal of decisions:

Questions strictly concerning the Universal Code of Conduct may be severed and appealed to the Universal Code of Conduct Coordinating Committee, which shall decide to hear it or not.

B: The following sentences shall be added to WP:ARBPOL#Policy and precedent:

Prior to publishing a decision, the Committee may refer questions of policy solely regarding the Universal Code of Conduct to the Universal Code of Conduct Coordinating Committee, which shall be required to answer, unanimously or by majority, in a reasonable timeframe.

I am petitioning these amendments in preparation for the upcoming U4C elections, which will establish the U4C. Part of their charter includes the option for projects to submit appeals concerning the UCoC, so I thought that might be helpful to add to ARBPOL.

These amendments are severable and may be adopted by themselves, so I have separated them into A and B.

_signed, SpringProof ^talk 04:31, 23 March 2024 (UTC)

_{Disclosure: I am currently a candidate for the U4C.}

Signatories for A

1. Petitioner, _signed, SpringProof ^talk 04:31, 23 March 2024 (UTC)

Signatories for B

1. Petitioner, _signed, SpringProof ^talk 04:31, 23 March 2024 (UTC)
2. Agree Slacker13 (talk) 13:04, 27 March 2024 (UTC)

General comments (ARBPOL U4C petition)

These proposals misunderstand what the U4C was created to do, and I hope they'll be withdrawn. The charter is very clear that the U4C doesn't generally have jurisdiction "when a NDA-signed, high-level decision-making body exists", and on en-wiki that's ArbCom. ArbCom should be interpreting the UCOC on its own (if necessary, which it rarely is), and the UCOC couldn't even hear appeals from those decisions if it wanted to except in extraordinary cases of "systemic failure". Anything else would be at odds with both the charter and this project's independence. Extraordinary Writ (talk) 05:52, 23 March 2024 (UTC)

@Extraordinary Writ: I understand that the U4C doesn't already constitutionally have jurisdiction over appeals. If there already was, this petitioned amendment would be moot (see above). I think the UCoC involves more disputes than it's chalked up to be. For example, the only open case right now is centered around a UCoC issue (What constitutes paid editing?). Love your name, by the way. :) _signed, SpringProof ^talk 07:38, 23 March 2024 (UTC)

Expanding the U4C's jurisdiction is even more problematic, I think. Even if it could be done without amending the U4C charter (which I doubt), giving the U4C additional authority over ArbCom would be a serious blow to this project's self-governance, and I think it's very unlikely that you'll find 100 editors who'll support doing so. (Paid editing is a Terms of Use issue, not a UCOC issue, by the way.) Extraordinary Writ (talk) 21:03, 23 March 2024 (UTC)

@Extraordinary Writ: You're right, I apologize. Nevertheless, the case also includes an issue of alleged doxing, which is further part of the UCoC. _signed, SpringProof ^talk 05:08, 24 March 2024 (UTC)

• This proposal misses the entire point of the UCOC, which is to provide a method of dispute resolution on projects that don't already have methods; in particular, smaller and newer projects. I fully expect to see medium- to large-sized projects without an arbitration committee creating one so that they don't have to deal with the U4C. Keep in mind that the UCoC itself is largely adapted from English Wikipedia policies and their corollaries on other large projects. This seems like massive overreach. Risker (talk) 23:11, 24 March 2024 (UTC)

  That's what I would like the UCoC to be. However, UCoC is more ambitious about its scope. Its main page claims that it may not be circumvented, eroded, or ignored by ... local policies of any Wikimedia project. It dictates that all who participate in Wikimedia projects and spaces will: [list of demands] and that it applies equally to all Wikimedians without any exceptions. Of course, any attempt to enact such arrogance may see significant numbers of us advise the WMF where to stick its encyclopedia, but those who wrote that text don't seem to be here to play second fiddle to ArbCom. Certes (talk) 23:47, 24 March 2024 (UTC)

Oppose this per others above: this is just more WMF stuff encroaching on enWP's jurisdiction. 🌺 Cremastra (talk) 22:48, 26 March 2024 (UTC)

I also oppose. UCoC may claim precedence over ArbCom, the laws of physics and all major deities, but U4C doesn't and shouldn't. Let us continue to answer to locally elected representatives rather than our new global overlords who have parachuted in uninvited. Certes (talk) 23:09, 26 March 2024 (UTC)

• This isn't useful. For (A) if something is within the scope of UCOC review it doesn't require a local policy to make it as such. For (B) local polices can't make global bodies act. — xaosflux ^Talk 14:21, 27 March 2024 (UTC)

  It seems to be suggesting that ArbCom defer to the U4C, which I suppose ArbCom could do if it wished, but it certainly isn't obliged to and I'd rather it didn't. Certes (talk) 14:30, 27 March 2024 (UTC)

  If I've understood correctly, then (A) would allow users to appeal some arbcom decisions to the U4C, whether to do so would not be a decision arbcom could make. If so then this is pointless as the UCOC and U4C determine whether the latter can hear appeals of ArbCom decisions, not local policy. It also attempts to mandate the U4C making a decision on whether to hear a specific appeal or not - legalistically it can't do that, but in practice the only other option is to ignore the request which I would sincerely hope they wouldn't do.

  (B) is really in two parts. The first part allows (but doesn't require) ArbCom to refer UCOC policy questions to the U4C if they want to. I don't have a problem with this in principle, but whether answering such questions is a function of the U4C is a matter for the UCOC and U4C to decide not en.wp policy, and I also don't think it is something that needs a policy amendment to allow given that ARBPOL doesn't restrict who the committee can consult. The second part attempts to require the U4C to answer arbcom's questions and to answer them in a "reasonable timeframe". English Wikipedia policy has no more ability to do this than it has to require the US Congress to answer arbcom's questions.

  Together that makes this whole thing a mixture of pointless and moot. Thryduulf (talk) 14:54, 27 March 2024 (UTC)

• It's credibly claimed above that in practice, our ArbCom disapplies the UCOC to en.wiki. If so, then we should make a clear declaration of this in a prominent place.—S Marshall T/C 16:00, 27 March 2024 (UTC)

  @S Marshall what doesn't apply to English Wikipedia is the Universal Code of Conduct Coordinating Committee (U4C). The community has never been given a chance to ratify the Universal Code of Conduct (UCoC) itself. This has always struck me as a mistake, though the WMF Board does seem to have the power to make it policy anyway. Either way, the UCoC is a set of minimums and it is my firm judgement that enwiki policies often go far above those minimums and in no place are our policies less than the UCoC. Barkeep49 (talk) 21:57, 27 March 2024 (UTC)

  On the basis of your last sentence, I modify my previous position to: "On en.wiki, our governance and policies make the UCOC nugatory." If that's right, it's rather important, and I do think we should say so.—S Marshall T/C 22:09, 27 March 2024 (UTC)

• This isn't useful. ArbCom is ArbCom. U4C has no supervisory jurisdiction over ArbCom. Stifle (talk) 10:08, 2 April 2024 (UTC)

Sources: clarify that they may be on a linked page

I wish to seek to change the wikipedia policy WP:SOURCE. Currently this states "Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source." in the section Responsibility for providing citations. I propose amending this with the additional sentence "Sources may be contained in a linked article."

RATIONAL FOR THE PROPOSED CHANGE

I believe that requiring sources on every page brings a number of problems: 1) it is onerous and inefficient and discourages linking relevant articles to pages, especially for new editors: 2) the relevant article may include more sources, mentions of the article might only include one, so anyone looking for useful information might not see it; 3) in a rapidly moving field sources may be updated in an article but that might be missed on linked pages. In any case it is easy for anyone to click on the link to see the article with all relevant sources. Hewer7 (talk) 13:50, 23 March 2024 (UTC)

I disagree with this suggestion. If the same information is sourced in a different article, it's much less onerous for the editor to copy the source to the new article than to expect readers to go to other articles to verify the information. And we can't rely on other articles being properly sourced because, too often, they're not. Schazjmd (talk) 14:00, 23 March 2024 (UTC)

Wikipedia:Wikipedia is not a reliable source. It has been long established that we cannot cite other Wikipedia pages to support content in a Wikipedia article. It may be fruitful to review sources cited in other articles, but Wikipedia:Verifiability#Burden states, The burden to demonstrate verifiability lies with the editor who adds or restores material, and it is satisfied by providing an inline citation to a reliable source that directly supports the contribution. That means that an editor who is using a citation to a source found in another article must have verified that the source does indeed support the content being added. You cannot change just the one policy point you targeting, other points in other policies and guidelines would all have to be changed. Donald Albury 14:10, 23 March 2024 (UTC)

I am not proposing that wikipedia be used as a source. My proposal is that sources may be contained in a linked article. Hewer7 (talk) 14:42, 23 March 2024 (UTC)

When you rely on another linked article to have the cited sources to support content, you are indeed using that other article as a source for that content. Donald Albury 18:03, 23 March 2024 (UTC)

An example of why we don't use WP articles as sources (or rely on sources cited in other articles without verifying their suitability): An article I'm drafting (User:Donald Albury/Trail Ridge) refers to the geological Hawthorn Formation. I found that our article on the Hawthorn Formation was a stub, saying it is a stratigraphic unit in South Carolina. On the other hand, our article on the Hawthorn Group said it was a stratigraphic unit in north Florida. In fact, the Hawthorn Group, formerly called the Hawthorn Formation, is a stratigraphic unit stretching from southeastern South Carolina through coastal Georgia and down the Florida Peninsula. I had to find new sources and cite them to correct that mess. You can only decide that a Wikipedia article is correct if you check out the cited sources, and search for and check out other sources (in case the cited sources are incomplete), and if you do that, you should just go ahead and cite those sources in the article you are working on. Donald Albury 18:28, 23 March 2024 (UTC)

There is a very simple reason why we require citations to be repeated in every article where information appears… articles can change. The “linked” article may currently contain a citation that supports the information at the article you are working on… but there is no guarantee that this will be the case in the future. The other article might, at some point in the future, be completely rewritten - and in the process the citation that supports what is in your article might be removed. You have to repeat the citation in your article to ensure that the information will always be supported, no matter what may happen at the other one. Blueboar (talk) 14:44, 23 March 2024 (UTC)

In addition to that (not that that isn't enough, mind you), there's the fact that while most of us most of the time experience Wikipedia online, it's not the only way it can be used. A printed copy of an article that contains proper referencing has those references listed at the bottom of the article. If we switch to relying on the mere fact that there are references on some other page, those references may not accessible to the person using the printed version. -- Nat Gertler (talk) 15:55, 23 March 2024 (UTC)

I would oppose this change in policy. Besides the other issues mentioned above, this would make it much more onerous and error-prone for a reader to verify content. Suppose there is a sentence containing links to 5 other Wikipedia articles, with no citation. If the reader wants to verify this statement, they would need to click on each of those 5 links, scrutinize the linked article to try to find a similar statement and see if there is a source there. If they can't find such a source after spending 10 minutes or whatever in this process, they still don't know if they have just overlooked the source or if the original statement was simply unsourced. Having the source for a statement at the point where the statement is made is essential.

The OP says that the current process is onerous for editors. That's fine; if there is a part of the process that is onerous, it should be onerous for editors, not for readers. CodeTalker (talk) 20:39, 23 March 2024 (UTC)

+1 Regards, Goldsztajn (talk) 21:24, 23 March 2024 (UTC)

• No, absolutely not. This would invite all sorts of problems. The most obvious one is that it would become easier for a source's meaning to drift via a game-of-telephone; a slight mistake or paraphrase on one article that isn't a problem there could become something drastically divergent from the source on another article that relies on the first one's citation. And worse, it makes it harder to verify - what source in the linked page, exactly, and on what page, do I look at if I'm not sure it's summarized correctly on the second page? Finally, on top of all this, what if the relevant section is edited or removed and the source replaced or removed itself? Someone making that edit may not even know the page that relied on that source existed, so it would quietly become unsourced with nobody realizing that it had happened. We already have a problem with "source drift", especially in uncited lead sections, where text starts out reasonably summarizing the source and yet repeated edits for WP:TONE or perceived NPOV issues or the like, each one a reasonable rewording of the phrasing immediately prior to them, collectively cause the text to drift further and further away from what the source actually says. This would make the problem far worse. --Aquillion (talk) 03:55, 25 March 2024 (UTC)

  +1 I strongly oppose this idea, but Aquillion said it far better than I could. 🌺 Cremastra (talk) 20:58, 26 March 2024 (UTC)

This would go against verification, Wikipedia is never a reliable source for verification and there is nothing to say that the details on the other page are reliable or will even stay in the article. -- LCU ActivelyDisinterested «@» °∆t° 21:18, 25 March 2024 (UTC)

• Oppose. Sources should be on the page they relate to, so that verifiability can be met. Stifle (talk) 10:09, 2 April 2024 (UTC)
• Oppose. I understand where you are coming from, but the information architecture of wikipedia isn't formed to make this a robust option. The longer-term solution, which has been discussed from time to time, would be to create a centralized "citation library" where, for instance, a book referenced by many articles has a central citation which is called from each of the articles using that citation. The only way this would work in practice would be to have a bot-based transfer of citations to the library with in-page replacement. This is a wish and not a reality today. --User:Ceyockey (talk to me) 00:36, 11 April 2024 (UTC)

• Oppose. Citing another article is just straight-up using Wikipedia as a source which isn't good. It's also confusing for readers to tell if the claims are cited since now they would have to read another article to verify it themself which may or may not be cited as well.

TheWikiToby (talk) 04:50, 19 April 2024 (UTC)

Preference of using OpenStreetMaps

Dear @User:Shannon1 before reverting my edits please discuss here. These maps are preferred because they are zoomable and rich of metadata. If you disagree please discuss. Hooman Mallahzadeh (talk) 15:19, 29 March 2024 (UTC)

@Hooman Mallahzadeh: Hi, can you link me to the Wikipedia documentation or discussion that indicates the OSM maps are "preferred"? The watershed maps are valuable to river articles because they show key information like drainage basin extent, tributaries and topography. I wouldn't be opposed to including both in the infobox, but there appears to be no way currently to display two maps. Shannon [ Talk ] 15:22, 29 March 2024 (UTC)

I should note that in French Wikipedia it is used correctly for Seine, In Japanese used for Arakawa River (Kantō). This is correct use of maps in the year 2024. Hooman Mallahzadeh (talk) 15:24, 29 March 2024 (UTC)

@Shannon1 Policies doesn't say anything. But I can discuss and defend about their preference. Just compare these images:

Traditional map	New Maps

Which of these maps is more clear? The new or the old? Hooman Mallahzadeh (talk) 15:38, 29 March 2024 (UTC)

I really think that we should create a policy for the preference of OpenStreetMaps over traditional ones. Hooman Mallahzadeh (talk) 15:40, 29 March 2024 (UTC)

I think they serve different purposes, and it would be ideal to have both in the infobox - but there appears to be no way to do this at the moment. The OSM map would be a fantastic replacement for pushpin locator maps like on Walla Walla River. However, it deletes a ton of important information that is displayed in the older watershed map. Can we hold off on any kind of mass replacement until this can be resolved? Shannon [ Talk ] 15:43, 29 March 2024 (UTC)

1. OpenStreetMaps presents the least but most important metadata at each level of zoom.
2. The ability of zooming is only provided by OpenStreetMaps
3. If any change occurs for the river, for example the path changes, this is rapidly applied for OpenStreetMaps
4. language of metadata changes automatically for each Wikipedia
5. and many others. Just let me some time to write them.
6. font-size of text of metadata is automatically adjusted

Hooman Mallahzadeh (talk) 15:44, 29 March 2024 (UTC)

You should have tried to get agreement for that policy before attempting to impose your preference across a large number of river articles. Kanguole 16:09, 29 March 2024 (UTC)

@Kanguole Ok, we are here for agreement about that. Hooman Mallahzadeh (talk) 16:14, 29 March 2024 (UTC)

@Hooman Mallahzadeh: Please revert the map changes you have made, since they have been challenged and there is so far no agreement for them. Kanguole 21:04, 29 March 2024 (UTC)

If it's an article about a river, the traditional map is more informative. 🌺 Cremastra (talk) 21:01, 29 March 2024 (UTC)

@Shannon1 See, we can have both maps by using "Hidden version of maps in infoboxes"

{{hidden begin|title=OpenStreetMap|ta1=center}}{{Infobox mapframe |wikidata=yes |zoom=6 |frame-height=300 | stroke-width=2 |coord={{WikidataCoord|display=i}}|point = none|stroke-color=#0000FF |id=Q1471 }}{{hidden end}}

that is rendered as:

OpenStreetMap

which yields: (here we hide topological and show OpenStreetMap, but the reverse can be applied)

Seine
The Seine in Paris
Topographical map
Native name	la Seine (French)
Location
Country	France
Physical characteristics
Source
• location	Source-Seine
Mouth	English Channel (French: la Manche)
• location	Le Havre/Honfleur
• coordinates	49°26′02″N 0°12′24″E
• elevation	0 m (0 ft)
Length	777 km (483 mi)
Basin size	79,000 km2 (31,000 sq mi)
Discharge
• location	Le Havre
• average	560 m3/s (20,000 cu ft/s)
Basin features
River system	Seine basin
Tributaries
• left	Yonne, Loing, Eure, Risle
• right	Ource, Aube, Marne, Oise, Epte

We can have both maps, one is hidden by default, and the other is shown by default. But I really think that we should show OpenStreetMap and hide others. But in many rare cases that the revert is true, we show topographic map and hide OpenStreetMap. Hooman Mallahzadeh (talk) 15:54, 29 March 2024 (UTC)

We want an edit for Template:Infobox river and use parameters hidddenMap1 and probably hiddenMap2 for implementing this idea. Hooman Mallahzadeh (talk) 16:07, 29 March 2024 (UTC)

I opened a thread on Template talk:Infobox river regarding this. Also pinging @Remsense: who has been separately reverting my edits. Shannon [ Talk ] 16:09, 29 March 2024 (UTC)

I'm merely concerned specifically with the articles I've reverted, I have no opinion on the issue at-large. Remsense诉 16:16, 29 March 2024 (UTC)

@Remsense: I've been on Wikipedia 15+ years and river articles have always used these watershed maps. I'm aware that policies can change but there has been no such discussion at WP:RIVERS or elsewhere. In my view, the watershed map on Yangtze for example is far more informative than the OSM map, which is essentially a better locator map. The Yangtze basin is immense, with dozens of major tributaries, and in this case the OSM map also leaves out the Jinsha that continues for more than 2000 km upstream of Sichuan. (Not because I made the watershed map, necessarily – I just noticed the reversions because of my watchlist.) Shannon [ Talk ] 16:25, 29 March 2024 (UTC)

I'll revert on these pages for now, thank you for the elaboration. Remsense诉 16:35, 29 March 2024 (UTC)

If you really want consistent guidelines (after working out technical issues), put them on WikiProject Geography. A global policy would just be MOS:BLOAT. SamuelRiv (talk) 16:39, 29 March 2024 (UTC)

@SamuelRiv I made a discussion for that here. Thanks, Hooman Mallahzadeh (talk) 16:51, 29 March 2024 (UTC)

@Shannon1:For my final word, I really cann't read the metadata of this map, because text on it is too small:

unless opening it. So its metadata is useless at the first glance, unlike OpenStreetMap.

