We have always treated policies as codification of existing practice. The recently closed Conflict of interest management cited failure to meet the conduct standards expected of an administrator, specifically as pertains to conflict of interest editing and conflict of interest disclosure as sufficient reason to revoke sysop status, with essentially unanimous agreement. So it seems to me it's time to make it official and upgrade this from a guideline to a policy. This would recognize that demanding adherence to this is indeed existing practice, and would be consistent with the community's increasing intolerance of COI editing. RoySmith(talk)16:26, 14 April 2024 (UTC)
@RoySmith: I'm pretty sure we would need an RfC for that. If you plan on starting one, I would most likely support, since I thought that it was already policy. QuicoleJR (talk) 17:08, 14 April 2024 (UTC)
I figured the place to start is with an informal discussion and move on to a formal RFC if it looks like there's broad support. RoySmith(talk)17:49, 14 April 2024 (UTC)
The general principles of the conflict of interest guideline are widely accepted. However, I'm inclined to think revision of the text would be necessary before accepting it as a policy would be possible, in order to have more clarity about what are conflicts and what are acceptable practices. I say this as someone who has been confused by gaps between what the guideline says and what some members of the community expect (and gaps between what different members of the community expect, as this interaction between Donald Albury, Teratix, and Levivich indicates). For instance, the current guideline's examples and language emphasize articles about current (being an owner, employee, contractor, rather than "having been") and close relationships, like an article about a business that one owns or an article about a direct family member. This has led, at least for myself personally, to being surprised to learn there are editors who consider since terminated institutional employment relationships a conflict as well. If this is going to be a policy, I think it's going to need to be shored up to avoid these confusions. I get desires to avoid rules creep, but if the community expects something as a rule but doesn't communicate it, that just creates more muddles and makes it harder to apply, follow, and enforce the potential policy. Hydrangeans (she/her | talk | edits) 18:35, 14 April 2024 (UTC)
Regarding previous employment, I'd say it's not black and white. If I retired from XYZ Corp on good terms and with a generous pension after 30 years of high-ranking service then I would likely still have a COI with the company for at least a few years, especially if I'm still in contact with current employees. Similarly, if I was fired by MegaCorp in a manner that I thought unfair then I would almost certainly still have a COI for probably some months (maybe a double digit number) after my employment ended. However, if I spent a short time temping at Joe Bloggs Ltd or had a summer job at Bob's Burger Franchise then any lingering COI would be measured in weeks at absolute most. If I am a self-employed plumber contracted to fix the toilet at Chain Restaurant then my COI with my client ends pretty much as soon as I walk out the door. This is all generally speaking of course - stronger COIs will last longer than weaker ones for example.
IMO another point point of clarification is what should disclose your COI actually means. I've always interpreted should to be opposed to must. So if an editor fails to disclose their non paid COI, this isn't actually a violation of anything and by itself is unsanctionable. Now if an editor fails to disclose their CoI and then we may have much lower tolerance for any problematic behaviours, so we may block or topic ban them or whatever much more readily but we do actually need something else for it to be sanctionable. However I know from discussions as far back as 15 years or so ago, there are plenty of people who interpret it as a "must" and consider failure to disclose a non paid CoI by itself sanctionable unless perhaps the editor can provide a sufficiently compelling reason for not disclosing. While I haven't followed this case that well, I'm fairly sure I read several commentators who still interpret it as a must and this also seems to be the direction arbconm has lent. Yet I read back in long ago discussions as well as ones when this blew up people who share my view. Nil Einne (talk) 22:09, 14 April 2024 (UTC)
I can envision a slew of problems with trying to make it a policy. It's important not to undercut the harassment policy by giving fuel to those who argue that outing is no big deal if there's a COI; there's been a ton of previous discussion at the harassment policy talk page about the balance between the two. And whereas undisclosed paid editing has an important place in the terms of use, COI covers such a broad spectrum of degrees of "conflict" (just look at some of the talk sections above this one) that it probably lends itself better to a guideline. I recently saw a thread on a good-faith editor's user talk page, where a POV-pusher asked the editor if they were a physician, implying that physicians cannot impartially edit pages about fringe medical theories or quackery. Just imagine where that could go if this were raised to policy-level. --Tryptofish (talk) 22:23, 14 April 2024 (UTC)
It wasn't so much NihonJoe's failure to disclose the COI that resulted in the desysop, but the manner in which he did not disclose. It's almost certain that had his initial response to a civil question of "Do you have a COI with X?" been "Yes, and I also have COIs with A, B and C." it wouldn't have reached arbitration let alone a desysop. Thryduulf (talk) 23:12, 14 April 2024 (UTC)
Quibbling about should? "Should" is not hard to understand, should means, you should do it. Should is used throughout the consensus policies and guidelines: eg. you "should explain the sides, fairly and without editorial bias" is the nutshell of perhaps Wikipedia's most fundamental editorial policy. That does not mean, it's just fine not to do it. Do it, if nothing else, because Wikipedia, which as there is a guideline means Wikipedians by consensus have asked you to. But really, why not take real stock of what we are doing here, Wikipedia is a publisher and responsible writers and publishers disclose COI. -- Alanscottwalker (talk) 23:52, 14 April 2024 (UTC)
'Should' does not mean 'must', and very few policies/guidelines use 'must'. COI isn't something that there should be any doubt around, so this should be one of the exceptions where 'must' is used. -- LCU ActivelyDisinterested«@» °∆t°16:04, 15 April 2024 (UTC)
I don't understand the concern people have about inviting harassment. Of course you shouldn't be subjected to harassment, and people who do harass you should be sanctioned. But as I said earlier nobody's making you edit COI articles. One perfectly good way to avoid being harassed about COI is to just not edit COI articles. We've got 6,812,642 articles right now; surely there's other topics which interest you that you could edit. And if you do chose to edit articles with a COI, all we're asking is that you disclose it. If you're unsure if your association meets the definition of COI, just say what your association is and let other people figure out for themselves. This really doesn't seem like an onerous requirement. And if you do find it onerous, then there's still those 6,812,641 other articles. RoySmith(talk)16:29, 15 April 2024 (UTC)
In addition to Hydrangeans comment, there was also a case recently of a good-faith editor being subject to borderline harassment, to the extent oversight was needed, when a different editor refused to take "no" for answer when accusing them of having a financial COI. We must be careful not to accidentally encourage or legitimise such behaviour. Thryduulf (talk) 20:49, 15 April 2024 (UTC)
That sounds like you are positively encouraging COI writing. "Don't edit COI articles" prevents readers and Wikipedia from being misled, and also editors who don't want to be publicly connected to the subject from being publicly connected. Alanscottwalker (talk) 17:25, 17 April 2024 (UTC)
Something that I don't think has really been addressed so far in this discussion, and needs to be carefully thought through and spelled out before making a formal proposal, is what is the problem that the proposal to make it a policy is going to address? Are we having a problem that results from it being only a guideline, that will be fixed by making it a policy? --Tryptofish (talk) 19:49, 15 April 2024 (UTC)
Yes, and that's the main reason I'd support a change to it being a policy. People frequently break out the "it's just a guideline, not a policy" argument when they disagree with a guideline. HJ Mitchell mentioned that in the case. Firefangledfeathers (talk / contribs) 20:02, 15 April 2024 (UTC)
Thanks. So if I restate that, it becomes: the problem that needs to be addressed is that users often treat COI as unimportant, because "it's only a guideline". I think I could get on-board with that. But I still have serious concerns about whether the community will have consensus about enforcing this as a policy. I understand the comments about "should". But consider someone who has a significant and problematic COI, leading them to make biased edits, and they are not disclosing it, and they have always been very careful not to make it possible to dox them. They should disclose that COI, and arguably they must, but if they don't, what do we do to enforce it? As a guideline, we focus on enforcing NPOV, which is is a policy, and enforcing NPOV as a policy also effectively enforces the COI guideline. So, if we take "COI is only a guideline" off the table, how does this change? --Tryptofish (talk) 20:14, 15 April 2024 (UTC)
I think the community is treating COI like a policy most of the time. This could be problematic for users who are treating it like a guideline and who might think their cases are exceptions. There are, and there will continue to be, COI cases that remain undetectable for a long period, or forever. I don't think upgrading COI to a policy will have much of an effect on those either way. I agree that the quality of the edits made by COI editors is a factor, but it's not just about NPOV. I don't think—if this is what you were implying—that NPOV enforcement is sufficient to handle all COI problems. Firefangledfeathers (talk / contribs) 00:55, 16 April 2024 (UTC)
Yes, just like there are NPOV violations that remain undetected for a very long time or forever, that's clearly not a reason for not having policy or guideline and enforcing them when they come up. Moreover, part of the purpose of the guideline is to promote ethics, and also honesty with readers, which should be what we always try to enforce and reinforce in almost every policy and guideline. Alanscottwalker (talk) 17:32, 17 April 2024 (UTC)
I agree that the quality of the edits made by COI editors is a factor I strenously disagree; the quality of their edits makes no difference at all (notably, this was a repeated defense made for Rachel Helps and she still got topic-banned.) There are clear red lines for COIs where, once crossed, nothing else can serve as a defense. We wouldn't accept "I like their edits tho" as a defense in a sockpuppetry case, I don't see how a COI case is any different. --Aquillion (talk) 23:44, 18 April 2024 (UTC)
The purpose of the COI policy is to maintain the quality of the encyclopaedia. If editors have found that the contributions of an editor with a COI are beneficial to the encyclopaedia (i.e. the articles they've contributed to are notable, neutral, due, etc) why would we want to block them? What benefits would we gain? Thryduulf (talk) 00:12, 19 April 2024 (UTC)
I think you're thinking about direct benefits. The more obvious benefits are indirect: expressing certain views about COI signals that you're part of the in-group, that you share values with those in power (that'd be most of us in this discussion), that you support the current social order, etc. Groups use words and intensification to help strengthen the group's power and identity.
