Wikipedia talk:No original research

Publisher website links and WP:PRIMARY

Moved to Wikipedia:No original research/Noticeboard § Publisher website links and WP:PRIMARY

– 15:34, 10 May 2021‎ (UTC)

Discussion at Wikipedia talk:Verifiability § Verifiability of animal habitat maps

 You are invited to join the discussion at Wikipedia talk:Verifiability § Verifiability of animal habitat maps. {{u|Sdkb}} ^talk 00:27, 9 March 2022 (UTC)

"Anywhere in the world, in any language" and other matters with the new table

This and especially this seems to imply that editors should be expected to somehow check other-language sources, even every other language, before something can be determined to definitely be original research (rather than just, say, failed verification). And do they have to scour 'the whole world' too? Ultimately, the WP:BURDEN of supporting a claim belongs to those adding it. If the cited sources don't support it, then it can be presumed to be original research by synthesis and removed.

Why would or did we ever define OR as based on 'all the RS on Earth' rather than whatever is being used to purportedly support a claim? If someone inserts material that is a synthesis of their sources, why shouldn't this be reverted per WP:SYNTH? It isn't my job, or even possible, to first prove that no sources make the point they are trying to make - I'm aware the new text doesn't specifically say I must, but why can't I use this policy to remove it?

Additionally, even if some RS does support whatever claim the person added, it can be removed per WP:UNDUE or WP:FRINGE if applicable (not just if no source exists for it). Crossroads ^-talk- 03:58, 17 March 2022 (UTC)

I don't interpret it as particularly radical. OR is just one of many tests we apply to content -- fringe that survives OR may still be removable under WEIGHT. But maybe I'm missing some subtle point of policy. Feoffer (talk) 05:50, 17 March 2022 (UTC)

The addition here of the phrase "anywhere in the world, in any language," is an example of WP:DUH, "pedestrian details the reader likely knows or would already assume". I'd remove that particular phrase. 13:46, 17 March 2022 (UTC)

I lean this way, it's overly specific in a way that's not particularly helpful. The WP:BURDEN is on demonstrating WP:V, not those challenging it. Perhaps a rewrite to make it clearer that the challenge can be made with any reasonable doubt that the claim has been published, and that the 'anywhere in the world, in any language' is the scope that the editor who may have added OR is permitted to search for a reliable source. Bakkster Man (talk) 14:13, 17 March 2022 (UTC)

I also disagree with adding the phrase. While basically true, it's stated out of context and without explanation, and I can see how other editors would find that phrasing misleading. There is a decent enough footnote in this policy already, which says "By "exists", the community means that the reliable source must have been published and still exist — somewhere in the world, in any language, whether or not it is reachable online — even if no source is currently named in the article. Articles that currently name zero references of any type may be fully compliant with this policy — so long as there is a reasonable expectation that every bit of material is supported by a published, reliable source." This is at least more clear. Shooterwalker (talk) 15:32, 17 March 2022 (UTC)

• We aren't talking about writing styles for articles here. We are talking about policies and guidelines. P&G need to be very clear, and precise; not concise. Huggums537 (talk) 11:22, 20 March 2022 (UTC)
• If it also appears in the footnote, then it is consistent with the whole of the policy. I agree with it. Huggums537 (talk) 11:29, 20 March 2022 (UTC)
  • "ever published" is still unnecessary and a pedestrian detail. I've removed it, and if I had half a mind more I'd probably remove the whole table until a version which pleases everyone can be agreed upon, because policy is the one place where it is almost always better to be safe than sorry (hence, no reason to be really bold about it), especially something as fundamental as WP:OR. RandomCanadian (talk / contribs) 00:38, 23 March 2022 (UTC)

@Crossroads, about Why would or did we ever define OR as based on 'all the RS on Earth' rather than whatever is being used to purportedly support a claim?: That's actually the point. The definition of OR was originally about a claim posted by an editor that had never been published anywhere (not even in unreliable sources). Early editors were trying to use Wikipedia as a publishing opportunity for "original research", with that name meaning approximately the same as thing as a "research paper containing a completely original idea that really ought to have been submitted to an academic journal instead of being posted at Wikipedia".

Under that model, we had these options:

• I create cold fusion in my kitchen, and Wikipedia is the first place that I write about how I did it: OR violation (no source)
• I read on social media about my neighbor creating cold fusion in her kitchen, and I write on Wikipedia about how they did it: WP:V violation (unreliable source)

Some years later, editors decided that if you copied something off an unreliable source (e.g., social media posts), then that should be considered the same as one that the editors made up themselves.

Under this newer model, we have these options:

• I create cold fusion in my kitchen, and Wikipedia is the first place that I write about how I did it: OR violation (no source)
• I read on social media about my neighbor creating cold fusion in her kitchen, and I write on Wikipedia about how she did it: WP:OR and WP:V violation (unreliable source)

There has never been a time in which OR had any connection to which sources happen to be cited in the article at any point. OR has always been about whether the idea originated from the Wikipedia editor or from external (reliable) sources.