• Not sure where to put this comment, because this section is broken with huge amounts of whitespace making it almost unreadable. I just want to mention that i have reverted three or four river map changes by Hooman Mallahzadeh, the summary of the diff indicated that the rather ugly and not as useful Open Street Map was preferable; my summary is "By whom is it "preferred"? Don't think there's a policy on this; until any discussion is finished the better map shouldn't be removed." I see now that a discussion (not a vote at all) has been started here. I'd like to suggest that Hooman Mallahsadeh reverts all the changes they have made of this type until this discussion comes to some conclusion. Happy days, ~ Lindsay^Hello 20:26, 29 March 2024 (UTC)

Proposal 1: Render both; prefer OSM; hide others

Ok, please vote for this scenario.

"Both topographic and OpenStreetMaps will be rendered in Infobox, but it is preferred to show OpenStreetMap and hide others by using "Template:Hidden begin" and "Hidden end".

For "vote", I asssume you mean "discuss"? 🌺 Cremastra (talk) 21:04, 29 March 2024 (UTC)

Agree with proposal 1 re OSM

1. Agree Hooman Mallahzadeh (talk) 16:23, 29 March 2024 (UTC)

Disagree with proposal 1 re OSM

1.  Disagree The OS map (in the way it is implemented here; don't know if layers in OS can be switched off for this kind of view) shows too much information that is not relevant for river articles (like roads, for example), and not enough information about what these articles are about - rivers. Plus, the watershed maps are just prettier IMO. Zoeperkoe (talk) 18:08, 29 March 2024 (UTC)
2.  Disagree Some maps are better for some things. For example in river or lake articles, the watershed maps are more helpful, but for city maps OSM is probably better. 🌺 Cremastra (talk) 21:03, 29 March 2024 (UTC)

   @Cremastra@Zoeperkoe Why OSM is preferred? Because it is more abstract, and for solving our problems, it is preferred to move from reality into concept. Please read the article Concept. In fact, we want to solve our problems by concepts that only includes main data and lacks redundant data. So certainly OSM maps are appropriately more abstract and finer concept.

   For example, in this image:

   

   The abstracted version of tree is preferred for many applications (question answering) like addressing and others over Cypress tree.

   So. in river Infoboxes, I even propose to use wider lines to remove elaboration of rivers and make a simpler map for its Infobox at the first glance. Hooman Mallahzadeh (talk) 05:22, 30 March 2024 (UTC)

   As someone who also likes the OSM maps in general cases: "read the Concept article" is not a very compelling argument.

   My argument would be that they are more flexible and more immediately maintainable by editors. We can theoretically better control the level of abstraction or detail we need for a given article. I don't mind cracking open the text editor to edit an SVG, but not everyone wants to do that. I've seen enough infobox crimes to know that dogmatism either for maximum abstraction or concretion is counterproductive. Remsense诉 05:28, 30 March 2024 (UTC)

3.  Disagree For users with Javascript disabled (either by choice or by force), OSM maps are useless. No movement, no zoom, and nothing drawn on top of the base tiles. Also no ability to swap between tiles. Please ensure that whatever choice you make fails safely without scripts. 216.80.78.194 (talk) 11:10, 31 March 2024 (UTC)

   When I disable JS in my browser, the maps above still render with the lines indicating the rivers' courses. They do miss the ability to click to see a larger interactive version, but they're not useless. Anomie⚔ 13:22, 31 March 2024 (UTC)

4. OSM map is much less informative for the topic of rivers. CMD (talk) 06:17, 7 April 2024 (UTC)

   @Chipmunkdavis Being less informative is an advantage. The purpose of an Infobox is providing some general information, not detailed information. In an Infobox, only the most important and most readable data should be shown. Other maps can contain details, not the Infobox map. Hooman Mallahzadeh (talk) 06:52, 7 April 2024 (UTC)

   While I think this position is preferable to the other extreme which is far more common in infobox disputes, I think it's a perspective being wielded too dogmatically here. While it's fun when I say things like "being less informative is an advantage" and there's a real sense where that's true, it also misses the point here that no one size fits all when it comes to presenting key information, and a watershed is important information one would like to know at a glance. It's being mischaracterized in my opinion as a detail, what others are arguing is that it is not so. Remsense诉 07:05, 7 April 2024 (UTC)

   @Chipmunkdavis@Remsense Yes. But the most abstract data version is in the first zoom, if you want more abstract version do "zoom out" and if you need more detailed version, do "zoom in",

   But at the first glance, if is not enough informative, then for example for "watershed", we can use "point locators" on the map. Or for areas we can use area locators. They are added very fast by using new items of Template:Maplink. The same as Shinano_River. Hooman Mallahzadeh (talk) 07:20, 7 April 2024 (UTC)

   I agree it's a potential solution. But we should judge the solution on a case by case basis, rather than making a swap across an entire class of articles now. Remsense诉 07:22, 7 April 2024 (UTC)

   An in this particular case, the watershed and to an extent tributaries is important and immediately visually readable. CMD (talk) 12:29, 7 April 2024 (UTC)

5. Disagree. I have just been reading a river article i happened to come across (River Wyre) which has made me feel so strongly that i have had to return here and protest these OSM maps, though i had planned not to. The map in that particular article, as well as other river articles i have looked at recently, is not sufficient: It gives no idea of the area drained by the river, there are unexplained dotted and faint grey lines all over it which apparently give no information, and (in this particular case) it is huge compared to the other images in the article. I am rather worried by Hooman Mallahzadeh's statement above, [b]eing less informative is an advantage, which i strongly disagree with; we should be giving our readers an abundance of information and allowing them, if they so desire, to choose what they wish to take away. Happy days, ~ Lindsay^Hello 07:42, 7 April 2024 (UTC)

   In the context of an infobox it is understandable what they mean. However, the point here is I think it's perfectly reasonable to display a river's watershed in the infobox. Remsense诉 07:54, 7 April 2024 (UTC)

   @Remsense See French Wikipedia at this page https://fr.wikipedia.org/wiki/Seine . It displays both start and end with pointer and then in the continuum of Infobox, it discusses start and end of the river. I think this convention of French Wikipedia describes rivers (and also Seine river) fantastic. Hooman Mallahzadeh (talk) 09:02, 7 April 2024 (UTC)

   Remsense, i agree that the infobox should contain the watershed ~ the thing is, if it doesn't, the information (presumably in the form of a map) would need to be elsewhere in the article. The infobox is indeed the logical place to look. Happy days, ~ Lindsay^Hello 13:19, 7 April 2024 (UTC)

   @LindsayH Please do not be surprised about my statement! Just see the Occam's razor article, ending line of the first paragraph:

   "The simplest explanation is usually the best one."

   And this sentence:

   In philosophy, Occam's razor (also spelled Ockham's razor or Ocham's razor; Latin: novacula Occami) is the problem-solving principle that recommends searching for explanations constructed with the smallest possible set of elements.

   And this sentence:

   Entities must not be multiplied beyond necessity.

   I don't know what is your major, but this principle is applied to all theories in science. Hooman Mallahzadeh (talk) 08:07, 7 April 2024 (UTC)

   Hooman Mallahzadeh, i think you're possibly misunderstanding Ockham's razor: It says nothing about withholding information to make things simpler, what it means is that given a certain number of observations or facts the simplest explanation which covers them all is to be preferred. So i am still concerned (maybe even more so now) about your desire to give our readers less information. Happy days, ~ Lindsay^Hello 13:19, 7 April 2024 (UTC)

   @LindsayH «Least information» but «most important information», in addition, it should be readable at the first glance, topological maps are usually unreadable at the first glance. Hooman Mallahzadeh (talk) 13:24, 7 April 2024 (UTC)

   My point is that this aphorism has exhausted its usefulness, and that this should be decided case by case, not as a class. Remsense诉 14:28, 7 April 2024 (UTC)

   Occam's razor has to do with problem-solving. If we apply to everything, then we get rid of everything as being too complicated. Cremastra (talk) 01:34, 11 April 2024 (UTC)

   It's always puzzling to me when people bring up Occam's razor as if it lends any credence to a particular philosophical argument, where it universally translates to "the right answer is probably the one that seems right to me". Remsense诉 01:38, 11 April 2024 (UTC)

   I think it's a useful metric when evaluating if an idea has a lot of edge cases or exceptions. If you can find a different idea that covers the topic without edge cases, it suggests that the "edge cases" aren't actually edge cases but rather refutations.

   That being said, I don't see how Occam's rasor applies to the question at hand. 104.247.227.199 (talk) 10:13, 20 April 2024 (UTC)

Neutral

1. I support the inclusion of both, but there is no need to hide one or the other. See the current documentation of Template:Infobox river. The OSM implementation would be a good replacement for the dot locator map, but it does not at all adequately replace a topographical map showing basin-level details. I am aware of the limits of image maps particularly regarding language, but 1) this is the English Wikipedia and this primarily concerns pages in English; 2) replacing existing .jpg and .png maps with SVG maps would enable maps to be easily edited for translation; and 3) if a map isn't available in a certain language, then just using the OSM version is fine. Shannon [ Talk ] 19:00, 29 March 2024 (UTC)

• Please see WP:NOTVOTE. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:26, 29 March 2024 (UTC)

• Yeah, that isn't how policy decisions are made. AndyTheGrump (talk) 21:32, 29 March 2024 (UTC)

• Im a huge OSM map fan, but to say that a it is preferred OVER a topographical map goes way too far. editorial discretion as always should apply, and blanket 'rules' for things like this almos always backfire. —TheDJ (talk • contribs) 10:19, 20 April 2024 (UTC)

Proposal 2: Include both (OSM and topographic maps) when appropriate

This seems like it best approaches existing consensus:

When appropriate, both a topographic map and OpenStreetMaps should be included in infoboxes.

Remsense诉 01:07, 30 March 2024 (UTC)

@Remsense Just see how beautiful Japanese Wikipedia introduced the river Shinano_River by this code:

{{Maplink2|zoom=8|frame=yes|plain=no|frame-align=right|frame-width=400|frame-height=600|frame-latitude=36.93|frame-longitude=138.48
|type=line|stroke-color=#0000ff|stroke-width=3|id=Q734455|title=信濃川
|type2=line|stroke-color2=#4444ff|stroke-width2=2|id2=Q11655711|title2=関屋分水
|type3=line|stroke-color3=#4444ff|stroke-width3=2|id3=Q11362788|title3=中ノ口川
|type4=line|stroke-color4=#4444ff|stroke-width4=2|id4=Q11372110|title4=五十嵐川
|type5=line|stroke-color5=#4444ff|stroke-width5=2|id5=Q11561641|title5=渋海川
|type6=line|stroke-color6=#4444ff|stroke-width6=2|id6=Q11437096|title6=大河津分水
|type7=line|stroke-color7=#4444ff|stroke-width7=2|id7=Q3304165|title7=魚野川
|type8=line|stroke-color8=#4444ff|stroke-width8=2|id8=Q11587633|title8=破間川
|type9=line|stroke-color9=#4444ff|stroke-width9=2|id9=Q11561259|title9=清津川
|type10=line|stroke-color10=#4444ff|stroke-width10=2|id10=Q11366441|title10=中津川
|type11=line|stroke-color11=#4444ff|stroke-width11=2|id11=Q11674896|title11=鳥居川
|type12=line|stroke-color12=#4444ff|stroke-width12=2|id12=Q11530256|title12=松川
|type13=line|stroke-color13=#4444ff|stroke-width13=2|id13=Q11571106|title13=犀川
|type14=line|stroke-color14=#4444ff|stroke-width14=2|id14=Q11626952|title14=裾花川
|type15=line|stroke-color15=#4444ff|stroke-width15=2|id15=Q11671931|title15=高瀬川
|type16=line|stroke-color16=#4444ff|stroke-width16=2|id16=Q11444998|title16=奈良井川
|type17=line|stroke-color17=#4444ff|stroke-width17=2|id17=Q11563522|title17=湯川
|type18=line|stroke-color18=#4444ff|stroke-width18=2|id18=Q59404662|title18=依田川
|type19=line|stroke-color19=#4444ff|stroke-width19=2|id19=Q59490451|title19=西川
|type20=line|stroke-color20=#4444ff|stroke-width20=2|id20=Q59537584|title20=黒又川
}}

This includes all sub-rivers. I think this type of maps should be a good sample for all other Wikipedia to introduce rivers. Hooman Mallahzadeh (talk) 13:18, 30 March 2024 (UTC)

I personally quite like this, yes. I'm sure if there's some argument against this, we will be hearing it—I like when other editors hone my aesthetic senses. Remsense诉 13:21, 30 March 2024 (UTC)

It looks very useful. I also stumbled across the Syr Darya page which manages to use both types of map in the infobox using the |extra= field. I would say that's a good, clean way to approach it going forward. Again, I think both types of maps are useful in different ways, and I see no reason to take an absolutist stance and say one or the other should be favored in all cases.

To add, I was kind of rubbed the wrong way at the start of this debate by OP's attitude that new and high tech is always better regardless of the context or usage (not to mention inventing an imaginary consensus which totally threw me for a loop), and as others have commented, this isn't how policy decisions on Wikipedia are made. Finally, as someone passionate about river topics, the auto generated maps just don't tell the full "story". It's nuance and individual approach versus cold standardization. Yes, there are a lot of poorly drawn and inaccurate user-made maps out there (including many of my older maps) which could do well with being replaced, but then there are beautiful ones like Rhine, which provide a value much harder to replace.Shannon [ Talk ] 16:53, 30 March 2024 (UTC)

@Shannon1 Even in the article of Rhine and in the selected map of Infobox, the font is too small and we can't read anything. So aside from choosing OSM or not, between existing maps, the second map i.e., File:Rhein-Karte2.png is more appropriate for Infobox map of this article. I think we should make a policy for selecting between maps, the one that is more abstract, i.e. we apply this policy:

The simplest and most abstract map is the preferred one for Infobox of articles

Hooman Mallahzadeh (talk) 17:56, 30 March 2024 (UTC)

I have already made my point, so I'll excuse myself from further argument on this thread. As I've stated, I support applying both maps where possible as I believe that provides the best value for the reader. I don't particularly mind if the OSM or topographic map is placed first or second in the infobox. However, I cannot agree with the assessment that "the simplest and most abstract map is preferred" in the context of rivers, which are complex systems that are much more than a simple blue line. Unless a broader consensus can be reached, I maintain to oppose any removal of useful content that have been considered standard on river articles for years. Shannon [ Talk ] 19:56, 30 March 2024 (UTC)

Proposal 3: Selection of varous types of "topographical maps" as background for OSM

I think this "alignment scenario" would be perfect:

OSM maps of rivers remains unchanged, but OSM white background could be changed to various topographical backgrounds by users.

Implementing this idea has challenges about setting correct size and challenges of alignment of two maps, but its implementation is not hard. Hooman Mallahzadeh (talk) 10:39, 20 April 2024 (UTC)

I'm sure it can work fine, but I still am not quite understanding why we would need to codify it as policy. Everyone has pretty much re-reiterated their preference for "just figure out what works on a per article basis", and you haven't really articulated why there's anything wrong with that. Remsense诉 10:41, 20 April 2024 (UTC)

@Remsense We should apply a policy is for "the selection of a map between various maps" for Infoboxes, which is for "First Glance Data". Wrong selection could give no data at the first glance. Hooman Mallahzadeh (talk) 10:45, 20 April 2024 (UTC)

I'm not sure I understand. Editors are currently free to decide what is best for each article, as per usual. Unfortunately, I don't think the type of arguments you've made are going to convince other editors that we should restrict editors' flexibility like that. If you want to improve the site, I think working on individual articles and discussing how to improve their maps for each would be more helpful to the site, because I still don't see a need to change sitewide policy. Remsense诉 10:51, 20 April 2024 (UTC)

You said

Editors are currently free to decide what is best for each article, as per usual.