As coroner I must aver I thoroughly examined her And she's not only merely dead She's really most sincerely dead
What's the difference between "only merely dead" and "really most sincerely dead"? Well, nothing, in practice. But the point isn't to change the practicalities; it's to indicate an emotional state. IMO the same thing happens here: We have said for years that you should normally disclose a COI before editing (except when that would require self-OUTING); now we want to say that you really most sincerely should disclose a COI before editing. The practical difference is zero. The psychological difference is significant. WhatamIdoing (talk) 01:56, 20 April 2024 (UTC)
They should disclose that COI, and arguably they must, but if they don't, what do we do to enforce it? We topic-ban them or block them as soon as sufficient evidence emerges. The recent ArbCom case and the ANI cases for Rachel Helps and Thmazing have made that obvious (though it was obvious already, since that has always been our practice.) Obviously some policies (like sockpuppetry) might not always be obvious or could be hard to prove, but once it comes out, what happens is clear. --Aquillion (talk) 23:44, 18 April 2024 (UTC)
But that wasn't what I asked. I asked about when they are very careful about not leaving any way to identify them, and therefore "as soon as sufficient evidence emerges" never arrives. You make a comparison with the sock policy, where we have the checkuser tool, but we wouldn't use checkuser to "out" someone who might or might not have a COI. --Tryptofish (talk) 21:06, 19 April 2024 (UTC)
@Aquillion, that hasn't "always" been our practice. The COI rules were very different back when we made our first edits. We have admins who created articles about themselves, their family members, or their employers, and nobody thought that was a problem back in the day. The idea back then was that if your personal interest aligned with Wikipedia's interests, then there was no "conflict", so writing an autobiography was acceptable, so long as the result was neutral. WhatamIdoing (talk) 02:03, 20 April 2024 (UTC)
Nonsense. A person writing their autobiography is POV writing. They are creating their life story -- now, part of their life story is the writing their autobiography. He wrote that, '[he] was born in a car seat', or he wrote a paragraph on his first job, is not neutral information no matter how it is written, it is important self-information, it is him, telling you about himself, indeed him telling you what's worth knowing. -- Alanscottwalker (talk) 13:18, 21 April 2024 (UTC)
You are conflating notability, neutrality, DUE, and other factors into a mess that accurately reflects nothing more than your personal, rather extreme, point of view:
Whether the subject of an article is notable or not is completely independent of who wrote the article.
It is more likely that an autobiography will be written by a non-notable person than a notable person (simply because there are fewer notable than non-notable people and a significant proportion of notable, living, people with the necessary skills and opportunity to write an article on the English Wikipedia already have an article) but that does not mean that everybody who writes an autobiographical article is non-notable.
What information about the subject of an article is encyclopaedic and due is completely independent of who wrote the article.
What information the article should contain is solely determined by what reliable sources write about the subject, and the consensus of editors deem is important. This always overlaps with what the subject of the article thinks it should contain - sometimes the overlap is tangential, sometimes it's (almost) complete, most often it is between the two extremes. The exact same is true when you substitute "fans of the subject", "haters of the subject", "the subject's family and friends" and many other groups of people.
What information is present in the article is present in the article is completely independent of who wrote the article, what should be in the article, and whether it is neutrally phrased.
Only a featured article contains everything it should contain and nothing else. Every other article contains either omissions or excesses. Given the overlap between what the subject wants in the article and what independent editors want in the article, an autobiography will contain at least some material that consensus says it should (e.g. both an autobiography and an independent biography of a notable actor will include information about their career and at least a partial filmography). How much (and what) is missing and how much (and what) is present that shouldn't be can (obviously) only be judged by the actual content of the specific article.
Whether the information in the article is written neutrally is completely independent of who wrote the article, and of what information is in the article.
It is obvious that this can also only be judged by the words on the page, but it is entirely possible to write both neutrally and non-neutrally about both encyclopaedic and non-encyclopaedic information, and all combinations are possible for both the article subject and for others. It is less likely that an auto-biography will be neutrally written, but that is not the same thing as impossible.
Your statements are the complete mess. Indeed, its as if you have no idea what you are responding to, it's like you just rattling off random talking points. Nothing I said addressed notability, nor encyclopedic. And your pretense of divorcing DUE form neutrality just demonstrates a complete lack understanding of neutrality. You appear to not even understand autobiography. -- Alanscottwalker (talk) 16:51, 21 April 2024 (UTC)
Now try reading what I actually wrote and responding to that rather than assuming that someone who disagrees with you must not be understanding you. Thryduulf (talk) 16:55, 21 April 2024 (UTC)
Oh, I read what you wrote. It is a complete mess of random irrelevant talking points, and nothing but misunderstanding or lack of understanding on your part. And as I said, which you apparently did not read, your misunderstanding goes far beyond me, to neutrality and biography. Alanscottwalker (talk) 16:58, 21 April 2024 (UTC)
I'm not misunderstanding anything, it's a detailed explanation of why you are wrong, but as you have no interested in listening I'll end the conversation here. Thryduulf (talk) 17:16, 21 April 2024 (UTC)
No, your comment is not. It is about other things, not responding to what I wrote, its either pretending that due is not part of neutrality or you are fundamentally misinformed or spreading misinformation. Your comment further betrays a misunderstanding or misinformation about biography and autobiography. Alanscottwalker (talk) 17:33, 21 April 2024 (UTC)
A person writing their autobiography is POV writing. They are creating their life story – I think this is overstated. If the article says "Alice Expert is the chair of the Expert Department at Big University", and that's what independent sources say about Alice, then the article is neutral, no matter who wrote it. (Articles were usually shorter back then. A single-sentence or two-sentence substub was pretty typical.) WhatamIdoing (talk) 00:00, 23 April 2024 (UTC)
No, not neutral, it's the thing the very subject thinks is important to say about the very subject.
(As for creating their life story, it is them living it, which now for your author includes writing their autobiography but deceivingly so, as independent biography.) Alanscottwalker (talk) 10:56, 23 April 2024 (UTC)
@Alanscottwalker You need to explain how the words "Alice Expert is the chair of the Expert Department at Big University" are neutral when written by an independent editor but the exact same words are non-neutral when written by Alice Expert. Until you can do that the foundation of your argument is baseless. Thryduulf (talk) 11:02, 23 April 2024 (UTC)
@Thryduulf the answer to your question is that when an independent editor writes that, they have made the independent (and presumably, unbiased) decision that it is important enough to put in an encyclopedia. The same cannot be said when Alice Expert writes the exact same words. RoySmith(talk)14:06, 23 April 2024 (UTC)
That still doesn't answer the question. If an independent editor makes a judgement that those words are neutral and DUE, but those identical words are not neutral and DUE when written by someone who isn't independent then there must be some way of distinguishing the words without knowledge of who wrote them. Given that the words are identical that isn't possible, therefore you still haven't answered the question. Thryduulf (talk) 14:17, 23 April 2024 (UTC)
I think Thryduulf is correct about the article. An article's neutrality is judged by its contents, not by its author(s). If a given string of words is neutral when I write it, then it's neutral when anybody writes it. WhatamIdoing (talk) 00:00, 29 April 2024 (UTC)
Well, since we're talking about nuance: the basic logic behind the COI policy, as I see it, is to recognise that, like all text, text in Wikipedia articles comes with paratext. The text "Alice Expert is the chair of the Expert Department at Big University" written by Alice Expert and "Alice Expert is the chair of the Expert Department at Big University" written by Bob Volunteer are identical; the paratexts are very difficult. One says this is an editorially independent encyclopaedia, the other says this is Who's Who. One imbues a degree of confidence that Alice being the chair of the Expert Department at Big University is neutral without qualification, the other raises the possibility that Alice might in fact be a disgraced former chair at Big University Degree Mill Inc, working in the hotly-disputed field of Expertise. And so on. In theory we could make the paratext irrelevant by rigorously assessing the verifiability and neutrality of every piece of text contributed. In practice we don't have time, so we take the reasonable shortcut of trusting some contributors of text more than others. – Joe (talk) 12:21, 23 April 2024 (UTC)
How does the paratext differ if both statements are supported by a reliable source? Especially if they are supported by the same reliable source? Thryduulf (talk) 12:39, 23 April 2024 (UTC)
So you are saying that identical text, supported by an identical citation to a reliable source, magically changes from neutral to non-neutral based on whether the person writing the words is the subject or not? I seriously don't understand how you can say that with a straight face. Thryduulf (talk) 13:26, 23 April 2024 (UTC)
Make sure you never find yourself in the literature department of a university then, it'll blow your mind.
But seriously, I agree that the text either neutral or it's not. What I'm trying to say is that in practice, we all know that it's not that easy to tell. It sounds neutral and has a reliable source, great, but does that source really check out? Has it been fairly represented? Is it independent? Has it been cherry-picked from other sources that say something different? Given time you can answer all those questions, but in day-to-day editing we usually take a shortcut by rely on contextual clues. WP:COI just extends that logic into a guideline. If I see a potentially non-neutral edit to Alice Expert from User:Thryduulf, with admin bits and 95k edits, I'm not going to blink an eye. If I see the same text as one of the first ten edits from User:McNewbie, I'll probably check out the source. If I see it from User:AliceExpertOfficial, I'm going to be so suspicious that I'd really rather they didn't make it in the first place.