I grant that it's more convenient for the patrollers when editors cite sources. It's much easier to spot OR violations if you give an edit summary of "I just proved this experimentally in my kitchen", and it's much easier for them to avoid the embarrassment of falsely claiming an OR violation if you cite sources that they are familiar with and believe to be credible. But the presence or absence of citations doesn't actually change the relevant facts about whether the Wikipedia editor is posting research that originated from the Wikipedia editor's own experience, and it is those facts alone that define OR. WhatamIdoing (talk) 21:34, 29 March 2022 (UTC)

As for why it's helpful to specify that this applies to any language, any time, and any place: Wikipedia editors are not born knowing that WP:NONENG, WP:PAYWALL, and similar statements are part of the core policies. When we make these things v-e-r-y clear, then diligent editors are less likely to find themselves making embarrassed apologies about their false assertions (e.g., demanding that articles only cite English-language sources). WhatamIdoing (talk) 21:37, 29 March 2022 (UTC)

Do not reinsert this when it clearly lacks consensus. While the historical background may be interesting, nowadays, our concern is not how exactly the editor got some idea. It doesn't matter. It makes zero material difference if the claim they are inserting was made up by them or was read by them elsewhere and they are merely propagating it - it looks the same when being added. And if I revert some junk per OR, and they come back with proper sourcing, I am not "embarrassed" in the least. They should be embarrassed from not properly citing sources to begin with.

This is not the place to add undue-weighted caveats about NONENG and the like. If something is unsourced or synthesizes their claimed sources, it can be presumed OR. In no way should we imply that it is a patroller's job to scour all the world before claiming OR. Crossroads ^-talk- 01:55, 30 March 2022 (UTC)

@WhatamIdoing has already pointed out that it doesn't make any difference where the OR comes from in both the old, as well as the new model, so your remark about the historical background, and that whole first paragraph is kind of pointless.

If there is any place where important caveats need to be clarified, it would be the P&G. It is absolutely false that just because something is unsourced means it can be presumed OR. It is flat out against policy to do that. It is also false that the insertion implies patrollers must scour the world before claiming OR. You have proven that by your own admission in your original post. If you suspect OR, all you have to do is remove per WP:UNDUE or WP:FRINGE, and leave the WP:BURDEN on the editor who inserted the OR. So, you don't need this bit taken out of the P&G. OTOH, WhatamIdoing has shown a practical reason why the bit is helpful to be inserted. Your complaint about what it implies is what really actually doesn't make any difference... Huggums537 (talk) 02:43, 30 March 2022 (UTC)

Against what policy? The WP:BURDEN is on the editor making the claim. And just because I know that I don't have to scour the world first, that doesn't mean newbie editors do, and this page should not put undue emphasis on that matter in such a misleading way. If I suspect OR, I can cite such in my edit summary or on the talk page, not just FRINGE or UNDUE (which don't always apply to bad text). It's much less useful otherwise. Also, I'm not taking anything "out" of the page; it's WhatamIdoing who inserted it both times recently - and neither time did it stick. Crossroads ^-talk- 04:24, 30 March 2022 (UTC)

What policy? WP:V has a section that links to a several pages with more information about the WP:V policies, where you find a certain supplement to the WP:V policy called When to cite. In the When to cite policy supplement, there is a section called When a source or citation may not be needed. It also links to WP:BLUE at the top of that section. These supplements to the policy tell us beyond doubt that there are certain content you simply are not allowed to presume as OR just because it is unsourced. Nothing about the insertion is preventing you, or any newbies from citing OR, FRINGE, and UNDUE other than these imaginary implications that don't make any difference. Huggums537 (talk) 05:14, 30 March 2022 (UTC)

• There has never been a time in which OR had any connection to which sources happen to be cited in the article at any point. OR has always been about whether the idea originated from the Wikipedia editor or from external (reliable) sources. The very first paragraph of WP:OR says To demonstrate that you are not adding original research, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented, which reflects both WP:V and WP:BURDEN. I also strenuously disagree with your assertion that whether the editor themselves or a buddy on Twitter came up with something makes a difference for whether something is OR or not - anything that is implied by a combination of sources and is not citeable to reliable, published sources is OR. If someone cites their buddy on Twitter for it, then it is both OR and poorly-sourced - it does not have to be one or the other; it can reasonably be templated either way and approached from either angle depending on which discussion seems more likely to be productive. This has always been the case and has always been central to how OR is defined, and your history implying otherwise is flatly untrue. Looking back slightly, I also take issue with the giant template you added recently, partially because it appears to be trying to force through this divide you're creating here, though also partially because it is huge and clunky and doesn't really seem helpful enough to justify the massive presence in the policy page. Since I feel your interpretation of WP:OR is a massive shift from the status quo, please do not add anything implying it is the case to any policy pages without an unambiguous consensus supporting it. At the very least, it's clear we need more discussion before such a massive addition to a longstanding policy page, especially if there's such a fundamental dispute over the policy's core purpose. --Aquillion (talk) 04:47, 30 March 2022 (UTC)