Editors should select what type of map for infobox? In the most cases (over 90%), the «simplest map» is the best for infobox. Do you agree? But in very special cases other maps should be used for Infoboxes. Isn't it better to be a «policy»? Hooman Mallahzadeh (talk) 11:00, 20 April 2024 (UTC)

I don't think so, no. Let editors make their own choices per article. You are working in generally correct principles, but this would be applying them too dogmatically, as mentioned above. Remsense诉 11:02, 20 April 2024 (UTC)

@Remsense But I really think that the selection of File:Bassin Seine.png for Seine river happened in English Wikipedia is wrong. Selection of French Wikipedia for this river is more appropriate, because it provides more data at the first glance. If we apply a «selection policy», such bad selections would not happen anymore. Hooman Mallahzadeh (talk) 11:18, 20 April 2024 (UTC)

...then discuss the merits for that particular map on that particular talk page, like I've suggested several times! That's how Wikipedia generally works. I don't know how else to illustrate that your suggestion seems overly restrictive, and the flexibility seems more worthwhile here, but please try to understand what I'm saying with that, I guess? Remsense诉 12:42, 20 April 2024 (UTC)

How to describe past events on the main page

Currently, the status quo for events listed on the main page is to use the present tense, even if the event in question has definitively ended. I didn't really notice this was an issue until yesterday when I noticed that the main page said that the Solar eclipse of April 8, 2024 is visible through parts of North America. Knowing that it was not currently visible and double checking that the article referred to the event in the past tense, I changed this to was visible. [1] I did not realize that this is against the current consensus at WP:ITNBLURB which says that these events must always be described in the present tense. If one is interested in further background, I encourage them to read this discussion here (scroll down to errors).

I think that this status quo is misleading to readers because it cases like this, we are deliberately giving inaccurate and outdated information. I believe this is a disservice to our readers. The eclipse is not visible anymore, yet we must insist that it is indeed visible. I think that we should also be consistent... If the article for a blurb is using the past tense, we should use the past tense on the main page. Therefore, I propose that events listed on ITN that have definitively ended should be described in the past tense if it would otherwise mislead readers into thinking an event is ongoing. Clovermoss🍀 (talk) 11:33, 10 April 2024 (UTC), edited 17:00, 10 April 2024 (UTC)

Note: Notification of this discussion was left at Wikipedia talk:In the news.—Bagumba (talk) 12:00, 10 April 2024 (UTC)

I propose that events listed on ITN that have definitively ended should be described in the past tense: But any blurb can be written in the past tense, e.g., a country was invaded, an election was won, a state of emergency was declared, etc. So if we did go to past tense, I don't understand why there is a distinction with needing to have "definitively ended".—Bagumba (talk) 12:07, 10 April 2024 (UTC)

I made the distinction because I felt our current approach was the most jarring in situations where we're literally misleading the reader. I don't really have any strong preferences either way on other situations and felt like it'd be for the best to make sure my RfC was clear and not vague. I'm not trying to change every blurb at ITN right now, hence the "definitive end date" emphasis. If someone wants more broader changes to verb tense at the main page, I'd say that warrants its own separate discussion. Clovermoss🍀 (talk) 12:16, 10 April 2024 (UTC)

Note The blurb currently reads A total solar eclipse appears across parts of North America[2]—Bagumba (talk) 12:33, 10 April 2024 (UTC)

I was about to suggest a rewording along these lines… so that the blurb is accurate while maintaining present tense. Blueboar (talk) 12:45, 10 April 2024 (UTC)

It's better than flat out saying visible, but this phrasing still implies that it is visible? Present tense when an event has ended implies that an event is still ongoing. Clovermoss🍀 (talk) 16:22, 10 April 2024 (UTC)

Appear means to start to be seen or to be present.[3] It doesn't say that it continues to be seen. Perhaps the previous blurb's problem was that it resorted to using is, incorrectly implying a continuing state, not that a present-tense alternative was not possble(??)—Bagumba (talk) 06:34, 11 April 2024 (UTC)

To be present is to continue to be seen (by those looking, at least). I think you're misreading that as to start to be seen or present. That second to be matters here (and so it appears bold). InedibleHulk (talk) 22:30, 11 April 2024 (UTC)

Support per nom, see no reason to oppose. Aaron Liu (talk) 13:19, 10 April 2024 (UTC)

Per below, there isn'ta clear way forward for this one. On one hand, "Liechtenstein wins the FIFA World Cup" should definitely remain that way, but this also causes situations like these. Maybe something like unless this wording directly encourages a misleading interpretation that the event is still ongoing., using an earthquake in present tense and this event in past tense as examples. Or maybe we should just IAR such cases. Aaron Liu (talk) 16:30, 10 April 2024 (UTC)

I don't think IAR is going to work as long as we don't have an explicit exemption because it'd be causing someone to explicitly go against consensus for their own ends. I switched the wording to "was visible" out of ignorance in regards to current standards, not because I was deliberately ignoring them. I think there might have been much more ado made about my actions if I had done this with a justification of IAR. I don't have issues with your proposed wording, because again, my biggest issue with all of this is intentionally misleading readers. Clovermoss🍀 (talk) 16:39, 10 April 2024 (UTC)

@Aaron Liu: I've changed the proposal to have "if it would otherwise mislead readers into thinking an event is ongoing". Does that address your concerns? Clovermoss🍀 (talk) 17:03, 10 April 2024 (UTC)

Support, though I find isaacl's alternative of including a time frame intriguing. Aaron Liu (talk) 17:11, 10 April 2024 (UTC)

Comment for a lot of blurbs, the present tense is fine, as it continues to be true. e.g. elections, "X is elected leader of Y" is correct and better than past tense, and same with sports matches that end up on ITN. A blanket change to past tense is disingenuous therefore, although swapping to past tense for events that happened (and aren't ongoing) seems somewhat reasonable. Joseph2302 (talk) 13:55, 10 April 2024 (UTC)

Isn't "Is elected" past tense? Though I agree that for situations where we can use the active voice, "Z legislature elects X as leader of Y" sounds better. Aaron Liu (talk) 14:05, 10 April 2024 (UTC)

"Is elected" is present tense, specifically present perfect. "Elects" is also present tense, simple present. Levivich (talk) 18:14, 10 April 2024 (UTC)

I thought "is elected" is passive voice. Voters are doing the electing, the elected person is passive in this situation. In passive voice "elected" is a past participle (also sometimes called the passive or perfect participle). (Side note: present perfect in English usually takes "have/has" as an auxiliary verb) —⁠andrybak (talk) 23:00, 23 April 2024 (UTC)

I think for time-bound events such as the eclipse, including a time frame would be the best approach to avoid confusion. Additionally, I think using past tense is fine. isaacl (talk) 17:09, 10 April 2024 (UTC)

I am in favor of past tense for everything. "Won the election," or "landslide killed 200" or "eclipse appeared" all read as fine to me. Newspapers using present tense makes sense because they publish every day (or more often). It doesn't make sense for ITN where items stay posted for days or weeks. Levivich (talk) 18:10, 10 April 2024 (UTC)

Something about ITN mostly using present tense just feels... righter. Regardless of staying posted for weeks, they are all quite recent compared to most other stuff we have on the main page. Also see historical present. Aaron Liu (talk) 20:30, 10 April 2024 (UTC)

I'll have what you're having. InedibleHulk (talk) 22:30, 11 April 2024 (UTC)

• Decide case-by-case: we can safely IAR in most cases. Cremastra (talk) 19:43, 10 April 2024 (UTC)
• No special rules for the main page: use the same tense we would in articles. We are an encyclopedia not a newspaper. (t · c) buidhe 20:37, 10 April 2024 (UTC)
• Object The present tense serves us well. It is the standard tense for headlines, certainly within the UK and I believe US too (though some MoS in the US is very different to the UK). I can't see anything in the proposal beyond change for the sake of change. doktorb _words^deeds 22:00, 10 April 2024 (UTC)

  Again, it is confusing to say that the solar eclipse is in the sky. Aaron Liu (talk) 22:05, 10 April 2024 (UTC)

  It would be confusing to switch from "is....was....did....has" in a single box on a typical ITN week. doktorb _words^deeds 22:28, 10 April 2024 (UTC)

  A typical ITN week does not have many blurbs that really need the past tense like the solar eclipse. Aaron Liu (talk) 02:37, 11 April 2024 (UTC)

• We should use the correct tense. Someone does not "wins" an election or sports match, they won it. The eclipse, after it ended, was visible over North America, but "is" visible is factually inaccurate at that point (and before it starts to happen, we should say it will be visible). A political leader does not "makes" a statement, they made it. On the other hand, it may be accurate to say that a conflict is going on, or rescue efforts after a disaster are underway. So, we should use the natural, normal tense that accurately reflects the actual reality, as it would be used in the article. Seraphimblade ^{Talk to me} 06:02, 11 April 2024 (UTC)
• Object I don't think I agree with the premise that ITN blurbs are phrased in the present in the first place. It's in the historical present tense. "A 7.4-magnitude earthquake strikes near Hualien City, Taiwan" doesn't give the impression that the ground is still shaking. Nor does "A solar eclipse appears across parts of North America" read as "a solar eclipse is happening right now." Likewise, "Nobel Prize–winning theoretical physicist Peter Higgs (pictured) dies at the age of 94." doesn't need to be changed to "died at the age of 94", we know it's in the past, we're not under any illusions that he's still in the process of dying. It's phrased in such a way that doesn't really imply either past or present and just kind of makes sense either way. If an event is still happening, the blurb makes sense. And if the event is over, the blurb still makes sense. I think that's intentional.  Vanilla  Wizard 💙 07:33, 11 April 2024 (UTC)

  Actually, I think that "A 7.4-magnitude earthquake strikes near Hualien City, Taiwan" does give the impression that the ground is still shaking, or at least that it was shaking very recently. Even newspaper headlines avoid that, especially after the first day. "Hualien struck by massive earthquake" is a perfectly normal headline style. In fact, I find these actual headlines in the past tense:

  • Taiwan Struck by Deadly 7.4-Magnitude Earthquake
  • Taiwan shaken but unbowed as biggest quake in 25 years spotlights preparedness
  • Taiwan hit by powerful earthquake
  • Taiwan hit by its strongest quake in quarter-century, but death toll is low
  • Earthquake in Taiwan blamed for at least 9 deaths as buildings and roads seriously damaged
  • Taiwan hit by strongest earthquake in 25 years, killing 9

  WhatamIdoing (talk) 01:00, 20 April 2024 (UTC)

• Keep present tense as general recommendation per above. Discuss individual cases when this is too jarring. —Kusma (talk) 07:43, 11 April 2024 (UTC)
• As an encyclopedia rather than a news agency, I would think past tense fits our vibe more. Archives of our frontpage would remain clearly accurate indefinitely. We are not reporting news, we are featuring a newly updated/written encyclopedic article on currently relevant events. ~Maplestrip/Mable (chat) 08:22, 11 April 2024 (UTC)
• Keep present tense. There is a difference between "X is happening" (which necessarily means right now, at this moment) and "X happens" (which os somewhat more vague). We should always use the second form, regardless of precise moment. As stated above, we even have statements like "an earthquake hits..." or "So and so dies", both of which are clearly over by the tine it gets posted. Animal lover |666| 19:12, 11 April 2024 (UTC)
• Object from a wp:creep standpoint To my knowledge there is no rule regarding this and it's just a practice. This would change it to having a rule. North8000 (talk) 19:25, 11 April 2024 (UTC)

  How? The present tense rule was always written down there and this proposal does not make ITN a guideline. Aaron Liu (talk) 19:42, 11 April 2024 (UTC)

• No, it should not – it's unencyclopaedic and ungrammatical. The Simple Present is used to describe habitual or continuous actions or states (the Sun sets in the West; he is a boot-and-shoe repairman; I'm Burlington Bertie, I rise at ten-thirty; Timothy Leary's dead etc). Events in the past are described using the Present Past when when no time is specified (the lunch-box has landed; London has fallen; mine eyes have seen the glory ...). When a time in the past is specified, the Simple Past is invariably used: in fourteen hundred and ninety-two, Columbus sailed the ocean blue, in fourteen hundred and ninety-three, he sailed right back over the sea; today, I learned; well I woke up this morning and I looked round for my shoes. This is not rocket science. Ours is not a news outlet with a profit target to meet, we have no reason to have 'headlines', which are simply bits of news given some kind of extra urgency by being in the wrong tense. "Wayne Shorter dies!" immediately begs the question "really? how often?" So "A total eclipse of the Sun has occurred; it was visible in [somewhere I wasn't] from [time] to [time]". It gives the information, it's written in English, where's the problem? (NB there are two distinct present tenses in English, the Simple Present and the Present Continuous; the latter is used for things that are actually happening in this moment or about to happen in the future (I'm going down to Louisiana to get me a mojo hand; I’m walking down the highway, with my suitcase ...). Justlettersandnumbers (talk) 20:22, 11 April 2024 (UTC)

  @Justlettersandnumbers: Reading your comment makes it sound like it supports of my proposal instead of opposing it? I don't understand the "no, it should not" unless there's something I'm not getting. Clovermoss🍀 (talk) 21:10, 11 April 2024 (UTC)

  @Clovermoss The title of your section begins with "Should the main page continue to use the present tense". Aaron Liu (talk) 22:49, 11 April 2024 (UTC)

  And then the actual RfC itself is my proposal to change that for situations where this would be misleading readers. I'm not sure it's necessarily the best idea to be messing around with section names at this point but I'm open to suggestions that would help make this less confusing for people. Clovermoss🍀 (talk) 22:53, 11 April 2024 (UTC)

  Eh, never mind. I decided to be bold and make it consistent with how CENT describes this discussion. Hopefully that helps things. Clovermoss🍀 (talk) 23:15, 11 April 2024 (UTC)

• Support Given that WP:ITNBLURB currently has the guideline that "blurbs should describe events in complete sentences in the present tense," it does not seem like instruction creep to modify an existing rule. isaacl recommends including a time-frame, but I find this impractical for events that occur over multiple time zones. While this eclipse's article reports the event's span over the overall planet in UTC, this level of detail is too cumbersome for a main page blurb. Clovermoss' proposal limits itself to cases where the present tense would be confusing, which is preferable to an individual discussion for each perceived exception to the current guideline. BluePenguin18 🐧 ( 💬 ) 20:50, 11 April 2024 (UTC)
• Yes, the practice should continue - this is a perfectly normal idiomatic feature of English. Headlines are written in the present tense, just like 'in which...' in the chapter sub-headings of old novels, the summaries of TV episodes in magazines and on streaming services, and lots of other places where a reported past action is summarised. GenevieveDEon (talk) 21:33, 11 April 2024 (UTC)
• How about, "is seen over North America" -- passive with present tense and past participle, anyone? :) Alanscottwalker (talk) 21:49, 11 April 2024 (UTC)

  That's a better solution than ending the practice of using the historical present tense. Though I think that suggestion is more likely to be implemented at WP:ERRORS than through a Village Pump policy proposal. (I'm also not entirely sure why this whole discussion isn't just at the ITN talk page since it doesn't affect any other part of the main page, but it's no big deal)  Vanilla  Wizard 💙 20:10, 12 April 2024 (UTC)

  ERRORS is not the appropriate venue, given that the discussion that was there was removed. As for why it's here specifically, I figured anything regarding the main page was important, that a discussion here would invite more participants, and avoid the possibile issue of a local consensus. Clovermoss🍀 (talk) 20:16, 12 April 2024 (UTC)

  I originally thought this suggestion was sarcastic, given the smiley face. If it is serious, I dislike it because "is seen" is extremely passive voice. Assuming there is a problem (which I don't think there is), the solution is not passive voice. CaptainEek ^{Edits Ho Cap'n!}⚓ 20:22, 12 April 2024 (UTC)

  I don't think passive voices are that bad; while I agree that the active voice is usually preferred, do you really think that "North Americans see a total solar eclipse" is better? Aaron Liu (talk) 21:32, 12 April 2024 (UTC)

  No. I think that the current iteration "A total solar eclipse appears across parts of North America" is perfect. CaptainEek ^{Edits Ho Cap'n!}⚓ 21:37, 12 April 2024 (UTC)

  I was illustrating why the passive voice doesn't deserve to be demonized. Aaron Liu (talk) 21:42, 12 April 2024 (UTC)

  In fairness, that discussion was removed specifically because ITN uses present tense and the discussion was proposing to change that, and ERRORS isn't the place for proposals to change how we do things. Alanscottwalker's suggestion also uses the present tense, so ERRORS would be a fine venue if they really wanted to see that change made. After all, that discussion at ERRORS is what resulted in the language being changed from "is visible" to "appears". I personally think appears is totally fine (I agree with CaptainEek that there is no problem), but if someone prefers "is seen", that's the place to do it.  Vanilla  Wizard 💙 20:33, 12 April 2024 (UTC)

  That discussion only happened because I changed "is visible" to "was visible", prompting an errors report. I'd prefer "appeared" over "appears" since that implies that it is still indeed visible per the above discussion. It's better than "is visible", though. Clovermoss🍀 (talk) 01:07, 13 April 2024 (UTC)

• Keep present tense as ITN is supposed to summarize and collect news headlines and the present tense is standard in headlines. Pinguinn 🐧 00:05, 12 April 2024 (UTC)
• Keep using historical present I think a lot of supporters here are confusing the historical present (often used in news headlines) for the simple present. I would agree that the eclipse would have made sense to be an exception to that general rule, as was the focus in the original proposal here, but I wouldn't change the general rule. Anomie⚔ 12:04, 12 April 2024 (UTC)

  Currently, in this proposal, I see a codified exception for when using the present tense would be confusing that would only apply in cases like the solar eclipse. Aaron Liu (talk) 12:43, 12 April 2024 (UTC)

  @Anomie, the lead of our article on the historical present says the effect of the historical present is "to heighten the dramatic force of the narrative by describing events as if they were still unfolding". I'm not convinced that making things sound more dramatic should be a goal for an encyclopedia, and I would not have guessed that you would support such a goal. Do you? WhatamIdoing (talk) 01:09, 20 April 2024 (UTC)

• Keep historical present tense Headlines are most compelling and appropriate in the historical present tense. The NYTimes provides that "Headlines are written in the historical present tense. That means they written are in present tense but describe events that just happened."
  Out of curiosity, I perused the AP Stylebook (56th edition, 2022-2024), which surprisingly had almost nothing to say on tenses, though its section on headlines is generally instructive.