And just as importantly, I think readers have a similar logic. Luckily, few people actually understand who writes Wikipedia articles or how to check, but everybody I know assumes that our articles are written by random nerds, because if they wanted to read what people and companies had to say about themselves, they'd be on Facebook. – Joe (talk) 14:04, 23 April 2024 (UTC)
So what you are actually saying is that, no, the neutrality of identical text supported by an identical citation to a reliable source doesn't change depending on who wrote it. What you are saying is that your assumptions about the text change based on things other than the words on the page. That is a very different claim to the one you were making previously. Thryduulf (talk) 14:20, 23 April 2024 (UTC)
Another way to look at this is to consider that if Bob Volunteer thinks about making that edit about Alice Expert, and then decides that it would be better not to, then Alice Expert making the edit herself is objectionable. But then if Alice Expert made the edit first, and it got reverted because of COI, after which Bob Volunteer decides to reinstate the edit and assume responsibility for it, then the edit has been vouched for. In that way of looking at it, Alice Expert making the edit by herself is not the same as Bob Volunteer making the edit by himself, although Bob Volunteer can take a positive action to make Alice Expert's edit effectively the same as if he had first made it. --Tryptofish (talk) 22:05, 23 April 2024 (UTC)
Joe, it sounds like your analysis is less about whether the article actually is neutral, and more about how the RecentChanges patroller feels about the edit (e.g., confident because it's a familiar face vs. suspicious because it's not). That's a popular heuristic for everyday patrolling activities, but as you say, it's not really about the article itself. WhatamIdoing (talk) 00:56, 29 April 2024 (UTC)
Oppose. The recent arbitration case highlights the complex interplay between managing conflict of interest in editing and upholding our privacy through the harassment policies. While the case underlines the necessity for clear COI guidelines, escalating these to a policy could rigidify processes that need to remain adaptable to context and sensitivity. The intricacies revealed through the case, where concerns about COI were mishandled or led to heightened disputes, show that a flexible approach is crucial. By codifying the current practices into a rigid policy, we risk not only an increase in administrative disputes but also potential misuse in contentious cases, thereby exacerbating challenges rather than providing clarity. The existing guideline allows for nuanced application that can be adapted as needed, which is more appropriate in handling the varied scenarios of COI that arise in an environment as diverse as Wikipedia. FailedMusician (talk) 23:26, 15 April 2024 (UTC)
Oppose The biggest reason is that the the COI def here is far too broad and vague. North8000 (talk) 00:56, 16 April 2024 (UTC)
Oppose. WP:COI is large, unweildy, and not well defined. The only "policy" we need, in my view, are a few items derived from the guideline: Don't write about yourself or associates in main space, don't cite yourself or associates in main space, don't link to places with which you have an association, use the talk page to propose any substantive change in any article where you have an association with the topic. Something like that, short and sweet. ~Anachronist (talk) 07:06, 16 April 2024 (UTC)
While I do think these shorter and sweeter principles are generally good, the last two I think still would need clarification before further elevation as policy. For instance, with don't link to places with which you have an association: I think it makes sense that we want to avoid promotionalism—the owner of a business wikilinking to their business, or someone who runs, say, a personal online warship database adding their website to external links. On the other hand, I think we go too far if the impression is given that a user who is a member of a professional association like the Society for Historians of the Gilded Age and Progressive Era or the American Association of Physics Teachers shouldn't add citations with DOI links or wikilinks to the organization's peer-reviewed journal, in these cases The Journal of the Gilded Age and Progressive Era or the American Journal of Physics. And any article where you have an association with the topic is a phrasing that's too capacious, imbricating as it would trans editors contributing to biographical articles about transgender people, or Americans editing U. S. president biographies, or maths teachers editing mathematics articles. Hydrangeans (she/her | talk | edits) 04:03, 17 April 2024 (UTC)
In addition to Hydrangeans poitns, any article where you have an association with the topic would prevent any (established) editor from editing the Wikipedia article. Thryduulf (talk) 09:54, 17 April 2024 (UTC)
Something can be "grounded in" something that is "based on" something that is good and still be problematic - especially when those things are proposed to be present outside of their original context as would be the case here. Thryduulf (talk) 00:15, 19 April 2024 (UTC)
Oppose per the above, and per the things I was pointing out in my earlier comments. (I was beginning to wonder if community sentiment had changed suddenly, and I was an outlier. In fact, I was starting to wonder if Jytdog had simply been ahead of his time.) I do want to say that I appreciate the views of those editors who want to treat COI more seriously, but I also want to suggest that trying to make a formal proposal of elevation to policy would be a waste of time and effort. --Tryptofish (talk) 21:20, 16 April 2024 (UTC)
This isn't an RfC so I won't !vote. But if this were an RfC then I would oppose making the current guideline a policy. There is too much in the current guideline - and how many editors interpret it - that assumes that COIs exist in a binary state and that all COIs are equal. In my professional experience, the organizations and people with whom I have worked recognize that COIs exist on a very large spectrum. Some bright lines are helpful and appropriate - WP:PAID is a good example - but this concept is much more nuanced than currently described and handled in this project. ElKevbo (talk) 22:08, 17 April 2024 (UTC)
I think we absolutely do need a policy about this, because in practice we are already using it as policy, because it is extremely serious and because there are clear red lines; in recent disputes over this many people who ultimately ended up topic-banned or blocked have breezily defended their actions (and had their actions defended by others) by arguing that the whole thing was not policy and therefore they weren't required to follow it. They were wrong, as can be seen from the numerous recent blocked; and fact that there gray areas makes it more important to have a well-thought-out, clearly-defined set policies for how we approach COI issues, not less. That doesn't necessarily mean that the current wording is ideal and should be translated directly into policies, but we need to think about what the red lines are, what the gray areas are, and when and how people have responsibilities to disclose potential COIs or to avoid editing in an area entirely. Clearly they do have such responsibilities; editors are regularly being blocked for failing to meet it. We need to make at least some effort to have a policy page that reflects current practice and makes it clear to people what their responsibilities are, because otherwise we're going to end up with more situations where editors go "ah I don't have any responsibilities, it's not policy" and then end up blocked by ANI or ArbCom. That is not ideal - the ideal situation is to have clear policies that make the requirements as clear as we can in advance, so editors don't find out where the red lines are only after being blocked. --Aquillion (talk) 23:38, 18 April 2024 (UTC)
I see several editors saying that we use this as a policy, and I wonder if we all agree on what the difference between policies, guidelines and essays actually is. I suspect that we don't, and what's meant is "we enforce this pretty strictly". We enforce a lot of "mere guidelines" pretty strictly. We even block people for violating "mere essays". On the other hand, RecentChanges patrollers violate the WP:PRESERVE section Wikipedia:Editing policy every hour of the day, and none of them get blocked. WhatamIdoing (talk) 02:08, 20 April 2024 (UTC)
Our enforcement is haphazard at best in most things, because many of our policies and guidelines rely on decentral enforcement, decentralized judgement, and fuzzy (or is that, nuanced) lines, content or editing policies or guidelines are probably the most haphazard, as there is minimal barrier of entry for any edit. Perhaps, the best we can do here is stress and reenforce for an amature encyclopedia writing publisher, the ethics of the matter. Alanscottwalker (talk) 12:14, 20 April 2024 (UTC)
That's true, but I think what's more relevant is that policies tend, on average, to articulate more general principles (e.g., "Wikipedia is not a place to publish new ideas", "Wikipedia is not a web directory", "Thou must not violate copyrights") and the other pages get into the details (e.g., all the different pages about identifying and dealing with POV pushers, spammers, and copyvios).
Perhaps, but that seems less relevant than, some things are easier and some things are harder, but it is the harder that are usually more central to complex tasks. And functionally it does not compare to decentralization as a reason for uneveness. Alanscottwalker (talk) 13:28, 23 April 2024 (UTC)
Per FFF above, I'd support promoting the guideline to a policy unchanged, and I'd also support some changes. (Shorter, clearer, probably stricter.) Levivich (talk) 06:40, 21 April 2024 (UTC)
Well, I'd need to know what problem this is solving. The case OP cites did end with the person getting the heave-ho on COI grounds, so it didn't seem necessary there. We have a lot of policies here, and a some of them suck, so I would support a trade: enact this as a policy in return for demoting another one, leaving no net change in number of policies. Which one to demote would be up to the community, I can think of several. Herostratus (talk) 01:33, 23 April 2024 (UTC)
Hi, I'm a paid editor who has recently been banned for a lack of COI disclosure (but which I believe was actually an issue of NPOV that other editors were unable to articulate well). I believe that we should get rid of this whole guideline. It promotes suspicion of editors, and since Wikipedia allows anonymous editing, requires some editors to be known publicly while allowing others to remain anonymous. This creates an environment that motivates outing and discrimination based on an editor's relationships and background. It is inherently unfair. Instead, we should focus on whether or not edits maintain NPOV. I would support making the policies surrounding NPOV clearer and having an environment of enforcing NPOV more strictly. Rachel Helps (BYU) (talk) 19:42, 13 May 2024 (UTC)
While the response is unnecessarily rude, I do see the humor in a paid editor arguing that COI editing can produce neutral results, as if people can be immune to implicit bias, confirmation bias, and, well, just bias. As if Wikipedia doesn't have a 20-year history of COI editing leading to NPOV problems. As if we didn't just prove that again in recent months. As if COI editing leading to NPOV problems isn't the reason why this particular paid editor is now topic banned. There's this very famous Upton Sinclair quote, "It is difficult to get a man to understand something, when his salary depends on his not understanding it." And, indeed, paid editors and COI editors seem to have difficulty understanding that paid editing and COI editing invariably leads to non-neutral editing. Funnier still is the argument that restrictions on COI editing are "inherently unfair" -- unfair to whom? The COI editors? The idea that fairness requires letting people edit about subjects with which they have a COI is, well, laughable. Levivich (talk) 17:22, 14 May 2024 (UTC)
invariably leads to non-neutral editing: Recent months don't indicate this is invariably true, per a finding of fact from an Arbitration Committee case about editing while having a conflict of interest [that] did not, in general, violate other content policies or guidelines; this would include WP:NPOV (the case maintained that conflict of interest editing is against guidelines; the point is that the claim about an invariable chain between COI and POV isn't, well, invariably true).