  "Headlines are key to any story. A vivid, accurate and fair headline can entice people to dig in for more. A bland, vague or otherwise faulty headline can push readers away. Often, a headline and photo are all that many readers see of a story. Their entire knowledge of the piece may based on those elements. Headlines must stand on their own in conveying the story fairly, and they must include key context. They should tempt readers to want to read more, without misleading or overpromising."

  How to best have a vivid headline? Present tense and active voice! One of Wikipedia's most frequent writing errors is using past tense and passive voice out of a misplaced assumption that it is more encyclopedic. But past and passive are weak. Present and active are better, and are what I have been taught in a wide multitude of writing courses and professional spaces. To add to the NYTimes, AP, and personal experience, I consulted my copy of Bryan Garner's Redbook (4th ed.), which while meant as a legal style guide, is useful in other areas. Regarding tense, in heading 11.32, it provides that "generally use the present tense." I then turned to the internet, which backed up the use of present tense in headlines: Grammar expert suggests present tense "Engaging headlines should be in sentence case and present tense." Kansas University on headlines: "Present tense, please: Use present tense for immediate past information, past tense for past perfect, and future tense for coming events."
  Using the historical present is best practice for headlines. That's not to say that there can't be exceptions, but they should be rare. As for the eclipse, it properly remains in the historical present. As a further consideration: if we are updating ITN tenses in real time, we are adding considerable work for ourselves, and we push ourselves truly into WP:NOTNEWS territory. CaptainEek ^{Edits Ho Cap'n!}⚓ 18:35, 12 April 2024 (UTC)

  I don't think we're adding considerable work for ourselves. It takes a second or two in the rare situations that require it, anything else regarding the main page has much more work involved. We already update the articles in question, just not the blurb, which is a bit of a jarring inconsistency in itself. I don't understand the argument that the tense we should be using should be comparable to newspaper headlines because we're NOTNEWS? Could you elaborate a bit on your thinking there? Clovermoss🍀 (talk) 19:43, 12 April 2024 (UTC)

  For the last part: they're mistaken that this proposal would require tenses to be updated to the past tense when any event ends, which is way too much effort to stay current which kinda does fall into NOTNEWS. (Note that this proposal would only require past tense if the historical present causes confusion) Aaron Liu (talk) 19:50, 12 April 2024 (UTC)

  We are NOTNEWS. But as my comment above alludes to, ITN is a de facto news stream. Each entry in ITN is effectively a headline. Why try to reinvent the headline wheel? I'm afraid I have to disagree with Aaron's clarification, because Clover did change the tense after the event ended. It would have been incorrect to say "was" when the blurb first posted...because the eclipse was presently happening at that time. I'll add further that "otherwise mislead readers into thinking an event is ongoing" is an unhelpful standard. I don't buy that the average reader is going to be confused by a historical present headline. We read headlines all the time, and the average reader understands the historical present, even if they couldn't define it. CaptainEek ^{Edits Ho Cap'n!}⚓ 20:18, 12 April 2024 (UTC)

  I have to disagree with you there. I think that when the main page stated that the eclipse "is visible", that was confusing to the average reader. It confused me, prompting me to check that the eclipse wasn't somehow ongoing. We were giving inaccurate information intentionally and I honestly don't see why we do this for the main page. Because it's interesting? Because newspapers do it before an event happens? Once the eclipse ended, newspapers referred to the event in the past tense as well. My decision to change it to "was visible" took one second (so not a considerable time investment, although everything that ensued certainly has been). Clovermoss🍀 (talk) 20:32, 12 April 2024 (UTC)

  Ah, that's my bad, the "is visible" language is also problematic for its passivity. I like the "appears" solution, and thought that was the original wording. But I think it would be improper to say "appeared." I'm not so sure I buy that newspapers were uniformly using past tense; again, the best practice for newspapers is to use the historical present. The time issue is ancillary to the best practice issue, I agree that the real time sink is the discussions that will surely result from implementing this rule. CaptainEek ^{Edits Ho Cap'n!}⚓ 20:42, 12 April 2024 (UTC)

  I could show some examples if you'd like, since you don't seem to buy that newspapers were using the past tense after the eclipse appeared.

  • "A total eclipse of a lifetime appeared for hundreds of thousands of visitors and residents in the Hamilton-Niagara region" – Canadian Broadcasting Corporation [4]
  • "In middle America, the eclipse was a phenomenon" – Washington Post [5]
  • "During the event on April 8, 2024, one of these arcs was easily visible from where I stood, agape beneath our eclipsed, blackened star, in Burlington, VT." – Mashable [6]
  • "The great American eclipse appeared Monday, bringing the nation to a standstill as photographers captured stunning shots of the rare celestial event." – CNET [7]
  • "The total solar eclipse that swept across Mexico, the United States and Canada has completed its journey over continental North America." – CNN [8]

  I think that "appears" is better than saying "is visible" like the previous phrasing was before my intermediate change of "was visible" but it still runs into the issue of implying the eclipse is appearing somewhere. I agree with what InedibleHulk said above To be present is to continue to be seen (by those looking, at least). I think you're misreading that as to start to be seen or present. That second to be matters here (and so it appears bold). Clovermoss🍀 (talk) 21:14, 12 April 2024 (UTC)

  The operative issue is that these are headlines from after the event. But the blurb got posted during the event. CaptainEek ^{Edits Ho Cap'n!}⚓ 21:19, 12 April 2024 (UTC)

  And the blurb stays days or weeks on the main page, where using the past tense would be more accurate than using present tense the entire time. I also think that having a clear exemption clause would prevent time sink discussions like this one, not cause them. It'd prevent us from needing to have a discussion every time something like this happens. Clovermoss🍀 (talk) 21:25, 12 April 2024 (UTC)

  I think that this discussion would prevent some time sink over reluctance to IAR. And again, only a small number of events would need their tense changed. Aaron Liu (talk) 21:34, 12 April 2024 (UTC)

• Drop present tense and use the tense we'd use anywhere else on Wikipedia. Wikipedia is not a newspaper, even on the Main Page, and there's no reason we should obscure the timing of events for stylistic reasons. Loki (talk) 21:18, 12 April 2024 (UTC)

  The tense we'd use anywhere else is, by default, present? WP:TENSE provides that By default, write articles in the present tense. CaptainEek ^{Edits Ho Cap'n!}⚓ 21:22, 12 April 2024 (UTC)

  MOS:TENSE says By default, write articles in the present tense, including those covering works of fiction (see Wikipedia:Writing better articles § Tense in fiction) and products or works that have been discontinued. Generally, use past tense only for past events, and for subjects that are dead or no longer meaningfully exist. We use past tense for past events like we do at the actual article linked in the ITN blurb: Solar eclipse of April 8, 2024. It's just the main page where we make the stylistic choice to not do that. Clovermoss🍀 (talk) 21:31, 12 April 2024 (UTC)

• The present tense makes the main page read like a news ticker, which we are often at pains to explain it is not (e.g. WP:NOTNP). I would favour the past tense for all events that are not ongoing. If we cannot agree on that, I support the proposal to use the past if there might be a misunderstanding (partly in the hope that familiarity will lead to the past tense being used more and more in the future!). JMCHutchinson (talk) 11:06, 13 April 2024 (UTC)
• Support Per WP:NEWSSTYLE, "As a matter of policy, Wikipedia is not written in news style ..." . ITN is especially embarrassing because its blurbs are often weeks old and so its use of the present tense is then quite misleading. It might help if the blurbs were dated to show how old they are. See OTD and the Spanish edition for examples. Andrew🐉(talk) 07:38, 18 April 2024 (UTC)
• Support the thing Clovermoss said we should do (to head off any confusion about whether "support" or "oppose" means to support or oppose making or not making a change, etc). jp×g🗯️ 06:30, 19 April 2024 (UTC)
• Oppose any firm rule. The same style is used in the not-so-current-events sections of year pages, or at least those I've checked so far:
  • From 520: The monastery of Seridus, where Barsanuphius and John the Prophet lived as hermits, is founded in the region of Gaza
  • From 1020: King Gagik I of Armenia is succeeded by Hovhannes-Smbat III.
  • From 1920: A woman named Anna Anderson tries to commit suicide in Berlin and is taken to a mental hospital where she claims she is Grand Duchess Anastasia of Russia.
  • From 2020: A total solar eclipse is visible from parts of the South Pacific Ocean, southern South America, and the South Atlantic Ocean.
• Now maybe I'm being a bit OTHERSTUFFy here and it's year pages that should be fixed, but until that's done, it would seem really weird to describe 1000-year-old events with "is", but events from last week with "was". Suffusion of Yellow (talk) 21:48, 19 April 2024 (UTC)

  None of these except the 2020 one can be mistaken as things that are currently happening. Aaron Liu (talk) 22:20, 19 April 2024 (UTC)

• I think we should use the past tense for some events (e.g., any event that is definitively "finished") and present tense for those that are ongoing. I didn't see a single clear argument above for using the present tense for things that are completely finished [correction: except for CaptainEek, who wants to use historical past for the "vivid" dramatic effect]. There are comments about what label a grammarian would apply to it, and comments saying that this is the way we've always done it, but no comments giving a reason for why it's better for readers if we say that a ten-second earthquake from last week "is" happening instead of that it "did" happen. WhatamIdoing (talk) 00:50, 20 April 2024 (UTC)

  Because the historical present is a convention in English, period. There's also consistency with lists of past events, which also blocks useful things like moving navboxes to the See also. Aaron Liu (talk) 00:53, 20 April 2024 (UTC)

  The historical present is a convention in English. It is not the only convention, which means we could choose a different one. Why should we choose this convention? WhatamIdoing (talk) 01:01, 20 April 2024 (UTC)

  For consistency and compactness. Aaron Liu (talk) 02:51, 20 April 2024 (UTC)

  The amount of compactness is usually one character – the difference between is and was, or elects and elected. In other cases, it's the same or shorter: shook instead of shakes for earthquakes, died instead of dies for deaths. I don't think that sometimes saving a single character is worth the risk of someone misunderstanding the text, especially since we get so many readers who do not speak English natively.

  As for consistency, I think that being easily understood is more important than having parallel grammar constructions across unrelated items. WhatamIdoing (talk) 05:28, 20 April 2024 (UTC)

  The historical present is not the convention anywhere on Wikipedia's main page. Just see today:

  • TFA: "The Nicoll Highway collapse occurred in Singapore...
  • DYK: "...librarian Amanda Jones won an award..."
  • OTD: "South African Airways Flight 228 crashed shortly after take-off ..."

  ITN is the only possible exception and it's not using the historical present because it's not referring to history.

  Andrew🐉(talk) 12:37, 20 April 2024 (UTC)

• I don't think anything needs to be changed here style-wise, we just need to write better ITN blurbs. "Solar eclipse is visible" isn't the historical present and it isn't sensible either. -- asilvering (talk) 06:21, 22 April 2024 (UTC)
• Not sure why this discussion isn't happening at WT:ITN, but stick with simple present as we have done for years. Stephen 09:49, 22 April 2024 (UTC)

  A notification has been at Wikipedia talk:In the news#Blurb tense for a while now. Putting this here attracts more attention.

  Most blurbs will not need to be changed to the past tense. Only things like "is visible" need to be changed. Aaron Liu (talk) 12:54, 22 April 2024 (UTC)

• The historical present should be taken behind the barn, shot, burned, and the ashes scattered to the four winds. --User:Khajidha (talk) (contributions) 14:02, 22 April 2024 (UTC)

Violently expressed dislike is not the same as a reasoned argument. The historic present is used widely in headlines, timelines, and other applications both on this site and elsewhere which are comparable to the ITN headlines. GenevieveDEon (talk) 14:16, 22 April 2024 (UTC)

Using present tense for completed events is ridiculous (which is even worse than wrong), no matter how much it may be used elsewhere. --~ User:Khajidha (talk) (contributions) 15:43, 22 April 2024 (UTC)

But Wikipedia cares about consistency, present tense saves characters, and most events will not be confused as ongoing. Aaron Liu (talk) 15:48, 22 April 2024 (UTC)

As I showed above, the present tense only occasionally saves characters, and the number of characters saved is most often one (1).

In my experience, the English Wikipedia cares more about clarity accuracy than about consistency. There are ~650 pages citing Emerson: "A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines." (And now there is one more.) WhatamIdoing (talk) 23:30, 22 April 2024 (UTC)

As you've said, I can't really articulate my thoughts on why we should use the historical present. I guess that's because not all grammar rules and conventions make sense either, yet they're usually prescribed. The most sense I could make is sort of "vividness": they emphasize that these events happen in the present day, as opposed to most of our content on the main page.

I also wish that Wikipedia didn't care so much about consistency, but it seems that we do, which has led to navboxes not being moved to the see also section and nearly all of them turned into the standard purple. Maybe that made me think to consistify the consistency. Aaron Liu (talk) 00:01, 23 April 2024 (UTC)

The rest of the main page uses past tense to refer to events that have occurred. The articles use the past tense to refer to past events. In the News isn't an up-to-the-moment news ticker; it points out articles that are related to current events. isaacl (talk) 00:14, 23 April 2024 (UTC)

Past, yes, but as you said they’re related to current events. These events are much more current than the rest of the main page and historical present emphasizes that.

Hopefully we have a rough consensus to at least put “otherwise confusing blurbs can you use the past tense” into the rules. Aaron Liu (talk) 00:17, 23 April 2024 (UTC)

The point is for the main page, using the same tense as the rest of the page, as well as the underlying articles, would be consistent. isaacl (talk) 01:26, 23 April 2024 (UTC)

The main page is just a reflection of the rest of the site. We don't need to force everything on the main page to be the same, and the underlying lists of stuff linked above also use historical present. Aaron Liu (talk) 11:14, 23 April 2024 (UTC)

The main page is just a reflection of the rest of the site. My understanding is that ITN blurbs are literally the only place we enforce this stylistic choice. It's inconsistent with the actual articles linked in the blurb. [9] I can't help but think that if this situation was the other way around (the status quo was to be consistent) that people would find the arguments for this unconvincing. Clovermoss🍀 (talk) 11:19, 23 April 2024 (UTC)

Most of the site doesn't use blurbs, but all the year articles do. See Suffusion of Yellow's comment above. Aaron Liu (talk) 12:27, 23 April 2024 (UTC)

I suppose then my question is if there's a consensus for year pages that things must be done that way then because it's not otherwise a stylistic choice you see outside of ITN blurbs. Clovermoss🍀 (talk) 15:02, 23 April 2024 (UTC)

You brought up consistency as an argument. I feel a reader will notice inconsistency amongst sections of the main page more readily than between the In the News section and the year articles. There's no navigation path between the latter two, but readers can easily jump between sections of the main page. isaacl (talk) 16:26, 23 April 2024 (UTC)

• Retain historical present. ITN blurbs are intentionally written in the style of news headlines, and that makes most sense given global usage on this point. It would be silly for Wikipedia to have a set of news items written differently from how every other outlet writes its news items. Cases like the eclipse can be handled on an individual basis, by rewriting the blurb into an alternative historical present form that removes the implication of ongoing nature. Arguably that blurb was simply badly structured in the first place as a normal headline wouldn't contain the word "is".  — Amakuru (talk) 09:50, 23 April 2024 (UTC)

  Wikipedia is not trying to be a news outlet; it's an encyclopedia. The correct comparison is then with a site like Britannica. Today, this opens with coverage of Passover:

  April 23, 2024
  Different from All Other Nights
  Last night marked the beginning of the Jewish holiday of Passover, which commemorates the Hebrews’ liberation from slavery in Egypt and the “passing over” of the forces of destruction, or the sparing of the firstborn of the Israelites, on the eve of Exodus. This year’s celebration occurs against a backdrop of conflict—today also marks the 200th day in the Israel-Hamas War—and heightened concerns of rising anti-Semitism.

  This makes the temporal context quite clear by dating the item and then using tenses accordingly -- the past tense for "last night" and the present tense for "today". Presumably tomorrow they will have a different item as their lead to reflect the fact that the present has moved on. This seems exemplary -- quite clearly explaining what's happening today specifically.

  Andrew🐉(talk) 11:05, 23 April 2024 (UTC)

  What about the present tense of "occurs"? I don't think a very long holiday is a good example.

  Looking at a few of their MP blurbs, most of them are anniversaries. Hopefully someone can find more examples of current events. Aaron Liu (talk) 11:13, 23 April 2024 (UTC)

Upgrade SCIRS to a guideline

Wikipedia:Identifying reliable sources (science) has been stable for years and is widely cited on article and user talk pages. It's in many ways similar to WP:MEDRS, which is a guideline. Isn't it time to bump SCIRS to guideline status too? – Joe (talk) 11:22, 11 April 2024 (UTC)

• I've left notifications about this proposal at the SCIRS talk page, WikiProject Science and WT:RS. Thryduulf (talk) 15:15, 11 April 2024 (UTC)

• I'm in general in favor of it, though it'll probably need some eyes going over it before going to guideline status, especially on cautions about using primary sources. Obviously a little more relaxed than WP:MEDRS, but not carte blanche use or outright encouraging primary sources either.

I have some guidance on my user page in the sourcing section that might be helpful there. In short, primary journal articles have their own mini-literature reviews in the intro and to some degree discussion/conclusions. When you are in a field that doesn't have many literature reviews, etc. those parts of sources can be very useful (e.g., entomology topics for me) for things like basic life cycle or species information. It's a good idea to avoid using a primary article for sourcing content on the findings of the study itself since it's not independent coverage though. That's not meant to be strict bright lines if it becomes guideline, but give guidance on how primary sources are best used if they are being used. If someone wants to use/tweak language from my page for updates, they'd be welcome to. KoA (talk) 16:20, 11 April 2024 (UTC)

I think we could say that in that circumstance, such as a paper is both a secondary source (in its discussion of other literature) and a primary source (in its results). I agree that the section on primary sources could be fleshed out, but I don't think it should be a blocker to giving it guideline status now (guidelines are never complete). – Joe (talk) 06:21, 12 April 2024 (UTC)

My thoughts exactly too in that it's an improvement that can be made independent of guideline or not. It would be a simple addition like you put, but it would also preempt concerns that sourcing would somehow be severely limited, which it functionally would not be.

If anything, much of what I mentioned here or at my userpage already addresses what has been brought up in a few opposes below. KoA (talk) 17:19, 12 April 2024 (UTC)

• Oppose. Maybe it's stable because we are free to ignore it. Maybe any useful advice in it is just what's already in other PAGs. Maybe we already have enough guidelines. WP:MEDRS was a bad idea too but at least had the excuse that dispensing bad health advice could cause legal problems; outdated cosmological theory has a somewhat smaller effect. Peter Gulutzan (talk) 17:51, 11 April 2024 (UTC)

• Support. This is necessary due to the huge and growing problem of the flood of unreliable research. As an engineer I edit scientific WP articles, and I waste an enormous amount of time dealing with noobs who come across some unsupported claim in a paper or sensationalist "science" website and are determined to put it in WP. And more time on pseudoscience advocates who dig up obscure papers that support their delusions. And more time on researchers trying to promote their careers by inserting cites to their own research papers in WP. In science today primary sources (research papers) are worthless, due to p-hacking the vast majority in even top journals are never confirmed. This needs to be reflected in our guidelines. --Chetvorno^TALK 20:51, 11 April 2024 (UTC)

• Support but... So unlike wp:ver & wp:rs (which require certain trappings and not actual reliability) we're going to require actual reliability for science articles? Requiring actual reliability puts it in conflict with wp:Ver and wp:RS.  :-) North8000 (talk) 21:06, 11 April 2024 (UTC)

  It's not my understanding that guidelines create "requirements", just offer best practices supported by consensus (WP:GUIDES). – Joe (talk) 06:09, 12 April 2024 (UTC)

• Support. Many longtime editors do not realize or refuse to acknowledge that primary sources should only ever comprise a small fraction of sourcing for an article. We also regularly have editors insisting various basic biology topics "aren't governed by MEDRS" because they don't have an immediate clinical relevance, and therefore the findings of primary research papers are acceptable. Having an actual guideline to point to that is more explicit on this would be helpful. JoelleJay (talk) 00:57, 12 April 2024 (UTC)

  This is also what I've found WP:SCIRS most useful for over the years. WP:PSTS is established policy, but it's not immediately obvious how to apply it to scientific topics without the extra guidance in WP:MEDRS or WP:SCIRS. We end up with sections that are just runs of "A 2017 study found, ..." then "A 2020 study found, ..." with no information on if any of those findings have achieved scientific consensus, because people see a journal article and assume that because it's reliable you can use it without qualification. WP:SCIRS clarifies which types of journal article are primary and which are secondary, and therefore how we should be using each type. – Joe (talk) 06:16, 12 April 2024 (UTC)

• Oppose Contra Joe Roe above, I think that Wikipedia:Identifying reliable sources (science) isn't an useful guidance on how to use primary vs secondary. In natural sciences, you tend to have articles that include a summary or review of existing science, followed by a paper's own conclusion - which by its very nature cannot say whether its findings have been widely accepted or not. That is, the same source is both primary and secondary, depending on which information you take from it. The essay isn't aware of this point. The problem with popular press isn't secondary/primary, either; rather that it tends to exaggerate and oversimplify i.e a reliability issue. Jo-Jo Eumerus (talk) 06:57, 12 April 2024 (UTC)

  We could just add that point? – Joe (talk) 07:06, 12 April 2024 (UTC)

  That would be a root-and-branch rewrite, as the idea of them being two separate kinds of sources is woven in its entire structure. In general, I think that WP:PSTS is a problem as it takes a concept mostly from history and tries to extrapolate it to other kinds of sources which often don't neatly map on it. Jo-Jo Eumerus (talk) 07:15, 12 April 2024 (UTC)

  If a primary source has a "summary or review of existing science", that existing science will be available in secondary sources, which are what we should use.--Chetvorno^TALK 07:36, 12 April 2024 (UTC)

  I don't see anything in WP:SCIRS or WP:PSTS that precludes a source being primary in some parts and secondary in others? WP:PSTS explicitly acknowledges that a source can be both primary and secondary at the same time: A source may be considered primary for one statement but secondary for a different one. Even a given source can contain both primary and secondary source material for one particular statement.. KoA observed the same thing above. It's a good point, and worth noting, but I think it can be easily achieved with an extra paragraph in WP:SCIRS#Basic advice, no rewrite needed. – Joe (talk) 08:26, 12 April 2024 (UTC)

  Only in well-covered fields. In less well covered ones like remote volcanoes, you often have one research paper that summarizes the existing knowledge before introducing its own point. But this guideline would apply to every field, not just the well-covered ones. ^It's not enough for the guideline to acknowledge the existence of "hybrid" sources; that's still assuming that most aren't hybrids and will mislead people into trying to incorrectly categorize sources. It's an undue weight issue, except with a guidance page rather than an article. Jo-Jo Eumerus (talk) 10:07, 12 April 2024 (UTC)

  I don't think I'm fully understanding you. A volcanology paper that summarizes the existing knowledge before introducing its own point is both a secondary source (in the first part) and a primary source (in the second part). How is this different from other fields? – Joe (talk) 10:20, 12 April 2024 (UTC)

  Sorry, that was addressing Chetvorno. Jo-Jo Eumerus (talk) 10:45, 12 April 2024 (UTC)

  I think the vast majority of people citing primary sources are citing them for their research findings, not their background sections. In the rare cases where they are citing the latter, if the material is contested on SCIRS/PSTS grounds then the editor can just point to where we say otherwise-primary sources can contain secondary info and say that's what they're citing. JoelleJay (talk) 16:18, 12 April 2024 (UTC)

  The trouble with the 'secondary' material in primary sources, is that the authors almost invariably spin it to align with their (primary) research conclusions. It should generally be avoided in favour of dedicated secondary sourcing. Bon courage (talk) 03:23, 16 April 2024 (UTC)

  I'm a bit short on time until next week, but I'd be willing to draft something based on my userpage (though a bit more flexible/advisory) if someone else doesn't get to it. KoA (talk) 17:29, 12 April 2024 (UTC)

• Oppose. <rant>The essay is an example of the primary vs secondary fetish that pollutes much of our policy. Actually there are very few things disallowed for primary sources that are not also disallowed for secondary sources. The rule should be "use the most reliable source you can find and refrain from original research". Instead, endless argument over whether something is primary or secondary replaces rather than informs discussion of actual reliability. So we get editors arguing that a newspaper report of a peer-reviewed journal article is better than the journal article itself, favoring the least reliable source for no good reason. Secondary reports of research are useful, for example they may contain interviews with experts other than the authors, but they are not more reliable than the original on what the research results were. Review articles are great, but rarely available. It is also not true that the existence of secondary reports helps to protect us from false/fake results; actually is the opposite because newspapers and magazines are more likely to report exceptional claims than ordinary claims.</rant> Zero^talk 10:28, 12 April 2024 (UTC)

  Well, it seems this is a common bugbear. Personally I've found the primary vs. secondary distinction very useful in doing exactly that, avoiding original research, but clearly others' mileage vary. Although it should be pointed out that, apart from discussing primary and secondary sources, WP:SCIRS strongly discourages using media coverage of scientific results (Wikipedia:Identifying_reliable_sources_(science)#Popular_press), so someone arguing that a newspaper report of a peer-reviewed journal article is better than the journal article itself would not find support in this essay.

  In any case, isn't the objection you and Jo-Jo Eumerus are articulating really against WP:PSTS, not WP:SCIRS? Not recognising a guideline because it fails to deviate from a policy would be... odd. – Joe (talk) 10:52, 12 April 2024 (UTC)

  If the policy is questionable, making a guideline that emphasizes the problematic aspects in a field where the problematic aspects are particularly problematic is making a problem worse. FWIW, while newspapers aren't my issue with SCIRS, I've certainly seen people claiming that news reports on a finding are secondary and thus to be preferred. Jo-Jo Eumerus (talk) 11:14, 12 April 2024 (UTC)

  And they cite SCIRS for that? It says the opposite. – Joe (talk) 12:21, 12 April 2024 (UTC)

  Joe, you are correct that my main beef is not with SCIRS. I haven't paid much attention to it, though I'd have to if it became a guideline. Mainly I severely dislike PSTS, which is full of nonsense, and I don't want more like it. Almost every word in the "primary source" section of PSTS is also the case for secondary sources. For example, "Do not analyze, evaluate, interpret, or synthesize material found in a primary source yourself" — since when are we allowed to do any of those things to a secondary source? And the only good thing about rule #3 is that it is largely ignored (unless "any educated person" knows mathematics, organic chemistry and Japanese). I could go on....I've been arguing this case for about 20 years so I don't expect to get anywhere. Zero^talk 14:32, 12 April 2024 (UTC)

  Zero0000 re: "...newspapers and magazines are more likely to report exceptional claims than ordinary claims". That is a different problem: what constitutes a reliable secondary source for a given field. SCIRS says: "Although popular-press news articles and press releases may tout the latest experiments, they often exaggerate or speak of 'revolutionary' results" So for scientific topics general newspapers and newsmagazines should not be considered reliable sources on a par with scientific journal reviews. WP:PSTS does not mention this issue; another reason SCIRS should be a guideline. --Chetvorno^TALK 23:32, 14 April 2024 (UTC)

• Oppose. Upgrading the "Identifying Reliable Sources (Science)" (SCIRS) to guideline status risks imposing unnecessary rigidity on topics that straddle the science and non-science boundary, and I believe that WP:MEDRS needs to be downgraded to an essay due to its frequent misapplication to part-biomedical topics, sometimes even in bad faith. As an essay, SCIRS provides useful advice without enforcing a strict approach that may not be suitable for all topics. By making it a guideline, we risk encouraging an overly simplistic distinction between primary and secondary sources, which may not always reflect the complexities and nuances of scientific inquiry, especially in interdisciplinary fields, or in burgeoning areas of research where established secondary sources may not yet exist. Furthermore, this rigidity could be abused, potentially serving as a gatekeeping tool rather than as a guide, particularly in contentious areas that intersect science with social or political dimensions, as seen with MEDRS in various topics. Maintaining the current flexibility that allows for context-sensitive application of source reliability is essential to ensure that Wikipedia continues to be a diverse and adaptable repository of knowledge. FailedMusician (talk) 23:42, 12 April 2024 (UTC)

  We already have flexibility in assessing secondary vs primary coverage within a source, per PSTS. Can you link some examples of MEDRS being misapplied? And if a topic has no secondary coverage at all, whether in review articles/books or in background sections of primary research papers, it certainly should not have its own article and likely isn't BALASP anywhere else either. JoelleJay (talk) 00:48, 13 April 2024 (UTC)

  I believe you are aware of cases where MEDRS has been misapplied. There are instances where primary sources, extensively covered in mainstream media, are challenged by misconstrued higher-quality sources, often to remove contentious content or editors. Introducing a new policy for scientific sources could further stratify our community, complicating contributions and inhibiting constructive dialogue. FailedMusician (talk) 10:14, 13 April 2024 (UTC)

  Where are these cases and why would you think I was aware of them? We haven't edited any of the same mainspace or talk pages or the same topics of any wikipedia- or user-space pages, so I don't know how you would get that impression. The only time I've seen editors claiming primary biomed sources need to be pushed into articles is in support of fringe views like the lab leak conspiracies. JoelleJay (talk) 01:58, 15 April 2024 (UTC)

  Scientific topics often cross multiple disciplines or emerge in fields where extensive secondary sources are scarce. Elevating SCIRS from an essay to a formal guideline could inadvertently restrict the integration of innovative or complex scientific information, making source verification a restrictive rather than supportive process, like in the example you cite. FailedMusician (talk) 23:34, 15 April 2024 (UTC)

  Other disciplines also need secondary sources to comply with NPOV and OR, so I don't see how SCIRS would affect such content negatively. Can you link some examples of disciplines where secondary sources are scarce but which still have DUE content? The example I cite is evidence in support of SCIRS as it would discourage use of unvalidated, potentially fringe research findings outside of medicine. JoelleJay (talk) 00:47, 16 April 2024 (UTC)

• Oppose. Large parts of SCIRS are copy-pasted from (an old version of) WP:MEDRS, but with some words changed. There may be a place for a SCIRS but it would need to be more specific and content-appropriate than this. Bon courage (talk) 03:08, 16 April 2024 (UTC)
• Could support if retitled. I find SCIRS is very useful. In my experience, when articles or sections are rewritten to use mostly SCIRS sources, they get considerably better. I've thought of proposing that this be retitled to "Identifying 'high-quality reliable sources (science)" and then made a guideline. With its current title, I have two concerns. One is the large grey area around what “science” is, which would need to be clarified. Another is the exclusion of factual encyclopedic content that is too new or too obscure to have been covered in secondary sources. Here’s a simple example from Orca: “A 2024 study supported the elevation of Eastern North American resident and transient orcas as distinct species, O. ater and O. rectipinnus respectively.”

I’m very concerned that a guideline would be used to revert any and all additions of content that “fails SCIRS” which is highly discouraging to newbies and would result in the rejection of a lot of good information along with bad.

The value of SCIRS sources is that they indicate the level of acceptance that claims have in the science community. This is useful for assessing controversial claims and for filtering out noise in fields where there are a lot of early-stage technologies clamoring for attention. Secondary sources are invaluable for ensuring NPOV in broad and/or controversial areas. However, I have never bought into the idea that secondary sources are essential for ensuring NOR in the sciences. A primary source in history is by definition written by a non-historian and requires a researcher to interpret it. A primary source in science is usually written by a scientist and summarizing it is not original research. Clayoquot (talk | contribs) 04:24, 16 April 2024 (UTC)

Yeah, but deciding a piece of primary research is worthy of encyclopedic content (i.e. is 'accepted knowledge') is OR. Primary research is really an interchange among researchers, and much of it is faulty/wrong/fraudulent. Wikipedia editors are in no position to pick and chose what's good and what's not. Bon courage (talk) 04:29, 16 April 2024 (UTC)

Clayoquot makes an important point about the uncertainty over what is the "science" as that can be exploited by advocates of certain issues to misrepresent emerging or part scientific topics as being on the fringe. This can impact the reliability and representation of such topics on Wikipedia, potentially either overstating or undervaluing their scientific validity. Therefore, clear guidelines are crucial to prevent the misuse of these definitions. FailedMusician (talk) 07:36, 16 April 2024 (UTC)

WP:FRINGE doesn't just apply to "science", however it is defined. In the humanities the secondary/primary considerations are completely different in any case (you don't get a systematic review of who wrote the works of Shakespeare!). Bon courage (talk) 07:43, 16 April 2024 (UTC)

How scientific aspects of topics are defined is important, especially in the face of editors engaging in strong advocacy on issues, and worse. That's why we need to exercise caution here. FailedMusician (talk) 09:56, 16 April 2024 (UTC)

FYI, SCIRS already addresses the scope in the lead The scope of this page includes the natural, social and formal sciences.