unfair to whom? Unfair to those who become targets of discrimination and harassment. If a Wikipedia guideline inadvertently motivates harassment or discrimination, I can understand why an editor might consider it unfair. I'm reminded of recent experience at Wikipedia:Miscellany for deletion/Wikipedia:No queerphobes, where it became apparent that some detractors to the essay believe that being LGBTQ+ constitutes a Conflict of Interest for LGBTQ+ topics (see for instance the comment claiming the essay is really only going to be [..] used to claim that editors do not have a conflict of interest even when it's obvious)—a claim that WP:No Queerphobes rightly says is a queerphobic claim—to the point that the deletion discussion was taken to Deletion review to overturn the keep decision on the claim that notifying WikiProject LGBT Studieshad the effect of prejudicing the discussion. While the XfD decision was not overturned, it was pretty chilling to see editors so willing to try to justify excluding LGBTQ+ editors from matters pertaining to LGBTQ+ topics. Among the comments opposing overturning, there were even a few that apparently agreed that notifying WikiProject LGBT Studies did wrongfully prejudice the discussion, which was unsettling.
laughable: I don't see what's funny about someone being concerned about harassment; that'd seem to entail supposing, however inadvertently, that people being harassed is funny. Hydrangeans (she/her | talk | edits) 20:48, 14 May 2024 (UTC)
It may be my inability to articulate well Rachel, but I fail to understand how the following sentences are logically coherent: "It promotes suspicion of editors, and since Wikipedia allows anonymous editing, requires some editors to be known publicly while allowing others to remain anonymous. This creates an environment that motivates outing and discrimination based on an editor's relationships and background. It is inherently unfair." Could you also please expand on which NPOV policies you think should be made clearer, and how they should be enforced? ~~ AirshipJungleman29 (talk) 00:38, 15 May 2024 (UTC)
Sorry if that wasn't clear. WP:PAID requires that paid editors reveal the identity of their employer; we assume, correctly, that any paid editor has a financial COI regarding their employer. However, there are other COIs, which the policy proposed on this page would cover. For example, it's a COI to edit the page of a friend or relative. However, this relationship is not apparent unless the identity of the author is revealed (encouraging authors who want to edit pages for people they have a COI for to be anonymous, because of the scrutiny declaring a COI often causes). Since so many people are editing anonymously with a COI, other editors become suspicious that their fellow editors have undisclosed COIs. This leads them to wonder what the identity of the other editors is, sometimes resulting in harassment. So here is a possible alternative. Let's say that an editor whose identity is unknown is making a lot of edits that seem to be promoting a living author. Rather than questioning them about a possible connection to the author, a fellow editor could ask them to comply with NPOV. If they disagree, they could ask for a third opinion, or take things to the NPOV noticeboard. When an admin agrees that they are not up to NPOV editing standards, they could give that editor a warning. After this formal warning, continued NPOV editing could result in a topic ban. That way we could use an existing policy (NPOV) to enforce what we actually care about, which is NPOV. We don't have to ask the editor what their relationship to the author is. There is no hidden information! We could make WP:UNDUE more detailed about what constitutes undue weight, with examples, so it is easier to understand. I think my own past editing had issues with NPOV, and in conjunction with my topic ban, and to demonstrate I understand what the problems with my editing were, I am in the process of drafting guidelines for NPOV specific to religious topics. To comply with my topic ban, this is off-wiki. Please email me if you would like to know more. Rachel Helps (BYU) (talk) 15:52, 15 May 2024 (UTC)
They ask for a third opinion and then a third editor comes and gives an opinion and turns out that third editor is also a friend of the article subject and doesn't disclose it. This is what actually happens in reality. Levivich (talk) 16:32, 15 May 2024 (UTC)
"So here is a possible alternative..." Yes, a perfectly normal procedure. Take a discussion I participated in today on Talk:Bachelor, where two editors got in a dispute, and I answered a third opinion request; one of the editors attempted to continue their non-neutral editing, and was subsequently page-blocked by an admin. Again, this is common. The "process" you seem to be pushing back against, where editors take the time to extensively investigate undisclosed COIs, is much rarer (I do sometimes have a life). I think you are overestimating the extent to which your topic ban represents a widespread problem (the case was in reality unique in nearly every way), among other things. ~~ AirshipJungleman29 (talk) 03:15, 16 May 2024 (UTC)
one of the editors attempted to continue their non-neutral editing The point is that it is irrelevant whether they were doing that because they have a COI with the topic or for some other reason. If they are unable or unwilling to edit neutrally they should not be contributing (to that article/subject), regardless of why that is. If they are able and willing to edit neutrally there is no benefit to preventing them contributing. Thryduulf (talk) 08:02, 16 May 2024 (UTC)
Comment It seems like this accidentally evolved into an RFC. Do we consider this discussion to be one? If so, we should probably update the formatting and list it at CENT. If not, we should probably go ahead and start one to figure out community consensus. QuicoleJR (talk) 16:44, 16 May 2024 (UTC)
Initially, I understood this discussion to be a sort of testing-the-waters, prior to starting any sort of formal proposal. Based on the discussion so far, it's not clear to me that there is enough support for upgrading to make a formal RfC worth the effort. (But of course anyone who wants to start an RfC is free to do so.) --Tryptofish (talk) 20:37, 16 May 2024 (UTC)
Updated proposed text for a shorter and clearer COI guideline
To preserve the integrity, neutrality, and public trust in Wikipedia, it is crucial to effectively manage conflicts of interest among editors. A conflict of interest arises when an editor's personal or financial connections might compromise the objectivity of their contributions. This guideline outlines the types of conflicts and specifies the conduct required for editors who may be affected by them.
Financial Conflict of Interest
An editor has a financial conflict of interest when they stand to gain, or can reasonably be expected to gain, financial benefits from the coverage of a topic on Wikipedia. This conflict arises in various forms:
Direct Financial Benefits: These include receiving direct payment for editing Wikipedia articles.
Indirect Financial Benefits: Such benefits are not as overt as direct payments but are significant. Examples include:
Business Exposure: Gaining from increased visibility when a product, service, or company is featured in an article.
Reputation Enhancement: Benefiting indirectly from an enhanced reputation due to having a personal article on Wikipedia or being cited as a source in oneor being prominently featured as an expert. This can lead to increased professional opportunities, such as book sales, speaking engagements, or consultancy work.
Non-financial Conflict of Interest
An editor has a non-financial conflict of interest when their personal or professional connections may compromise their ability to present a subject objectively. This type of conflict arises in various forms:
Personal Relationships: Editing articles about friends, colleagues, family members, romantic partners, or personal adversaries can lead to biased content, whether overly favorable or unduly negative.
Professional Connections: Editing articles related to one's employer or competitors in the industry can introduce biases that may either unfairly promote one’s own organization or undermine others. Similarly, citing oneself or ones close acquaintances as sources can introduce bias, influencing the content to unduly favor personal or professional interests.
Political and Ideological Beliefs: Editing articles that cover political or religious movements in which an individual holds an active role. It is important to note that simply being a supporter or member of such movements does not constitute a conflict of interest. For example, a member of the Democratic Party does not inherently have a conflict of interest when editing articles related to the party unless they hold an official role within it.
Managing Conflicts
Editors with a Financial Conflict: Must not directly edit affected articles. Instead, they should propose changes using the Edit COI template, disclosing the nature of their conflict on their user page and in any location they discuss the topic.
Editors with Non-financial Conflicts: While not strictly forbidden from directly editing affected articles, they must disclose their conflict in the edit summary and in any location they discuss the topic.
Exceptions
No exceptions, except those explicitly listed here, exist to this guideline; edits made in violation of it are indefensible, regardless of their perceived harmlessness or quality. Editors who wish to avoid disclosing their conflict of interest may do so only be avoiding topics affected by it.
Aside from those explicitly listed here, no exceptions exist to this guideline; edits must abide by it regardless of their perceived harmlessness or quality. Editors who wish to avoid disclosing their conflict of interest may do so only be avoiding topics affected by it.
General exceptions
Reverting obviousvandalism—edits that any well-intentioned user would agree constitute vandalism, such as page blanking and adding offensive language.
Removal of clear copyright violations or content that unquestionably violates the non-free content policy (NFCC). What counts as exempt under NFCC can be controversial, and should be established as a violation first. Consider opening a deletion discussion at Wikipedia:Files for discussion instead of relying on this exemption.
Removing contentious material that is libelous, biased, unsourced, or poorly sourced according to Wikipedia's biographies of living persons (BLP) policy. What counts as exempt under BLP can be controversial. Consider reporting to the BLP noticeboard instead of relying on this exemption.
Reverting unambiguous spam, where the content would be eligible for page deletion under criterion G11 if it were a standalone page.
Wikipedians in residence
A Wikimedian in Residence (WIR) is a professional role in communications for an organization to share its knowledge within the Wikimedia platform, measure the impact of the same, and promote Wikipedia through training, education, and edit-a-thons. While WIRs can greatly benefit Wikipedia, there is a risk of edits that could unduly benefit their employers. To manage this risk, the following guidelines apply:
Scope of Editing: WIRs may edit articles related to but not directly involving their institution. They must not edit articles where their institution is the primary subject or could reasonably be seen as directly benefiting from the article's content; if they wish to make changes to such articles they must follow the instructions at #Managing Conflicts for Editors with a Financial Conflict.
Disclosure Requirements: WIRs must clearly disclose their role and association with their institution on their user page, as well as in any discussions or edits related to their role as a WIR
I have updated the proposed changes per the discussion above. In particular, this has included the expansion of exceptions based on the exceptions listed at WP:3RRNO, and changed this from a proposed policy to a guideline - this would replace the current content, with most of the current content becoming part of a supplementary essay. Personally, I do believe this would be better as a policy, but better to handle one change at a time. BilledMammal (talk) 02:41, 10 May 2024 (UTC)
This is clearer, but it doesn't reflect the actual enforcement. It is trying to propose something that is much stronger than actual practice, which is something we generally do not do here. In particular, the "Editors with a Financial Conflict: Must not directly edit affected articles.", which either means "outlawing" all paid editing which the TOU currently allows. Plus it bans all non-problematic edits in articles someone just has a weak connection to (my AT&T example), and isn't really enforceable. "Significant" isn't a very good delineator, because it doesn't mean the same thing to everyone. For example, when I did a rewrite about "tanning beds", I was wise enough to do it with someone who had no COI (I had a strong COI at the time, but that was the source of my expertise) who had final say. I had announced my COI on my user page, but I wasn't going to just make requests on a talk page, that is too cumbersome, particularly when I had already gone to great lengths to make sure any bias that I might inject had oversight. In other words, it would have made someone either not improve the article, or more likely, lie about their COI and not get any oversight from a disinterested party. Neutrality would have been reduced. And I use my own editing because it isn't theoretical, it can be examined by anyone here, and I would argue that I used best practices. I think the community needs to accept they will never stop Paid or COI editing, and instead focus on managing it, because it is impossible to prove or verify and it is pretty easy to game strict rules. The harder you squeeze, the less neutrality (and disclosure) you will have. Dennis Brown - 2¢03:41, 10 May 2024 (UTC)
Regarding your AT&T example, my understanding is that we have already established that it wouldn't constitute a conflict of interest.
We didn't establish the "signficance" with AT&T, as I never said how much I had, people just assumed one way or another. As for the tanning beds, I had previously worked for a very small tanning bed manufacturer. Even the guys on WPO have talked about and found humor in it. I've never hidden the fact, had it on my user page at one point. At the time, I owned Solacure, which designs and sells UV lights for several industries (animal husbandry, cannabis, luthier, etc), but not human tanning, so unrelated to the article. I had that on my user page for a time as well. People will see it however they want, but there was no financial motivation or gain that I could have possibly realized by editing it. The COI meant I have some expertise in the field, however, so it made sense for me to be a part of the editing, along with SlimVirgin, who took the lead.
Real world practice is more complicated than you can put in a paragraph of "should"s and "must"s. I support firmer disclosure rules, and I actually understand how it affects the editor. I'm against making rules that no one will follow and will lead to less disclosure, but make us "feel good". I'm pretty sure a consensus won't approve of a ban of Paid and COI editing anyway, so it wastes time that could be spent on practical changes. Hard rules that ban COI editing simply won't work even if passed. It would just drive COI into the shadows, with less oversight. That is harder to manage than having a set of reasonable rules that don't ostracize editors and allow for oversight of COI edits in plain view. If the edits are promotional or spam, this is already covered by other policies and easily dealt with the same as we do any other spam. Dennis Brown - 2¢05:30, 10 May 2024 (UTC)
I like these changes. Some feedback:
I would rephrase "No exceptions, except...," to something that doesn't repeat the word.
I would change "indefensible" to something else. The important part of that sentence is "regardless of their perceived harmlessness or quality," that message should be kept, but "indefensible" sounds too combative or judgmental to me; like the policy/guideline shouldn't suggest that it's immoral, just explain the rules matter-of-factly.
I know we want to keep it short, but perhaps it could benefit from some examples of direct/indirect financial COIs, like perhaps some (not necessarily all, not sure which are best to include if any) of the following situations:
owning a business, or being a director, officer, executive, or other high level position v. low level positions
being paid to edit v. working in marketing v. just working in a business
working in a small biz v. a large one (ie being one of few v. one of many)
"de minimis" connections, like owning one share of stock in a company (which probably shouldn't be considered a COI)
"attenuated" connections, like having a retirement plan that invests in a mutual fund (that the editor does not direct) that owns some shares in a company (prob not a COI)
"stale" connections, like having worked in a company many decades ago (as opposed to left employment last year) (prob not a COI)
Maybe it should explain COI is a continuum: the closer the connection and the more significant the relationship, the stronger the COI. So indirectly owning a few shares ten years ago isn't as serious as currently owning a majority stake that provides most of your income or constitutes most of your assets.
What's the logic behind having different rules for financial and non-financial COI? I feel like people might have a stronger COI/bias editing about family members than employers.
Does anybody like the "if you put it on your resume/CV, you have a COI with it" formulation?
Does anybody like the "if you put it on your resume/CV, you have a COI with it" formulation? I'm not very fond because it seems prone to the unconstructive interpretation 'because you are a member of/participate in this academic society (since you want to attend its conferences and/or receive its periodical so as to be aware of and have access to scholarly knowledge in this field), adding citations to its periodical is conflict-of-interest editing'. To reiterate my comment in the earlier discussion, I think we go too far if the impression is given that a user who is a member of a professional association like the Society for Historians of the Gilded Age and Progressive Era or the American Association of Physics Teachers shouldn't add citations with DOI links or wikilinks to the organization's peer-reviewed journal, in these cases The Journal of the Gilded Age and Progressive Era or the American Journal of Physics.Hydrangeans (she/her | talk | edits) 11:36, 10 May 2024 (UTC)
Thank you; I've implemented #1 and #2 - what do you think?
I also like the idea of #3, but I can't yet see how to do it. I will give it some more thought.
With #4, I'm concerned that editors will result in disputes about whether it was appropriate for an editor to not report; I think it is better to simply set up clear lines and say that regardless of whether you are an inch or a mile over the line you have to comply with the restrictions. However, I'm not set on this position, and if other editors feel strongly I am willing to change it.
The logic is mainly that I don't think we will get a consensus to merge the rules; editors won't agree that editors who are paid to edit should be allowed to edit directly, and editors won't agree that editors who are writing about a colleague shouldn't be allowed to edit directly. BilledMammal (talk) 18:34, 13 May 2024 (UTC)
The #1/#2 changes LGTM, thanks. I see #3 and #4 as being related... overall, #3 lays out that continuum, and to your point, also provides various places to draw lines. Insofar as we want the COI guideline/policy to draw clear lines rather than leaving things fuzzy, perhaps it can draw lines that would clearly include/exclude certain things listed in #3. One thing I'm thinking of in particular is the "de minimis/attenuated/stale" trifecta: the guideline could specify that none of those are COI. That would take care of the "AT&T issue" raised by Trystan below. Trystan's "significant roles in organizations" language would be an example of drawing lines around some of the categories in #3 (first and third bullet points). I understand the logic of #5, thanks for explaining that. "The art of what is possible." Levivich (talk) 20:41, 13 May 2024 (UTC)
The current guideline notes that COI investigation doesn't justify violating the policy against harassment. This proposed draft does not. As this same absence was pointed out in the earlier discussion about your earlier drafts, is this an intentional elision? The importance of the policy against harassment has been affirmed in an Arbitration Committee finding of principle, and the boldness with which editors have violated it makes this a bad time for obscuring the intersection of WP:COI and WP:HARASS. Any version of the COI guideline or elevation of it to policy should maintain our strong reminders that editors must abide the policy against harassment.
I'll add that I find the claims about indirect benefits, particularly This can lead to increased professional opportunities, such as book sales, speaking engagements, or consultancy work, unconvincing when applied to both articles and citations. I also notice that an equivalent doesn't seem to appear in the current version of the guideline (in fact, it contradicts the guidance that citations to one's own work are allowed within reason, but only if it is relevant, conforms to the content policies, including WP:SELFPUB, and is not excessive), making this not a shorter rephrasing but a novel alteration). I could believe there is potentially some benefit a person gain from there the existence of a biographical article about them—a Wikipedia article about a person is relatively visible on the Internet, and an editor making or sans-exception contributing to a biographical article about themselves is definitely a bright line—but it's my understanding that few non-editor readers look at the citations in articles. The notion that appearing in the citations of a Wikipedia article could so measurably benefit a writer's career strikes me as an inflated sense of readers' engagement with Wikipedia citations. Wikipedia citation sections aren't being combed by event coordinators at universities and libraries but by high schoolers and undergrads trying to follow rote instructions about media literacy who probably forget the authors' names within days of turning in their homework.
I continue to share, as I described in the earlier thread, ElKevbo's skepticism of trying to handle Conflict of Interest in such a bright line, black-and-white way as this proposal would—especially one that makes no effort of caution about adhering to the the policy against harassment. Hydrangeans (she/her | talk | edits) 06:54, 10 May 2024 (UTC)
I've made a few changes to address your concerns; below I've created a draft section on how to handle suspected violations, and I've adjusted the draft above to consider citing oneself, one one's close acquaintances, as a non-financial COI rather than a financial one. BilledMammal (talk) 19:00, 13 May 2024 (UTC)
To a degree, no editor should be adding any source to any article unless they think, 'this is a source to be read on this subject (and indeed, the source to read on this bit)' -- add in a financial and probably more so, a reputational or personal motive, to make the source the editor had published, one of the few sources to read on this subject out millions and billions of sources, self-cite is in need of a check. Few publishing writers would not want be relevant, just to be relevant. In American Fiction (film), one of the funny and truthful bits for writers was the lengths the author would go to get his writing out there (and it's certainly not just financial motive, it is personal). Mostly we don't ban self-cite for the sake of writers with expertise, who may actually be the author of the source you should read on this subject, so I agree we should bring back some the wording (perhaps modified) of the current guideline (even though the exception also applies to, for example, journalists where there may be dozens sources just as good -- and for journalists there is an added conflict, as much as we try to say we are NOT, Wikipedia is probably in direct market competition with their publication - eyes on)). -- Alanscottwalker (talk) 11:29, 10 May 2024 (UTC)
As revised, I think there is still the "AT&T" issue, where someone with a relatively trivial financial conflict of interest would be barred from editing, which isn't how I read the current guideline. I think keeping the bright line between paid editing and all other COI editing is more feasible.