As for something being too obscure, that would indicate a WP:WEIGHT issue with inclusion. If it's too new, weight issues come into play too. For MEDRS topics, that means waiting to see if sources indicate the results are due to include or not, and that's worked well in practice. We don't have a WP:CRYSTAL ball to decide what new or late-breaking news (i.e., WP:RECENTISM) should be included. Generally our WP:PAG have us being behind the ball on new information like that. WP:NOTJOURNAL policy comes into play here too where an encyclopedia is not where we have essentially recent news on primary research like us scientists are used to writing in real life.

Point is, a lot of things being brought up are things we are supposed to be avoiding in existing policy/guideline. SCIRS is just explaining why (or intended to) and how to navigate that with relative flexibility compared to something like MEDRS. KoA (talk) 21:56, 16 April 2024 (UTC)

• Oppose: Like @Clayoquot, I am an active contributor to WikiProject Climate change, and I can say with confidence that certain scientific subjects, such as, in fact, climate change, are so fast-moving that an application of this policy would cripple most of our articles on this topic. Even the primary peer-reviewed papers are, by necessity, several years behind the real-world processes due to the time it takes to first analyze the climate data, and then to get the paper through peer review. To give an example I have had to deal with recently - a research paper (i.e. a primary source) on trends in oceanic carbon storage published in August 2023 was only able to cover trends up to 2014! Yet, if SCIRS were to become a guideline, we would need to exclude it and rely on even older data!

As @Bon courage points out, much of SCIRS stems off MEDRS. Primary research in climate science is in a very different position to primary medical research. It's one thing to p-hack an observational study among a few dozen patients. It's quite another when you have to reserve months of computing time from room-scale supercomputers (lead image here shows what a typical climate model looks like nowadays, for reference) - often multiple ones in different research institutions across the world - in order to be in a position to even test your hypothesis in the first place. Likewise when your primary research involves field work like sending robotic submarines underneath glaciers.

It is actually a lot easier to write a review in climate science if you don't mind about the journal which would accept it - and the current guideline text has very little to say about differences between journals, even ones as obvious as those between Nature and Science vs. MDPI and Frontiers. As best as I can tell, this notorious piece from MDPI would be accepted for inclusion due to being a secondary source, while a paper in Nature like this one would be rejected, if going off SCIRS as currently written. Even if this were amended, a lot of primary climate research is unlikely to make it into reviews for reasons that have nothing to do with reliability. I.e. it's not really realistic to expect that scientific reviews, or even the ~4000-page IPCC reports (published in 7-year intervals) would include every good paper about climate impact on every species that could be studied, or about every geographic locale. For lesser-known species/areas in particular, it would often be primary research or nothing.

Finally, I can only assume that if this guideline were to be applied consistently, then graphics taken from primary sources would be affected as well, wouldn't they? That would be a disaster for so many of our articles, which would stand to lose dozens of illustrations. This is because only a handful of reliable climate journals use the licensing compatible with Wikipedia terms, and those overwhelmingly publish primary research. Secondary scientific reviews tend to either lack suitable illustrations in the first place, or to have incompatible licensing (i.e. the graphics in the IPCC reports). The precious exceptions are nowhere near enough. InformationToKnowledge (talk) 10:54, 16 April 2024 (UTC)

As someone who works in the climate change research realm IRL, this doesn't seem like a very accurate description of the literature in terms of how SCIRS would work in practice. If a late-breaking primary study is important and worth mentioning on Wikipedia, other papers will cite it, even primary ones, and establish due weight for us as well as give us a summary we could use. Climate change is one of the more well published fields, so that won't really be a problem. Even in my main area of entomology, there's a lot of variation, but even the "slower" fields really wouldn't have issues with this. SCIRS is not the same as MEDRS, though like I mentioned above, the essay would benefit from some tightening up I hope to do soon that would make it redundantly clear. At least the intent behind SCIRS really isn't being reflected in the oppose !votes so far, so there's a disconnect somewhere. KoA (talk) 18:15, 16 April 2024 (UTC)

If the intent behind a policy/guideline/essay isn't being reflected in people's comments about it then it's likely that the intent isn't being well communicated, which is an argument against formalising it. Thryduulf (talk) 19:14, 16 April 2024 (UTC)

That's a part of it I mentioned early on where some reworks are likely needed. The other concern I have of that coin is that people are assuming things about MEDRS and applying that to SCIRS rather than focusing on specific parts of what SCIRS actually does say. There's a point where an ungrounded oppose really isn't even opposition to SCIRS.

I was getting a hint of the latter in ITK's comment where they said Yet, if SCIRS were to become a guideline, we would need to exclude it and rely on even older data! in reference to this primary source. Nothing is mentioned about SCIRS specifically there that's at issue though. Normally you'd want to stay away from the results section of such a paper outside of very limited use, but in the absence of full secondary publications, using the introduction there absolutely would not be a problem in a limited fashion. I also see at least 15 papers citing it if you wanted to include WP:DUE information about the results of that paper. There's a lot of flexibility in terms of what SCIRS says right now for that example, so that's why I'm asking for specifics to see where the disconnect is.

I'm personally more interested in fine-tuning SCIRS than guideline promotion right now, so I'm trying to sort out what concrete concerns there may be (that could potentially be worked on) versus assumption. If there's something specifically in SCIRS that's at issue, this would be the place to iron that out, so I'd ask folks to point out specifics in SCIRS. If it's something someone thinks is in SCIRS but isn't, then I don't know what to say. When I see some comments here that basically amount to saying they wouldn't be allowed to do what SCIRS specifically gives guidance on and allows, I have to wonder if it was something they skimmed over, something that needs to be outlined better, etc. rather than jumping to a more extreme conclusion that someone didn't really read SCIRS. Tl;dr, I'd like to see specifics to work on. KoA (talk) 21:39, 16 April 2024 (UTC)

At least the intent behind SCIRS really isn't being reflected in the oppose !votes so far, so there's a disconnect somewhere. - Well, nobody reasonable opposes the intent to make scientific citations more reliable. That does not indicate agreement with the methods used to get there.

There's a lot of flexibility in terms of what SCIRS says right now for that example, so that's why I'm asking for specifics to see where the disconnect is.

Well, here's an example. I also see at least 15 papers citing it if you wanted to include WP:DUE information about the results of that paper. - Firstly, the paper, which I'll link to again only has 5 citations according to CrossRef, which is directly integrated into the journal page itself. Needless to say, SCIRS most definitely does not say anywhere "You should choose Google Scholar over the papers' own preferred citation tools when it comes to assessing notability."

Secondly, there is absolutely nothing in the current text of SCIRS which suggests that ~15 citations is the magic number which would satisfy the "widely cited" part of In general, scientific information in Wikipedia articles should be based on published, reliable secondary sources, or on widely cited tertiary and primary sources. It is left wholly ambiguous what "widely cited" should mean. Who says it's not 30 citations or 50, or perhaps even more? (Papers on certain subjects in climate science can hit such numbers very quickly - here is a research paper which got to 604 citations in less than two years, for example).

At the same time, I think that even if SCIRS did codify the recommended number of citations + the recommended citation tool, that would not be much of a step forward on its own. You may say that SCIRS would allow that AGU Advances paper, for instance, but you seem to concede my other example: As best as I can tell, this notorious piece from MDPI would be accepted for inclusion due to being a secondary source, while a paper in Nature like this one would be rejected, if going off SCIRS as currently written. I don't think it's good practice to effectively say we know better than the editors of Nature flagship journal and effectively impose a freeze on citing their latest research (which, in this case, operates with decades of data and has very important implications for a range of ecosystems) until an arbitrary number of other publishing researchers end up citing it as well.

I'll give one more example of personal relevance for me. This January, I have put a lot of work into cleaning up and generally expanding the Southern Ocean overturning circulation article. It is still far from perfect, but I hope nobody will object to the idea that it is now MUCH better than what it used to be. Yet, the research on ocean circulation is overwhelmingly focused on the Northern Hemisphere, particularly on the AMOC (also rewritten by yours truly the other week, for that matter.) There has been a drought of research on this southern counterpart to AMOC until very, very recently (literally the last couple of years), and the research which is now coming out still has a (relative) difficulty getting cited, because again, AMOC is a much "sexier" research topic. I have a concern that a not-inconsiderable number of papers I used for that article would be considered "insufficiently cited" if the current text of SCIRS were elevated to guideline status, and I really struggle to see how excluding them, even temporarily (but potentially for years) would improve that article.

If I were to name one modification to SCIRS I would want to see the most, it would be de-emphasizing the number of citations of individual papers and emphasizing CiteScore/Impact Factor of the journals which published them. For the flagship journals with absolute highest Impact factors, I don't think any number of citations should be demanded. In contrast, the number of required citations would scale as the impact factor/journal reliability decreases: I might well oppose citing anything from MDPI/Frontiers that has not hit ~50 citations in general and/or a citation in something like an IPCC report or a flagship journal publication. InformationToKnowledge (talk) 23:12, 16 April 2024 (UTC)

I interpreted I also see at least 15 papers citing it if you wanted to include WP:DUE information about the results of that paper as a suggestion to use those citations' descriptions of the paper's results, as those would be secondary. JoelleJay (talk) 16:14, 17 April 2024 (UTC)

Correct, the whole point was that there were plenty of citing papers to potentially use as a secondary sourced content about that example study. Following citations is almost exactly what we do in MEDRS topics too looking for papers that cite a primary research article if someone initially wanted to try to add it, except there we're limited to full secondary sources like lit. reviews or meta-analyses. For a SCIRS topic though, it's much more relaxed. This is illustrating some of my caution I just mentioned about reading what's actually in SCIRS because I'm not aware of anything about citation counts, and that's rightfully left out of both this and MEDRS. As far as I'm aware, only WP:NPROF does anything like that, which isn't without controversy. KoA (talk) 21:34, 17 April 2024 (UTC)

This is illustrating some of my caution I just mentioned about reading what's actually in SCIRS - Have you considered that maybe SCIRS is just poorly written? Reams and reams of passive-voice text, often chock-full of equivocations and qualifiers, and frequently packed into 8-12 line paragraphs that make it hard to pick out the important from the self-explanatory at a glance. SCIRS makes the actual literature reviews look positively exciting and easy-to-read.

Correct, the whole point was that there were plenty of citing papers to potentially use as a secondary sourced content about that example study. Following citations is almost exactly what we do in MEDRS topics too looking for papers that cite a primary research article if someone initially wanted to try to add it, except there we're limited to full secondary sources like lit. reviews or meta-analyses.

So...how do you functionally tell apart an otherwise primary source being used a secondary source for another primary source... from simply a primary source being used as a primary source, when you are an editor reviewing another's edit? This idea seems to fall apart if you think about it for a minute.

In fact, after taking a second look...

what we do in MEDRS topics too looking for papers that cite a primary research article if someone initially wanted to try to add it

So, if applied wiki-wide, that would seem to translate to baby-sitting every single attempt to add a primary paper reference and demand either proof it's an indirect citation for something else or a citation hunt to cite that paper indirectly? All while we still have enormous issues with both unreferenced passages and those relying on deeply obsolete references? That would seem to be incredibly counterproductive.

I'm going to concur with a quote from Peter Gulutzan up above: WP:MEDRS was a bad idea too but at least had the excuse that dispensing bad health advice could cause legal problems I don't see the justification for this additional, stifling layer of Wikibureaucracy where that risk does not exist, and where there is a much greater chance of important context being lost in translation from a full study to a citing sentence. InformationToKnowledge (talk) 04:10, 18 April 2024 (UTC)

So...how do you functionally tell apart an otherwise primary source being used a secondary source for another primary source... Well, you go look at the source of course. That's normally what most of us editors do when we're reviewing any article. If someone isn't checking sources when they make a citation or are verifying content, that's a problem in terms of existing WP:PAG, which is what we based WP:CONSENSUS on. KoA (talk) 04:38, 23 April 2024 (UTC)

@InformationToKnowledge, I appreciate you looking at the practical side here. You might be interested in readhing WP:PRIMARYINPART.

In the MEDRS context, it's usually pretty easy to tell when a primary source (e.g., a journal article whose primary purpose is to report the results of a randomized controlled trial) is being used as a secondary source. The first thing to look for is whether the content comes from an "introduction" or "background" section. Those sections take information from previously published sources, and are selected and combined to present a new(ish) way of looking at the information. So that part of the paper is usually secondary, and the thing for editors to remember is that "Secondary" does not mean "good". Even though there are secondary, they can be somewhat biased (the authors present only the background information that is relevant to or supportive of their specific research, rather than trying to write an unbiased and comprehensive overview – for example, the surgeons only talk about surgery, the drug companies only talk about drugs, etc.). WhatamIdoing (talk) 17:51, 23 April 2024 (UTC)

I interpreted "I also see at least 15 papers citing it if you wanted to include WP:DUE information about the results of that paper" as a suggestion to use those citations' descriptions of the paper's results, as those would be secondary.

I have already written my objections to this idea in the other reply, but I decided I might as well humour this suggestion and see where it takes us. I'll begin with the 5 citations you actually see on the journal page itself, since people will almost certainly do that first, and pull up Google Scholar second (if ever).

Citation 1: Shares some of the same authors - according to the current SCIRS text, that seems to be allowed? (Unless I missed a line tucked away within one of those huge paragraphs which accounts for that.) If it is, that kinda makes one wonder what the purpose of this whole rigmarole is, if the researchers citing their previous work somehow immediately makes it more reliable. Anyway, it cites the study in question (Müller et al., 2023) four times:

Currently, the global ocean takes up 25%–30% of all human-made CO2 emissions (DeVries, 2014; Friedlingstein et al., 2022; Gruber, Clement, et al., 2019; Gruber et al., 2023; Khatiwala et al., 2009; Müller et al., 2023; Terhaar et al., 2022)

Although they contain data from similar GOBMs and pCO2 products, the compiled database of RECCAP2 (Müller, 2023) goes well beyond that used in the framework of the Global Carbon Budget (Friedlingstein et al., 2022).

The RECCAP2 database (Müller, 2023) provides model output from 1980 to 2018 from four simulations (called simulations A, B, C and D) that aim to quantify the different components of the oceanic CO2 flux. (A bunch of equations follows.)

To compare the net sea-air CO2 fluxes from the GOBMs with observation-based estimates, we utilize the RECCAP2 data set of pCO2 products (Müller, 2023), including AOML_EXTRAT, CMEMS-LSCE-FFNN, CSIR-ML6, JenaMLS, JMA-MLR, MPI-SOMFFN, OceanSODA-ETHZ, UOEX_Wat20, and NIES-MLR3 (see Supplementary Table 2 in DeVries et al. (2023) for references and further details).

Citation 2

Multiple lines of observation-based evidence support climate-change effects on the ocean carbon sink (Keppler et al., 2023; Mignot et al., 2022; Müller et al., 2023)

It is unambiguous that the ocean carbon sink has increased over recent decades in line with increasing atmospheric CO2 levels as its primary driver (Ballantyne et al., 2012; DeVries et al., 2023; Gruber et al., 2023; Müller et al., 2023).

Citation 3

This finding agrees with previous studies that find an important role for Pinatubo in preindustrial carbon variability (Eddebbar et al., 2019; Fay et al., 2023; McKinley et al., 2020), and gives us additional confidence that observation-based estimates of changing anthropogenic carbon distribution (e.g., Gruber et al., 2019; Müller et al., 2023); (also Müller, Jens Daniel, Gruber, Nicolas, Carter, Brendan R., Feely et al., Decadal Trends in the Oceanic Storage of Anthropogenic Carbon from 1994 to 2014, in preparation for Authorea) are relatively unaffected by the Pinatubo climate perturbation.

Citation 4 (also involves some of the same authors as the original study)

How is the ocean carbon cycle changing as a consequence of sustained increases in emissions of carbon to the atmosphere? Important steps toward answering this question over the last several decades have been provided via estimates of ocean carbon uptake from both interior hydrographic measurements (Gruber et al., 2019; Müller et al., 2023; Sabine et al., 2004),...

For all of the oceanic studies within RECCAP2, a discrete number of ocean biomes based on Fay and McKinley (2014) are used to facilitate consistent intercomparison between regions (described in the supplement to the Müller (2023) publication of the RECCAP2 data).

Thus, our six aggregated biomes (Table 1 and Figure 1) (their precise boundaries given in Supporting Information S1 of the RECCAP2 data release of Müller, 2023) consist of

Citation 5 (also involves some of the same authors as the original study)

A recent update of the eMLR(C*) results by Müller et al. (2023) resolves decadal trends in the anthropogenic carbon accumulation from 1994 to 2014, but was published after the completion of this study and could thus not be considered here.

Nevertheless, a recent update of the eMLR-C* estimates by Müller et al. (2023) also suggests substantial climate-driven variability in the oceanic storage of anthropogenic carbon similar to that shown in Figure 7f.

None of these citations mention the finding of the original study where carbon storage in the North Atlantic specifically had declined by 20% (at most, you can kinda sorta see a decline in Figure 1 of Citation 4, but you can't really get the specific percentage from there), which is the whole reason why I cited that study in the first place. For that matter, the additional citations from Google Scholar (some of which are either preprints or paywalled) do not do that either. So, if SCIRS were to become a guideline, that specific finding, which, lest we forget, was derived from

the JGOFS/WOCE global CO2 survey conducted in the 1980s and 1990s (Key et al., 2004; Wallace, 1995), the repeat hydrography program GO-SHIP that began in 2003 and is now completing its second cycle (Sloyan et al., 2019; Talley et al., 2016), as well as a number of additional programs, including INDIGO, SAVE, TTO, JOIS, and GEOSECS (Key et al., 2004, and references therein).

Would somehow become too unreliable for inclusion in a Wikipedia article, purely because no other article felt the need mention this particular detail yet, even as they cited the study itself? REALLY?

So, I once again don't understand what this is meant to achieve. If poorly reviewed papers attempting to overturn academic consensus are supposedly the problem, then an Impact Factor bar set at the right level would efficiently block basically all of them without forcing this onerous rigmarole whenever attempting to cite valuable research findings. InformationToKnowledge (talk) 05:16, 18 April 2024 (UTC)

The idea that prestigious journals are more likely to be correct is debatable. See The Economist which explains that there's a winner's curse effect. Prestigious journals like Nature have the most choice and and may publish the papers which are most sensational rather than those which are most accurate. Andrew🐉(talk) 07:57, 19 April 2024 (UTC)

If no academic sources are discussing that particular finding, then neither should Wikipedia. If it's not relevant enough to be mentioned in other studies then it's certainly not at the level of accepted knowledge we need in an encyclopedia. JoelleJay (talk) 08:22, 19 April 2024 (UTC)

I'll echo that too. If someone is opposed to a potential guideline because it won't let them add something that all signs are pointing to not currently being WP:DUE, that's not a good reason to oppose a guideline.

I'm seeing a lot of potential introductions to pull from in general in that little exercise above though. KoA (talk) 04:58, 23 April 2024 (UTC)

The only reason that finding wasn't "due" in those citations is because all those citations to date were focused on the entire World Ocean, so citing certain text about a specific ocean region certainly wasn't relevant in the context of their research - as opposed to our wiki pages on the North Atlantic region or the Atlantic meridional overturning circulation specifically.

The idea that how often a certain paper is cited in general (let alone the general reputation of its publisher and/or research team) does not matter, and specific findings only "become" reliable once another paper happens to have enough overlap with a certain topic to cite not just the paper as a whole, but that specific phrase, is unintuitive and counterproductive and is likely to remain so.

I am still not seeing a good reason for why a combination of (independent) citation count and journal metrics like Impact factor would not be a better alternative for assessing WP:DUE than this proposal. The only counterargument I have seen so far is "big journals make mistakes too" - which is easily countered by how often papers, particularly at bad journals, can be found to mangle their citations, saying something subtly yet significantly different from the original text. At least, when the major journals screw up, it is reasonably likely that there would be pushback in WP:RS, which we can then mention to place the work in context. In fact, it is many times more likely than to see criticism of bad referencing post-publication, so I remain unconvinced this suggestion adds, rather than removes, reliability. InformationToKnowledge (talk) 14:00, 23 April 2024 (UTC)

How would your proposed citation+prestige system work for a paper like the MMR fraud perpetrated by Andrew Wakefield? I'd think it would rate quite highly, even though nearly all of the sources citing it have done so to say that it's a disgraceful fraud. WhatamIdoing (talk) 17:56, 23 April 2024 (UTC)

InformationToKnowledge raises crucial points about the practical implications of applying SCIRS as a guideline to rapidly evolving fields like climate science, especially for non-controversial facts. The concern about excluding valuable primary research due to the proposed guidelines' stringent requirements is well-founded, especially when considering the time lag in publishing comprehensive secondary sources in such dynamic areas. This emphasizes the need for SCIRS to accommodate the unique challenges of different scientific disciplines, ensuring that Wikipedia remains an up-to-date and reliable resource without unnecessarily excluding relevant and recent research findings, observations and commentary. FailedMusician (talk) 01:35, 23 April 2024 (UTC)

I think we're looking at two different problems:

• Primary vs secondary sources: Secondary is not another way to spell "good", and primary is not a fancy way to spell "bad". A source can be primary and be a good source. Whether it's a good source depends partly on the source itself (e.g., is it self-published?), but it also depends significantly on the WP:RSCONTEXT. For example, editors will probably want a secondary source for a statement like "Wonderpam cures _____", but a primary source might be accepted for a statement like "Wonderpam was the first drug in its class" or "Wonderpam has a shorter shelf life than other treatments".
• Strong vs weak sources: Sources need to be strong enough to bear the weight of the claim they're being cited for. Wikipedia:Extraordinary claims require extraordinary evidence. Lightweight claims don't. If a claim is truly non-controversial, then we don't really need a strong source at the end of the sentence. For example, MEDRS accepts WebMD for non-controversial content. It's a secondary source, but it's not a strong source.

The more controversial the claim, the better the source(s) we should be citing. WhatamIdoing (talk) 18:15, 23 April 2024 (UTC)

• Oppose per WP:CREEP. There's no simple algorithm for determining The Truth and complex advice tends to be so equivocal that it is no help and just results in endless Wikilawyering. Andrew🐉(talk) 08:04, 18 April 2024 (UTC)
• I feel like the various subject-specific RS essays are more in-line with supplement, but I'm not sure it would make much of a difference either way. Alpha3031 (t • c) 13:55, 18 April 2024 (UTC)
• Oppose; people have given a pretty broad range of rationales for opposing this, so not sure that I can contribute a lot. But one thing I will note is that the most recent extremely-high-profile back office brouhaha we had about WP:MEDRS and WP:BIOMED (to wit the giant years-long covid slapfight) did not convince me that having a bunch of additional rules for what sourcing guidelines to use and when to apply them would make it easier to deal with conflict. jp×g🗯️ 06:45, 19 April 2024 (UTC)
• Regardless of which tags end up at the top of the page, I'd like to see the Wikipedia:Identifying reliable sources (science)#Formatting citations section deleted as redundant to Wikipedia:Citing sources. WhatamIdoing (talk) 01:29, 20 April 2024 (UTC)
• Provisionally oppose upgrade, but not use as an essay for now This guideline lacks references to support the claims that it makes. (I accept that WP:V does not necessarily allow me to delete all unreferenced content in the project namespace, but that does not mean that I have to agree to making it a guideline). It tells us to prefer peer-reviewed sources, despite the fact that this is apparently not completely uncontroversial: [10] [11] [12] [13] and all the other sources that come up on a search for "peer review flawed process" and Scholarly peer review#Criticism. It fails to answer the apparent controversy. It fails to consider whether the purpose of peer reviewing is to determine accuracy (which is relevant to reliability) or to determine importance/originality etc (which is not relevant to reliability). (I am under the impression that scientific "proof" consists of being able to reproduce the results of an experiment by repeating it over and over again, and the peer reviewer is presumably not doing this). We are told to use textbooks. I was once told that the average physics textbook is two years out of date the moment it reaches book shops, and that you cannot do physics properly without reading papers. (You'll have to take my word for this for now, as I don't have time to verify it.) James500 (talk) 09:20, 21 April 2024 (UTC)

  I'm among those scientists that think the peer review system is broken and should be thrown out, but that's a red herring here. Peer review is currently the universally-accepted quality control mechanism in academia. There is debate other whether it should continue to be so, but until then tertiary sources like Wikipedia have to rely on peer reviewed literature, because there is simply no alternative.

  With textbooks, Wikipedia is supposed to be at least two years out of date, because our goal is to document and explain major points of view. New research does not become a "major point of view" in science in the first few years after it is published, because the scientific community needs time to assess the arguments and the evidence. In other words, it is impossible to summarise cutting-edge research without falling afoul of WP:SYNTH. We can give readers a summary of accepted knowledge; they should go elsewhere to learn about current debates or the state of the art. – Joe (talk) 09:30, 21 April 2024 (UTC)

  (Answering both James and Joe) Though there are many problems with peer-review, it is correct to treat it as the best process we have because it really is the best process we have. If some better system takes over, we'll adopt that as the gold standard instead. Peer review is about both accuracy and importance, but the degree to which it can address accuracy depends on the field of study. It's true that reviewers are not expected to repeat experiments, but they are supposed to look for flaws in the experimental protocol, check that competing theories are adequately cited, and they are supposed to insist that enough details are given that others can repeat the experiment. In theoretical fields like mine, reviewers are supposed to check calculations, but they are not expected to repeat computer calculations. I strongly disagree that we can't describe recent publications without violating SYNTH. All that's needed is to present the results as a theory proposed by a named person, without making our own judgement of its validity. The only real problem is deciding which of the thousands of new publications to cover, but that's a NOTABILITY problem not a SYNTH problem. I'm sure that many readers visit Wikipedia looking for accessible explanations of new scientific ideas that they saw in the press or somewhere and I believe that one of our roles is to provide them. Incidentally, lots of important science never appears in textbooks and research monographs about them can take much longer than two years to appear. Zero^talk 14:04, 21 April 2024 (UTC)

  You've said it yourself: the problem is selecting which primary findings should be covered. Call it SYNTH or call it OR or call it notability (though that seems a stretch?) – it's a problem. If we don't retain an emphasis on secondary over primary sources, how do you propose that we identify which new papers are "important science" and which are garbage that somehow sneaked through peer reviewer but will be forgotten about in a year, without engaging in original research? – Joe (talk) 15:15, 21 April 2024 (UTC)

  It's absolutely a problem of BALASP and SYNTH to cover the results of primary papers using those papers as sources. If the wider academic community hasn't contextualized it with the existing mainstream consensus, through reviews or at the very least summaries in the background of other, independent, primary research articles, then it does not belong on wikipedia. JoelleJay (talk) 19:00, 21 April 2024 (UTC)

  I don't know how you can match this situation to the definition of SYNTH. There isn't anything necessarily SYNTH about it. We can judge notability by the usual way we judge notability: by the attention something gets in "reliable sources". Arguments about the notability of particular cases will happen but that's true across all fields, not just in science. I'm not saying that we should pick journal articles that are citation orphans and make articles out of them, but once a journal article makes a "splash" we are free to cite it. The suggestion that Wikipedia must be endlessly years out of date is horrifying and definitely was not the intention when the rules were written. Zero^talk 07:04, 22 April 2024 (UTC)

  This is completely unrelated to notability. The problem with covering splashy new science results is that it is not possible to summarize what the scientific community's assessment of them is in relation to the existing understanding of the topic. We are therefore synthesizing its relevance to the topic when we describe it in an article based only on its primary report.
  And we don't necessarily have to be years out of date, but WP definitely is intended to operate as an encyclopedia summarizing accepted knowledge, not as a EurekAlert stand-in. JoelleJay (talk) 17:33, 22 April 2024 (UTC)

  The way to avoid summarizing "what the scientific community's assessment of them is in relation to the existing understanding of the topic" when we have no reliable sources for that assessment is to not do it. To avoid SYNTH, don't violate SYNTH. You have not given a reason why citing a recent publication is necessarily SYNTH. Zero^talk 01:31, 23 April 2024 (UTC)

  You would be engaging in original research as an anonymous editor. In the real world, you generally need a degree in a semi-related field and familiarity with research statistics to interpret the validity of results in the study. None of us are qualified to do that as anonymous editors, even those of us who do have credentials in specific fields IRL. That's what makes using primary research so different here than if you are writing your own summary for work in a research lab. KoA (talk) 04:47, 23 April 2024 (UTC)

  The same argument could be used to eliminate all articles on technical subjects even if all the sources are secondary. Do you think an average editor can understand a graduate textbook in differential geometry or quantum mechanics? The only reason we have many excellent articles on scientific subjects is that we have editors who can understand the sources. You are actually criticising something that nobody has proposed. The role of an editor is to summarise what the researchers (and experts other than the original authors) say about it, not to "interpret the validity" of it. Zero^talk 09:40, 23 April 2024 (UTC)

  The role of an editor is to summarise what the researchers (and experts other than the original authors) say about it, not to "interpret the validity" of it. Exactly. IRL, you are assessing the validity of statements in the results section if you are summarizing them in any way or saying they are worth mentioning. We as anonymous editors don't get such special privileges, so that's why we rely on secondary sources who are qualified to do that for us.

  If I'm reading a primary article IRL and citing the results, I'm supposed to be checking if their methods actually let them say that, the statistical tests are valid, etc. That gets taught pretty early on in introductory college level courses, and especially on how scientific literature is misused when people don't do that. That reality remains regardless of guideline or not. KoA (talk) 16:34, 23 April 2024 (UTC)

  KoA, I think you have the right intuition here, but it's neither OR nor SYNTH.

  First of all, it is actually impossible to violate SYNTH when you are looking at a single source. SYNTH begins with the words Do not combine material from multiple sources. One source is not multiple sources; ergo, SYNTH does not apply.

  Second, deciding that some material is worth mentioning is not an example of material—such as facts, allegations, and ideas—for which no reliable, published source exists – which is our definition of OR.

  I find that understanding the reason that WP:NOR was created helps people understand it. That policy exists because, 'way back in Wikipedia's earliest days, a Usenet personality (read: physics crackpot) thought that Wikipedia would be an excellent place to tell the world about his proof that Einstein was completely wrong. He couldn't get the scientific journals to publish his nonsense, and he got laughed at on Usenet, but he was just so convinced that he had figured out something that nobody else knew, that he really wanted to tell the world. Wikipedia was one of his targets. We didn't accept his nonsense, either, and we wrote NOR to draw a line in the sand, and say to all the other crackpots in the world: if you can't get your idea published in the real world, we don't want it here, either.

  The flipside, which has probably occurred to you, is that if you did get your idea published in the real world (e.g., as a primary source in a scientific journal), then we might want it here. But what's important for this discussion is: If the material in question was actually published in a reliable source, then it's not NOR. It might be a violation of every single other policy and guideline, but it's not NOR.

  I think what you're looking for is Wikipedia:Relevance, or, in the more general case, NPOV. Deciding whether the contents of a source is worth mentioning is fundamentally not about an editor making stuff up, but about an editor finding the right WP:BALANCE in the article. WhatamIdoing (talk) 18:31, 23 April 2024 (UTC)

  No. You're not generally looking at a single source when writing an article, you are looking at multiple sources, and indeed you can imply something about them in the ways you put them together. Alanscottwalker (talk) 18:43, 23 April 2024 (UTC)

  But incorporating material from the primary source into the context of the rest of the article is synthesizing from multiple sources... JoelleJay (talk) 18:53, 23 April 2024 (UTC)

  @Alanscottwalker and @JoelleJay, that isn't what the policy says.

  WP:SYNTH does not restrict itself to primary sources. If you combine any sources to reach or imply a conclusion that does not appear in any source, then you violate SYNTH. Combining two high-quality secondary sources, if you combine them in ways that reach or imply a conclusion that has never been made in a reliable source, is a SYNTH violation.

  For example, this is a N classic SYNTH violation:

  • String theory is correct.^{[excellent source]}
  • Newtonian physics is correct within limits.^{[great source]}
  • Therefore, I say Einstein is wrong!^{[Wikipedia editor's own conclusion]}

  Using two sources next to each other – so long as you are not reaching or implying a conclusion that has never been published in a reliable source – is not a SYNTH violation.

  For example, this pairs two primary sources, and it is Y 100% non-SYNTH and acceptable per policy:

  • Smith stated that Jones committed plagiarism by copying references from another author's book.^{[op-ed in a magazine]}
  • Jones responded that it is acceptable scholarly practice to use other people's books to find new references.^{[Jones' blog]}

  Alan, you're correct insofar as we (and the policy) agree that you can imply something that isn't present in any source, but there is nothing inherent about using a primary source, or using multiple sources in the same article, that means you actually are reaching or implying a previously unpublished conclusion. If you haven't combined multiple sources to create a new conclusion, it's not SYNTH; if everything in the article comes from sources, then it's not any type of OR. WhatamIdoing (talk) 23:33, 23 April 2024 (UTC)

  Joelle, I don't know if the implications of your comment were clear to you – maybe it doesn't say quite what you meant – but if it were actually true that "incorporating material from the primary source into the context of the rest of the article is synthesizing", then WP:PRIMARY would be much shorter, since all it would need to say is "Citing primary sources is banned". Either it's possible to cite a primary source in an article without violating SYNTH, or primary sources are banned by SYNTH. This is a strictly either-or situation; we cannot have it both ways, so that we claim out of one side of our mouths that primary sources are permitted and out of the other that using them is a violation of SYNTH because using them (correctly) is synthesizing their contents into the context of the rest of the article.

  Given that the word primary doesn't appear anywhere in SYNTH, and given that editors cite primary sources every hour of the day, including in Featured Articles, I think it's clear that primary sources are permitted (when used appropriately) and do not violate SYNTH (except when used in ways that would equally violate SYNTH if they were secondary sources). WhatamIdoing (talk) 23:41, 23 April 2024 (UTC)

  What's your point? No one said SYNTH is restricted to primary sources. You were just plain wrong that we are only dealing with one source in almost every case that matters, and yes the implication from the way you combine them is what needs to be looked into, that is SYNTH analysis. -- Alanscottwalker (talk) 23:51, 23 April 2024 (UTC)

  JoelleJay said incorporating material from the primary source into the context of the rest of the article is synthesizing from multiple sources. Note the absence of any restrictive clause like "if that's done to reach or imply a conclusion that is not present in a reliable source". A plain reading of her sentence indicates that she believes using a primary source is a SYNTH violation.