The "Political and Ideological Beliefs" subsection is a bit confusing to me, and I think potentially risks creating confusion between COI and bias. As I understand it, COI is about a personal connection to the subject. A significant role in any organization, whether or not it is religious or political in nature, creates a COI. Religious and political beliefs may create bias, but are not relevant to whether a COI exists. I would suggest considering replacing it with something like: Signficiant roles in organizations: Editing articles that cover organizations in which an individual holds a significant role. It is important to note that simply being an ordinary member of such organizations does not constitute a conflict of interest. For example, a member of the Democratic Party does not inherently have a conflict of interest when editing articles related to the party unless they hold an official role within it.--Trystan (talk) 19:10, 13 May 2024 (UTC)
How about:
Signficiant roles in organizations: Editing articles related to organizations in which an individual holds a significant role, or recently held a significant role, may introduce biases and a lack of objectivity in content related to the organization's interests. This is a spectrum, with whether an editor has a conflict of interest depending both on the level of authority or influence their role granted them, and the recency of the role. For example:
A general volunteer for the Democratic Party would have no conflict of interest.
A precinct captain would have a conflict of interest for a few years after they hold the role.
When violations of this policy are suspected or identified it is crucial to address them with transparency and caution, balancing protecting the encyclopedia with respect for the editors involved. Always adhere strictly to our no-outing policy; only post personal information if the editor has disclosed it on Wikipedia.
User talk page: Non-urgent issues can be raised on the editor's talk page, using the COI warning template as appropriate.
COI noticeboard: If issues remain unresolved after user talk page discussions, or if the user talk page is an unsuitable venue, escalate the matter to the Conflict of Interest Noticeboard. This noticeboard also provides guidance for editors dealing with their own conflicts.
Private communication: For issues requiring confidentiality, including where posting the information on Wikipedia would violate our policies on the posting of personal information, email evidence to the appropriate channels: for general COI issues, contact functionaries-en@lists.wikimedia.org, and for paid editing concerns, reach out to paid-en-wp@wikipedia.org. Always consult these channels for advice before sending private information.
This addition would improve the proposal, though I still don't support it overall. Additionally, I object to the phrasing of the first sentence, address them with transparency and caution, balancing protecting the encyclopedia with respect for the editors involved, as it suggests that respecting editors is somehow in tension with "protecting the encyclopedia", entrenching an attitude of aggrieved defensiveness and hostility. Respecting editors is protecting the encyclopedia, and we're ignoring WP:5P4, WP:CIV, and WP:HARASS if we neglect civility and respect as core pillars and policies of the project. Hydrangeans (she/her | talk | edits) 20:25, 14 May 2024 (UTC)
I don't see the issue you see here; the two aren't in opposition or alignment; they are complementary and we can often achieve both, but sometimes we need to carefully balance them.
We respect editors by avoiding baseless COI accusations while protecting the encyclopedia by allowing editors to express COI concerns supported by evidence, even when the editor has denied a COI. We respect editors by forbidding the public disclosure of personal information they haven't shared on Wikipedia, while protecting the encyclopedia by allowing private presentation of such evidence when necessary, and allowing the public sharing of previously disclosed personal information.
Proposed text for a shorter and clearer COI policy
To throw out a possible summarized and clarified wording, either for this guideline or for a hypothetical policy. This is only a first draft, and feedback would be appreciated. BilledMammal (talk) 07:41, 21 April 2024 (UTC)
To preserve the integrity, neutrality, and public trust in Wikipedia, it is crucial to effectively manage conflicts of interest among editors. A conflict of interest arises when an editor's personal or financial connections might compromise the objectivity of their contributions. This policy outlines the types of conflicts and specifies the conduct required for editors who may be affected by them.
Financial Conflict of Interest
An editor has a financial conflict of interest when they stand to gain, or can reasonably be expected to gain, financial benefits from the coverage of a topic on Wikipedia. This conflict arises in various forms:
Direct Financial Benefits: These include receiving direct payment for editing Wikipedia articles
Indirect Financial Benefits: Such benefits are not as overt as direct payments but are significant. Examples include:
Business Exposure: Gaining from increased visibility when a product, service, or company is featured in an article.
Reputation Enhancement: Benefiting indirectly from an enhanced reputation due to having a personal article on Wikipedia or being cited as a source in one. This can lead to increased professional opportunities, such as book sales, speaking engagements, or consultancy work.
Non-financial Conflict of Interest
An editor has a non-financial conflict of interest when their personal or professional connections may compromise their ability to present a subject objectively. This type of conflict arises in various forms:
Personal Relationships: Editing articles about friends, colleagues, family members, romantic partners, or personal adversaries can lead to biased content, whether overly favorable or unduly negative.
Professional Connections: Editing articles related to one's employer or competitors in the industry can introduce biases that may either unfairly promote one’s own organization or undermine others.
Political and Ideological Beliefs: Editing articles that cover political or religious movements in which an individual holds an active role. It is important to note that simply being a supporter or member of such movements does not constitute a conflict of interest. For example, a member of the Democratic Party does not inherently have a conflict of interest when editing articles related to the party unless they hold an official role within it.
Managing Conflicts
Editors with a Financial Conflict: Must not directly edit affected articles. Instead, they should propose changes using the Edit COI template, disclosing the nature of their conflict on their user page and in any location they discuss the topic.
Editors with Non-financial Conflicts: While not strictly forbidden from directly editing affected articles, they must disclose their conflict in the edit summary and in any location they discuss the topic.
No exceptions exist to this policy; edits made in violation of it are indefensible, regardless of their perceived harmlessness or quality. Editors who wish to avoid disclosing their conflict of interest may do so only be avoiding topics affected by it.
The first section talks about "financial conflict of interests" but the third talks about "paid" and "non-paid conflicts". That the existing text uses these terms interchangeably was a major source of confusion in the recent arb case; I think we ought to stick to one or the other. – Joe (talk) 09:05, 21 April 2024 (UTC)
I appreciate this attempt although I'm very skeptical that any attempt to approach this inherently nuanced topic without nuance is fatally flawed. For example, a proposed policy must account for (a) employment that isn't directly related to editing Wikipedia or promoting the employer (e.g., marketing, public relations) and (b) editing by experts who are connected to a topic but perhaps not directly connected to a specific article subject (e.g., scholars). The draft language for "Non-Financial Conflict of Interest" (why the weird capitalization?) would seem to disallow edits by scholars and experts which is certainly not what we intend. And in the areas in which I focus - US colleges and universities - we also deal with edits made by students and alumni who arguably have a COI. And what about GLAM editors and Wikipedian-in-residence programs who have project-approved exemptions to our typical COI policies and practices? ElKevbo (talk) 12:16, 21 April 2024 (UTC)
Thank you for the reply; I agree that the text can be improved, but I hope that we will be able to do so.
employment that isn't directly related to editing Wikipedia or promoting the employer (e.g., marketing, public relations)
I think that is accounted for by the section on "Professional Connections"
The draft language for "Non-Financial Conflict of Interest" (why the weird capitalization?) would seem to disallow edits by scholars and experts which is certainly not what we intend.
My intent was to allow such edits; for example, if you are an expert in Austro-Hungarian history, then there should be no issue contributing to the article October Diploma, so long as you avoid citing yourself or your colleagues. However, the fact that you think the wording would prevent this means that the wording can be improved; why do you think the wording would prevent this? (Also, edited the capitalization; feel free to edit it further)
And what about GLAM editors and Wikipedian-in-residence programs who have project-approved exemptions to our typical COI policies and practices
Regarding GLAM and WIR editors, that's a good point; I've drafted an expanded "Exceptions" section below, which would exempt them from the financial COI restrictions with limited exceptions; the intent is to allow them to share the knowledge of their institution without promoting it, in line with meta:Wikimedian in residence. Do you feel it meets this intent? BilledMammal (talk) 12:57, 21 April 2024 (UTC)
If you are an expert in Austro-Hungarian history, we actually do want you to cite yourself and your colleagues, within reasonable limits. WhatamIdoing (talk) 00:10, 23 April 2024 (UTC)
Why? In this circumstance, I don’t think you are well placed to decide whether your work, or the work of your colleagues, is WP:DUE; better to make an edit request and let editors without a conflict of interest figure that out. BilledMammal (talk) 15:24, 24 April 2024 (UTC)
Honestly, if an article is getting to the point where only a tiny community is knowledgeable enough to edit it accurately, the content is no longer suitable for a general encyclopedia anyway. We shouldn't "need" experts to edit in their sub-sub-specialty. JoelleJay (talk) 14:29, 27 April 2024 (UTC)
As well-meaning as this sentiment is, this sense the project doesn't "need" experts who are well-versed in their specialities seems too near to the anti-intellectualism that permeates Wikipedia and contributes to the the project's imbalanced content coverage (this linked article focuses on exclusion of certain kinds of sources based on reliability, but I would take this observation to instead point out that experts can be aware of academic sources in other languages, or that are available primarily through academic libraries rather than convenient Google searches). Furthermore, there are plenty of plainly notable topic areas that suffer from the lack of expert intervention, such as the soft-pedaling of Anglo/British settler colonialism (permanent). The project does need, desperately, for experts to contribute in their specialities and sub-sub-specialities. Hydrangeans (she/her | talk | edits) 19:27, 27 April 2024 (UTC)talk
It may or may not have been the point you intended to make, but it does not miss the point that you actually made. Thryduulf (talk) 21:33, 27 April 2024 (UTC)
I don't see where I said Wikipedia does not need editors with expertise. I said a page should not require editors with such super-specialized expertise that only a tiny handful of people would be qualified to evaluate it. This isn't a controversial statement. JoelleJay (talk) 22:21, 27 April 2024 (UTC)
Except that the editors of Wikipedia does not represent the universe of people who would be interested in a topic. A topic may have plenty of interest to be worthy of a Wikipedia article, but only tiny handful of Wikipedia editors are qualified to evaluate it. Rlendog (talk) 13:56, 29 April 2024 (UTC)
A subtopic that is so complex and niche that only the people who have a conflict of interest with it are capable of summarizing it accurately is clearly too technical to be described on WP. JoelleJay (talk) 01:33, 2 May 2024 (UTC)
I don't disagree if everyone was an active Wikipedia editor. But since most people are not active Wikipedia editors, there are topics that only a few Wikipedia editors are capable enough of summarizing it accurately, and those might have some conflict of interest, especially if interpreted broadly. Rlendog (talk) 00:33, 3 May 2024 (UTC)
This won’t prevent those experts using those non-English sources, unless those non-English sources are written by the expert or their colleagues - and in such a circumstance, it’s worth the additional oversight of the edit request process to make sure they’re adding them in accordance with our core policies and not to promote themselves or their work. BilledMammal (talk) 03:03, 30 April 2024 (UTC)
In my experience, academics are aware of their place in their field's hierarchy, sometimes acutely so. In particular, when their views are rejected or ignored, they know it. They might still think they're right, but they are aware. Consider Katalin Karikó, who won last year's Nobel Prize in Medicine: she'd been demoted, had her pay cut, and was told she wasn't good enough for tenure. There is no way that she would have thought her views were the most dominant in her field. But if were were lucky enough to have her as a Wikipedia editor, having her cite a paper or two of her own would not have been a significant problem.