  Do you agree with her that citing a primary source in an article always involves synthesizing it with the other sources in the article? WhatamIdoing (talk) 00:30, 24 April 2024 (UTC)

  It always depends on how you use the source. I understood her point to be you are using more than one source (Though one would surmise, one would need to have a very rigorous explanation for throwing primary info into the middle of secondary analysis, of course). Alanscottwalker (talk) 01:00, 24 April 2024 (UTC)

  Yes, my point was that contextualizing a primary source with the rest of an article is combining multiple sources to reach a conclusion not implied in any of them. "Contextualizing" being a key synonym of "interpreting" or "analyzing" here. JoelleJay (talk) 02:50, 24 April 2024 (UTC)

  All sources can (and must) be cited without "interpreting" or "analyzing" them. You seem to think that merely citing a source involves interpreting or analyzing it, but that is not true. See SYNTH is not mere juxtaposition and other items on that page. Here is a hypothetical example of a properly cited source: "XYZ has proposed that black holes evaporate more quickly than previously assumed.[cite]" If the source satisfies RS, this ticks all the boxes and does not involve any interpretation or analysis, nor does it imply that XYZ is correct. It is mere reporting of what is in a source and there is nothing whatever wrong with it. Zero^talk 03:17, 24 April 2024 (UTC)

  You don't judge SYNTH on one source alone. It remains, primary sources are not interpretation/analysis so therefore you can't use them to recast, remix, redo, update, shade, shape, bolster, critique, bring new contextualization, make new implications, etc., for secondary analysis. Alanscottwalker (talk) 06:34, 24 April 2024 (UTC)

  The biggest problem that I have here is that this essay is a TLDR wall of uncited text. Every time I read it, I find new issues. For example, the section "definitions" contains a link to the article Primary source, which says that it is about "the study of history as an academic discipline". (While the article does contain a one line mention of "scientific literature", it is referenced to a source that is about "research" generally, rather than science). The link to the article is clearly not relevant to the essay and ought to be removed. Another example: The essay tells me to use "reviews published in the last five years or so". Why five years? Is this just a round number? Where has this number come from? Who says five years is up to date? Has this essay been systematically checked for errors? It might be better to start a new proposal from scratch, and build it up one line at a time, carefully checking (and preferably citing) the claims as you going along. James500 (talk) 22:41, 22 April 2024 (UTC)

  the article Primary source, which says that it is about "the study of history as an academic discipline" – Um, no, it doesn't. It says that "in the study of history as an academic discipline", a primary sources is a particular thing. It does not say that the article is about the study of history. (Compare "In the field of medicine, cancer is a disease, but in the field of astronomy, cancer is a constellation".) The link is there to help people who don't know what that jargon means. Reasonable people could disagree over whether it is more useful to link to the encyclopedia article, the policy, or the explanatory essay, but I don't think anyone believes it's best to leave unfamiliar terms undefined.

  For your other questions:

  • Why five years? Is this just a round number? – Three to five years is recommended to medical students based on the length of time it takes for sources to get published in that field. This is based on the idea of a "cycle": You publish your research, I publish my review of your research, and someone else publishes a response to my review. You want the whole cycle to happen. Because it takes weeks or months to write the papers, and months (sometimes, even longer than a year) to get the paper published, it usually takes at least one year, and it often takes three to five years, to get an understanding of how the scientific community has reacted to a paper.
  • Where has this number come from? – Straight out of WP:MEDRS.
  • Who says five years is up to date? – Medical researchers, but as a Rule of thumb, not as an absolute statement that applies in all circumstances. Some information (e.g., names of diseases) rarely changes, and other information changes rapidly.

  WhatamIdoing (talk) 23:55, 22 April 2024 (UTC)

  I don't think it's possible to uncritically apply the norms of the medical research cycle to all science - different disciplines move at different speeds and have different considerations and conventions. Thryduulf (talk) 00:12, 23 April 2024 (UTC)

  And that is why I think citations would be desirable: to avoid the risk of a Wikiality definition of reliability. James500 (talk) 00:22, 23 April 2024 (UTC)

  Citations are typically used in mainspace. Wikispace generally doesn't have citations outside of very specific needs. An essay/guideline not having much for citations is the norm. KoA (talk) 04:53, 23 April 2024 (UTC)

  The rule is at WP:NOTPART.

  Thryduulf, I agree with you. History isn't science, but it makes a good example: their fundamental unit of scholarly output is the book, and the cycle is consequently much longer. I don't expect the hard sciences to be wildly different (anything in the last five years is likely to be reasonably current under normal circumstances in any hard science, no matter how fast it moves, and under abnormal circumstances, sudden shifts can happen overnight even in medicine). I am more concerned about subfields that move more slowly. Sometimes niche information is relevant and appropriate, and the best source is six or ten years old, rather than two or five. WhatamIdoing (talk) 00:24, 24 April 2024 (UTC)

  Indeed, and then you get into fields that don't fit completely into a single box, like the history of science, where you might need to cite decades old research, such as when a mainstream theory is proven incorrect conclusively and repeatedly and so nobody touches it again. Luminiferous aether is the first thing that comes to mind (although probably not the best example as that's been the subject of much ley coverage). Thryduulf (talk) 00:58, 24 April 2024 (UTC)

• I think it's obvious that there isn't consensus to upgrade SCIRS right now, but I'm also not hearing a hard no forever and there's been a lot of potential points of improvement raised. I'll try to summarise those at WT:SCIRS when I get a chance – but if anyone can beat me to it, please be my guest. The trickiest issue seems to disagreement over the desirability of applying WP:PSTS to scientific topics, but since that's already a policy I don't see much room for manouvre. – Joe (talk) 06:40, 23 April 2024 (UTC)

  On the contrary, I do not see any disagreement over WP:PSTS. After all, this is its current first paragraph.

  Wikipedia articles should be based on reliable, published secondary sources, and to a lesser extent, on tertiary sources and primary sources. Secondary or tertiary sources are needed to establish the topic's notability and avoid novel interpretations of primary sources. All analyses and interpretive or synthetic claims about primary sources must be referenced to a secondary or tertiary source and must not be an original analysis of the primary-source material by Wikipedia editors.

  Appropriate sourcing can be a complicated issue, and these are general rules. Deciding whether primary, secondary, or tertiary sources are appropriate in any given instance is a matter of good editorial judgment and common sense, and should be discussed on article talk pages. A source may be considered primary for one statement but secondary for a different one. Even a given source can contain both primary and secondary source material for one particular statement. For the purposes of this policy, primary, secondary and tertiary sources are defined as follows

  A primary source may be used on Wikipedia only to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge.

  So, perhaps it is the pro-SCIRS editors here who need to be reminded of the actual PSTS text. They are the ones who are suddenly turning to a lesser extent, on tertiary sources and primary sources into "secondary and tertiary sources only" and arguing that no, primary sources should not be used for straightforward statements of facts, instead proposing an alternative which would often run counter to common sense (I am yet to a see a response to the fairly obvious downsides I identified in an earlier comment here).

  I also want to highlight that this would be a very disruptive change if adopted and there were actually serious attempts to enforce it. To give a personal example: so far, I have successfully nominated a total of three articles for GA. In each case, the article was what I (and apparently, the reviewer) considered to be a healthy mix of primary and secondary sources. Further, each reviewer was a veteran editor with ~67k, ~267k and ~22k edits, and two of them have made extensive contributions both to creating and reviewing GA-class articles. If the people responsible for much of the GA article creation and maintenance are acting counter to the spirit of the policy you propose, you may want to reconsider something. InformationToKnowledge (talk) 13:17, 23 April 2024 (UTC)

  I really thought SCIRS was a low-profile and uncontroversial essay before this – I'm surprised there is already a camp of "pro-SCIRS editors" who need to be reminded of anything! WP:PSTS does indeed say that primary sources can be used, as does WP:SCIRS, under #Respect primary sources: a primary source, such as a report of a pivotal experiment cited as evidence for a hypothesis, may be a valuable component of an article. A good article may appropriately cite primary, secondary, and tertiary sources. When I observed that there is disagreement over PSTS, it is precisely because the rather moderate attempt to apply it in SCIRS (as opposed to say WP:MEDRS, a guideline, which says Avoid primary sources) has provoked such strong reactions. – Joe (talk) 13:43, 23 April 2024 (UTC)

  Well, only a few posts up, multiple editors are telling me that what may appropriately cite in this "Respect primary sources" wording really means is, apparently, "A primary scientific source can only be cited when it cites something else, and never for its own findings." This really is not the way many of us have thought of WP:PSTS before, so I question the idea that this is "moderate". InformationToKnowledge (talk) 14:09, 23 April 2024 (UTC)

  But the purpose of this discussion was to gauge support for upgrading SCIRS to a guideline, not their opinions to a guideline. – Joe (talk) 17:58, 23 April 2024 (UTC)

  At least for most of us who work on science topics, it generally isn't anything controversial in practice in my experience.

  To be blunt though, this has highlighted how many who would benefit from additional guidance of scientific sources are often opposed to it, so there's a catch-22 there on the wiki-process side of things. Some arguments that have come up here are just plain misconception or just making something simple we normally do when dealing with primary sources seem really complicated somehow. I mentioned earlier too how it's not an uncommon problem for people with a science background to have trouble adjusting to working as an anonymous editor when it comes to using scientific literature, so there are a few systemic things to address.

  That said, SCIRS in concept is fairly well primed to be a guideline, but there is some work to be done on structure, broadening concepts that were addressed in the narrow MEDRS sense, etc. I didn't get around to it yet, but I have a few edits I've been working on putting in that I'll get to soon. KoA (talk) 16:48, 23 April 2024 (UTC)

  I'm with you there. Reading this discussion, I can certainly see where people are coming from with the fear of too harsh a prohibition, but I cannot really imagine my own usual topic area functioning at all if primary and secondary sources were given complete parity. – Joe (talk) 18:00, 23 April 2024 (UTC)

  I cannot really imagine my own usual topic area functioning at all... One thing that seems clear to me but not to everyone in this discussion is that "science" isn't a single topic area. Medicine is different to climate change, both are different to archaeology, and all of them are different to astronomy. They have commonalities, but there are such fundamental differences in the nature of the research, the speed of the field, the conventions, etc. that I don't think it's going to be possible to produce a single guideline that both covers every scientific discipline and has anything useful to say that more general policies and guidelines don't already. MEDRS works because it's focused on a single topic area, but at least some of it's provisions just don't translate to other sciences. Thryduulf (talk) 20:27, 23 April 2024 (UTC)

  SCIRS would be applicable to any field that has primary research papers and secondary reviews/meta-analyses/books. They all would benefit from stronger guidance discouraging inclusion of primary results, as suggested in PSTS, which itself apparently needs to be enforced better in certain topics if editors are routinely citing splashy new papers. JoelleJay (talk) 21:44, 23 April 2024 (UTC)

  I agree and I think that Joe Roe's initial proposal to upgrade SCIRS to a guideline was a commendable effort, but there is a big undercurrent of opposition as it would be subject to abuse, and the project has suffered from that in the past. It would be wielded as a gatekeeping tool by editors experienced in wikilawyering, keeping out new ones with valuable contributions, stagnating the project. The key to Wikipedia's success is open collaboration, as written in the lead sentence of the Wikipedia article. FailedMusician (talk) 01:16, 24 April 2024 (UTC)

  and the project has suffered from that in the past. Citations? JoelleJay (talk) 02:52, 24 April 2024 (UTC)

Support, though I'd be in favor of some tweaks/changes, eg I think it's too long, and should also say more about sources being a mix of primary and secondary (eg a novel study might be a good source for current state-of-the-science background). But the core of it, identifying the difference between primary and secondary in science, would be useful to have as a guideline, particularly to prevent against the misuse or overuse of primary sources, basically the same thing that medrs does for medicine. Levivich (talk) 02:06, 24 April 2024 (UTC)

The analogy between MEDRS and SCIRS has problems, mostly to do with why they exist. MEDRS was invented because we knew that people would read WP for medical advice and we don't want to kill anyone. This gives a very strong motive for strict rules on medicine that does not apply to general science articles. I don't see a strong reason why science is different from, say, history. Zero^talk 02:59, 24 April 2024 (UTC)

These rules aren't very strict IMO and apply to all fields, which to say, lots of fields could use a ___RS. MEDRS to explain different levels of reliability in medical sources, SCIRS for general science sources, I think there should be a LAWRS for legal sources (that would explain about citing court opinions v. treatises, etc.), and yes, a HISTRS for history sources (that would explain about citing historians for history v. non-historian scholars v. news media, for example). (Btw I never bought into the whole people-look-at-WP-for-medical-advice-and-might-die argument.) Levivich (talk) 04:48, 24 April 2024 (UTC)

I don't have a memory any more, but my recollection is that both MEDRS and BLP were more or less forced on us by the higher-ups as they were afraid of being sued. In the early days Wikipedia was not filthy rich like it is now. Zero^talk 04:54, 24 April 2024 (UTC)

Unpopular opinion: at least the money has bought better higher-ups; I think the current management team is the best one so far. Levivich (talk) 05:04, 24 April 2024 (UTC)

Oppose - I think that even MEDRS is too restricting, although I understand that in that case we need to be concerned about people taking medical advice from Wikipedia (despite Wikipedia:Medical disclaimer, which hopefully prevents any liability but certainly won't stop most people). In the case of science, this concern is irrelevant. Animal lover |666| 07:09, 24 April 2024 (UTC)

References

Making COI policy

Please see the discussion I started at WT:COI#Should we upgrade this to policy? RoySmith (talk) 16:29, 14 April 2024 (UTC)

Servant of God and Saint With Respect to WP:N.

Quick question here. Does being Servant of God and Saint count as notable, and if so, should we either add this to outcomes or notability? I can think of roughly three options here. 1. No effect on notability, in addition, consider any coverage of their canonization as WP:1E coverage. 2. No effect on notability, but consider coverage of their canonization as establishing notability. 3. Presumed notable. Allan Nonymous (talk) 20:35, 23 April 2024 (UTC)

Just clarifying, I assume that you mean it by the Catholic Church use of those terms. North8000 (talk) 20:48, 23 April 2024 (UTC)

Nothing the Catholic Church (or any other religious institution - it would be deeply inequitable to distinguish between such institutions and treat them as more of an authority than others in this regard) does directly determines notability, which for this, like anything else, is assessed through the depth of coverage in reliable sources. Our decision, not theirs. AndyTheGrump (talk) 21:24, 23 April 2024 (UTC)

The Catholic Church designating someone a Servant of God or Saint (or other religious institutions doing their equivalent) is a claim of significance for A7 purposes, but it doesn't automatically confer notability. It is one point to consider when determining notability, but do note that the Catholic Church is a reliable but not independent source for Catholic saints. Thryduulf (talk) 21:43, 23 April 2024 (UTC)

@Allan Nonymous, the answer is "it depends". Specifically, it depends on how much coverage the individual(s) have received in reliable sources.

So here's something that sometimes surprises people: According to our rules, the Presidents of the United States are not inherently notable. Every single one of them is obviously notable, but none of them are inherently notable. That's because every single article about every single human has to be justifiable individually, based on coverage. If, over the course of several years, journalists and scholars write thousands of words, across dozens of news articles, about an individual, then that individual is actually notable. If nobody writes anything, then that individual is not notable, no matter how important they are. If we could somehow have a US president with no coverage, then that president would not qualify for a Wikipedia article.

Given that I occasionally see news headlines about the Catholic Church declaring someone to be a saint, I assume that basically all of the individuals canonized in recent years will – like US presidents – turn out to be notable. But it's the media coverage that matters for notability, not why the media coverage happened. WhatamIdoing (talk) 00:03, 24 April 2024 (UTC)

While I'm here: You have nominated many articles for deletion recently, and a larger than usual proportion have been kept. This suggests that your personal views about notability do not align closely with the wider community's views. You are obviously doing a WP:BEFORE search, but it's not helping you avoid incorrect AFD nominations.

I think that it would be helpful for you to reflect on what the Wikipedia:Editing policy says: "Wikipedia summarizes accepted knowledge. As a rule, the more accepted knowledge it contains, the better."

Every deletion makes Wikipedia have less knowledge. Sometimes deletion is necessary or desirable, but usually – as a rule of thumb – the long-standing policy says that more accepted knowledge is better than less. Consequently, if an article appears to be summarizing accepted knowledge, then deleting the article will probably make Wikipedia worse. You might have better results at AFD if you restrict yourself to nominating only articles that do not summarize accepted knowledge. WhatamIdoing (talk) 00:13, 24 April 2024 (UTC)

That's some good advice, thank you! Making my AfD noms more consistent has been a goal of mine, and one I have struggled to meet. Advice like this is always appreciated. I try to keep an open mind on AfD noms and try to be quick to admit mistakes when I make them so when I screw up, it at the very least doesn't waste too much time. Allan Nonymous (talk) 00:27, 24 April 2024 (UTC)

From my experience, I agree with "It depends" above. There isn't a one-size-fits-all. People who have been sainted recently are very likely, practically guaranteed, to have coverage of them. Saints from non-Mediterranean regions in the first millennium tend to be more iffy in terms of coverage, but even this is not a good one-size-fits-all judgement because you have saints like Bede and Boniface. Curbon7 (talk) 01:04, 24 April 2024 (UTC)