How would we know if it isn't working out for us? I know at least two people in my department who have added their own findings to numerous articles and even created articles based on their research papers. Wikipedia currently strongly emphasizes primary results that we now know were incorrectly interpreted and misleading, but the paper that recently came out showing some of the deficiencies in their previous findings is also primary and not suitable to use for context. I can't edit the affected articles myself or even suggest edits under this username, so a splashy unreproducible narrative will persist for a topic that itself never would have even been mentioned on here if we had more stringent COI rules/patrolling and stricter discouragement of primary sources. How many other pages reflect a highly distorted perspective on what is relevant, accepted science in a subfield due to undetected self-promotion? JoelleJay (talk) 06:39, 29 April 2024 (UTC)
More stringent rules wouldn't have changed anything. People who aren't following the existing rules can't be counted on to follow other rules.
More stringent patrolling is difficult. If someone edits as "User:AliceExpert" and cites a paper by Alice Expert, that usually gets reverted. If someone edits as "User:InnocentVolunteer", then patrollers don't wonder whether that's Alice, unless our InnocentVolunteer cites multiple papers, or the same paper in multiple articles, all of which were written or co-written by Alice.
I have seen several academics over the years who cite not only their own work, but others as well. I have seen them write articles that say there are two camps, A and not-A, and cite strong sources on both sides. I would not want to give that up. WhatamIdoing (talk) 15:04, 29 April 2024 (UTC)
Can you explain more about why you are proposing defining Editing articles related to one's employer or competitors in the industry to be a Non-financial Conflict of Interest, which differs from the current policy? To be fully transparent, I'm someone who falls into this category and I've declared a Financial Conflict of Interest accordingly. I strongly believe I am not biased toward my employer in this case as, for example, I hold no fear whatsoever that I could be fired or punished by my employer for "disparaging" the business in an article. That being said, I agree with the widely shared sentiment that this is not the case the vast majority of the time. I fully support the current practice of submitting the article to AfC and not touching it afterward, so I don't support language that says not strictly forbidden from directly editing affected articles in this scenario. Mokadoshi (talk) 03:04, 23 April 2024 (UTC)
Fear of retaliation is not the only concern about bias. The above type of editing is also problematic due to what it emphasizes; what might seem important to an employee may not reflect what is important in independent sources. JoelleJay (talk) 14:41, 27 April 2024 (UTC)
That would be highly restrictive for some large companies. Apple, for example, is involved in computing, AI, photography, videography, filmed content production and distribution, telecommunications, cloud computing, music, electronic payment, ... The list goes on and on. isaacl (talk) 02:38, 28 April 2024 (UTC)
Apple isn't involved in every aspect of those topics. An employee could easily edit in those areas without touching material related to Apple. JoelleJay (talk) 06:45, 29 April 2024 (UTC)
Editing aspects of topics unrelated to your employer is not the same standard as not editing pages related to one's employer. Which are you actually proposing? Thryduulf (talk) 07:08, 29 April 2024 (UTC)
If your standard is "do not edit aspects of topics related to your employer" then at least some parts of all those articles would be able to be edited.
The two standards are very clearly not the same, which standard are you proposing we should use?
Do you intend to prohibit an Apple retail assistant from adding third-party sources to the LocalTalk article, because both of your above standards would do that. Thryduulf (talk) 22:48, 29 April 2024 (UTC)
Its "related", not "related, broadly defined". However, I think we can improve the proposed wording and make things clearer by saying "editing content about/related to" rather than "editing pages about/related to"? BilledMammal (talk) 01:34, 30 April 2024 (UTC)
No exceptions, except those explicitly listed here, exist to this policy; edits made in violation of it are indefensible, regardless of their perceived harmlessness or quality. Editors who wish to avoid disclosing their conflict of interest may do so only be avoiding topics affected by it.
Wikipedians in residence
A Wikimedian in Residence (WIR) is a professional role in communications for an organization to share its knowledge within the Wikimedia platform, measure the impact of the same, and promote Wikipedia through training, education, and edit-a-thons. While WIRs can greatly benefit Wikipedia, there is a risk of edits that could unduly benefit their employers. To manage this risk, the following guidelines apply:
Scope of Editing: WIRs may edit articles related to but not directly involving their institution. They must not edit articles where their institution is the primary subject or could reasonably be seen as directly benefiting from the article's content; if they wish to make changes to such articles they must follow the instructions at #Managing Conflicts for Editors with a Financial Conflict.
Disclosure Requirements: WIRs must clearly disclose their role and association with their institution on their user page, as well as in any discussions or edits related to their role as a WIR
I also share ElKevbo's skepticism of trying to handle nuance in a bright line, black-and-white way. Both proposals also seem hasty in light of the lack of consensus in the thread further up for policy elevation in the first place. This seems well intended but misbegotten. More specifically, I'm also troubled by the draft's elision of the policy on harassment. The current guideline reminds editors that COI investigation does not justify harassment. Any policy version of COI should include similar reminders of similar strength. Finally, edits made in violation of it are indefensible seems contrary to our guideline for being patient with newcomers: a new user's ignorance doesn't make COI editing not a violation, but it does make the behavior defensible in the sense that it's a mitigating factor for how we respond to said hypothetical newcomer's COI editing. We educate before expelling. All this reminds me of the reasons Thryduulf and Tryptofish laid out for not supporting elevation to policy at this time, and right now I share their view. Hydrangeans (she/her | talk | edits) 18:27, 22 April 2024 (UTC)
(edit conflict) I was about to write something substantially like Hydrangeans and WhatamIdoing (edit: before their most recent comment) did, but probably less well articulated. Stating "no exceptions exist", even with an accompanying list of exceptions, is naive at best. The real world is complicated and messy, and ill-suited to such inflexible language. For example everybody should be allowed to revert very obvious vandalism, fix obvious typos, and make similar edits that are identical to ones an editor without a COI would always make. Not every organisation has a clear delineation between those with and without official roles within them, and even when they do it's not always the case that someone with an official role is more conflicted than someone without (e.g. a volunteer, unofficial spokesperson for a charity vs the directly employed HR person). This is a good faith attempt, but it is fundamentally flawed. Thryduulf (talk) 00:29, 23 April 2024 (UTC)
No comment on the "Exceptions" section, but regarding the WIR section, I support the overall effort to clarify when a WIR has a conflict of interest. I understand the background to this proposal is due to WIRs not declaring a COI, but from my experience recently I've seen the exact opposite problem: I've seen WIRs submit a ton of COI edit requests in order to make sure everything is above board when they aren't always necessary. I'm not naming names because I am happy they are being so transparent, but my concern is that a WIR creating 20+ COI edit requests take a lot of time to review. Mokadoshi (talk) 03:12, 23 April 2024 (UTC)
I would oppose this. This is a radical change, but more importantly, it is unenforceable and clearly doesn't have consensus. The best example I can give is the one I gave at the COI Arb case. I own AT&T stock. I have edited History of AT&T. Should I now stop? What exactly is "significant" benefit mean when it comes to the dividends I earn on that stock? Significant compared to the average worker where I live, in the Philippines? That wouldn't take much. Significant compared to the average stock broker on Wall Street? That would require a great deal more than I have. Where do you draw the line in defining "significant"? More importantly, how do you enforce it? You don't know if I own 1 share or 1 million shares, and I'm not going to declare it, so should I be blocked for refusing to declare my holdings? "Financial benefit" is so vague, so nebulous, that it doesn't work. Paid editing is simple. Editing for your employer is simple. After that, it gets muddy.
You can't make a black and white policy to "outlaw" COI editing and have it be effective. You CAN look at my edits and say "they look normal" or say "your edits look promotional, full of puffery" and remove those edits based on existing policy, and then block me if it is a pattern. Existing policy covers this. What matters is the CONTENT of my edits and whether they are obviously designed to improve the encyclopedia, or if they appear to serve another purpose. Regardless, forbidding COI editing is just going to guarantee that people with a COI will simply stop declaring the COI, so you will have less transparency, so this would backfire spectacularly. What the COI guideline needs is MORE disclosure, not less. If you treat all COI editors as the enemy, they will respond in kind and simply ignore policy and sock. You are better off engaging and getting them to declare and become invested in their editorship (and reputation) here. Dennis Brown - 2¢03:39, 23 April 2024 (UTC)
It seems neither the extensive rebuttals on Wikipedia nor on Wikipediocracy have convinced you to give up the "I own stock" line of argument. Levivich (talk) 03:49, 23 April 2024 (UTC)
No, because there hasn't been a single successful rebuttal that address the actual circumstance. The real question is "where do we draw the line?", and it serves as an excellent example because it isn't hypothetical and is extremely common. Writing bad policy without considering real world implications is what I'm trying to get the community to avoid, just as I tried to get Arb to avoid. Now, do you have an actual argument against the example, or are you just wasting everyone's time and saying "I don't like this example". Dennis Brown - 2¢07:58, 23 April 2024 (UTC)
All policy and guidelines here are line drawing, and regularly the lines are fuzzy (or nuanced). As presumably (you say you support the guideline) your not saying there is no such thing as conflict of interest, it means even you are aware that lines can be drawn for conflict of interest, indeed probably some are easy to draw. And as for Wikipedians being able to hash out lines, that's most of what Wikipedians do all the time. So, let's say your position is you don't have a conflict, submit your situation to the community, and no doubt they will guide you as best they can. Alanscottwalker (talk) 13:50, 23 April 2024 (UTC)
There's room for refinement but overall I think this is a significant improvement over the current guidelines, especially in terms of readability. The thing is, strategically, you're never going to get consensus to adopt it, because any attempt is going to be blocked by people who oppose the current guideline (see the !vote directly above...). This is a project-wide phenomenon and the reason why we have so many PAGs that everyone agrees are shit but that haven't changed substantially in a decade. But I digress. The upshot is I think it's more realistic to make these changes incrementally. – Joe (talk) 08:13, 23 April 2024 (UTC)
I don't oppose the currently policy, Joe. You've been told again and again by others, YOUR interpretation is not correct and you are very much in the minority, in particular at the last Arb case. I think refinement is a good idea, but refinement that makes people want to lie about any conflict is not in the best interest of the encyclopedia. Dennis Brown - 2¢09:21, 23 April 2024 (UTC)
There is, clearly, room for differing interpretations of the current guideline. I am not interested in battling it out to see whose is "correct". However, I've been active in enforcing COI for nearly a decade, so what I can tell you is that what actually happens (e.g. at WP:COIN) is that when someone has an obvious financial stake in a topic, editors don't fuss around trying to establish whether they've been "paid by an entity for publicity" (because how would they even do that?) before acting to protect the project's integrity. That this practice differs from how some interpret the guideline is a problem that needs to be solved, one way or the other, for sure. But I think all you achieve by going around talking about your AT&T stock is to make people wonder why on earth you think "this can't be paid editing, because Dennis Brown is doing it!" is an argument that would convince anyone except Dennis Brown. – Joe (talk) 10:39, 23 April 2024 (UTC)
There is, clearly, room for differing interpretations of the current guideline. whether that is true or not, there is a very clear consensus that your interpretation is not one that is correct. Thryduulf (talk) 10:44, 23 April 2024 (UTC)
"this can't be paid editing, because Dennis Brown is doing it!" is s strawman argument, and someone would have to be rather simple to believe that is an actual position. If you sincerely believe that my example is paid editing, surely you wouldn't show favoritism, and you would instead block me. Your stance on COI smacks of politicking. Dennis Brown - 2¢12:14, 23 April 2024 (UTC)
Given that you haven't blocked Dennis, it seems safe to assume that you concluded either that he wasn't violating policy or that a consensus of uninvolved administrators would conclude that he was not violating policy. That sounds rather similar to your interpretation of the policy either being different to what you say it is in discussions or, more likely, different from the community consensus interpretation of the policy. Thryduulf (talk) 14:26, 23 April 2024 (UTC)
Or maybe I'm just too much of a coward to apply the same rules that are applied to regular editors to an admin with a tonne of social capital in the midst of a contentious arbitration case about the selfsame issue. – Joe (talk) 14:34, 23 April 2024 (UTC)
Those edits are ones that even an indisputably paid editor should not be sanctioned for making, further demonstrating the "no exceptions" is incorrect. Thryduulf (talk) 15:46, 23 April 2024 (UTC)
Some of us live on the other side of the globe, and slept during most of the buffoonery. The claims of "non-trivial" amounts of stock were absurd, since I never said I owned non-trivial amounts. I was intentionally vague because it isn't anyone's business. The concept of an absurd de minimis example are lost on some. Again, I can only conclude that Joe's reference to blocking me here, and then dragging out an ANI case, are to try to silence me, which won't work on me, but it might work on others, which is what made me question his competency to be an admin. He knew I only had a few minor edits to AT&T, but the value of dragging someone through the mud wasn't to get action on me, it was to silence critics of his unique interpretation. Even above, he seems worried that others will cite my example "this can't be paid editing, because Dennis Brown is doing it!", so it seems the ends justify the means in his eyes. As for being a "coward", that is an amazing display of gaslighting Joe. I'm not trying to silence debate, Joe, but you clearly are and have made it personal. Dennis Brown - 2¢20:17, 23 April 2024 (UTC)
It's gaslighting that my own explanation for why I didn't do something differs from yours? Okay. I didn't know that you only made minor edits to AT&T, because you strongly implied otherwise, and I foolishly took you at your word. Now I am wondering why on earth you brought this up at all, then, given there is already a well-established exception for minor COI edits.
In the mean time, for the crime of daring to take you seriously when you repeatedly asked "is this paid editing?", and so asking for the community's answer to that question, you have:
Accused me of "politicking" and trying to "silence" you[8][9]
I eagerly await the evidence you will be providing for these aspersions and personal attacks. – Joe (talk) 08:00, 24 April 2024 (UTC)
You say "we have a well-established exception for minor COI edits." But the proposed wording says "No exceptions, except those explicitly listed here, exist to this policy; edits made in violation of it are indefensible, regardless of their perceived harmlessness or quality." And the WP:COIADVICE exceptions are not listed in the proposed wording. Rlendog (talk) 14:03, 29 April 2024 (UTC)
I'd flip it: why do we need them? Why do we need someone with undisclosed COI to be able to "fix spelling, grammatical, or markup errors" or "add independent reliable sources when another editor has requested them, although it is better to supply them on the talk page for others to add"? I mean I'm fine with having WP:3RRNO-like "obvious" exceptions sitewide--same exceptions not just for 3RR and WP:INVOLVED but also for TBANs, COI and PAID, but we actually don't have that right now. Here's an interesting wiki policy question: why should certain edits be exempt from COI if they're not exempt from TBANs? Levivich (talk) 04:58, 24 April 2024 (UTC)
I think they're most helpful for editors with declared COI. Our COI requested edit backlog is usually significant, so keeping uncontroversial changes out of it is helpful. Interesting thought experiment about TBANned editors, but they're ones who have already been disruptive in that topic area. The burden of evaluating the acceptability of their edits falls back on those who were affected by the disruption. Firefangledfeathers (talk / contribs) 12:12, 24 April 2024 (UTC)
In what possible world is that is good thing!? We want COI editors to engage with the edit request process, and the best way to do that is to make it responsive - anything else incentivises them to ignore it. When requested edits are good delaying making the changes harms the encyclopaedia, when requested edits are bad leaving them hanging around benefits nobody and indeed makes it harder for the editors making the requests to learn which edit requests get accepted and which don't. If your primary goal is eliminating COI rather than improving the encyclopaedia then you are here for the wrong reasons. Thryduulf (talk) 12:20, 24 April 2024 (UTC)
It is a good thing in the world where we don't want COI editors to edit at all. I understand that you don't agree with that aim, but you must surely acknowledge that many others do, and let's not forget that the nutshell of this guideline—its message stripped of all the ifs and buts—is do not edit Wikipedia in your own interests, nor in the interests of your external relationships. Eliminating COI editing is an important goal for many people, even if it's not our primary one. – Joe (talk) 12:55, 24 April 2024 (UTC)
If we didn't want COI editors to edit at all, we would ban COI editing rather than regulate it. The reason that we don't do that is twofold - first we recognise that in the real world we can't actually ban it, people will just edit without disclosing and the encyclopaedia will be worse off. Secondly we recognise that COI editors can make valuable input that improves the encyclopaedia. I'm increasingly of the opinion that you've become so blinded by your view towards COI editing that you've lost sight of the bigger picture. Thryduulf (talk) 13:17, 24 April 2024 (UTC)
Your use of a collective "we" is a bit misleading here because there is a lot of disagreement about COI and there has been for as long as I can remember. I'm aware that my view is on one of the end of the spectrum—I think yours is too, to be fair—but I'm not alone and I don't think I'm blind. I think this is just another one of those judgement calls that make consensus difficult to achieve: is the risk presented by COI editors worth the value they can add? Is a total prohibition of paid editing something that we could enforce within an acceptable margin for error? I'm more and more realising that one of the major shortcomings of our decision-making model is that we lack mechanisms for resolving disputes like this when different editors bring different and incompatible premises to the discussion. – Joe (talk) 13:28, 24 April 2024 (UTC)
is the risk presented by COI editors worth the value they can add? Yes. Is a total prohibition of paid editing something that we could enforce within an acceptable margin for error? No, but yet again it's worth noting that you are conflating paid editing and COI editing - the former is just a small subset of the latter. I'm also uncertain why we would want to attempt to completely prohibit either, given the harm it would do to the encyclopaedia. Thryduulf (talk) 13:46, 24 April 2024 (UTC)
Conflating them there was a slip of the tongue. I meant to write COI in both cases. But there's no objectively determined answer to either question. Others, like me, answer no and yes. That's the basic problem here. – Joe (talk) 14:02, 24 April 2024 (UTC)
WP:COIU isn't clear as to whether those are examples of edits that editors with a COI can make if they disclose the COI, or if they're examples of edits that editors with a COI can make without disclosing the COI. I think it's the latter? I don't really have a problem with the former, but I do have a problem with the latter. Because why not make the same exceptions for UPE, or even socking? The reason why not is that it'd be next to impossible to police, and rife with abuse if it was on the honor system. "I'll make an undisclosed alt account but only use it to revert obvious vandalism and BLP vios," or "I'll UPE but only take payment for edits that are reverting obvious vandalism and BLP vios" wouldn't fly with anyone, and I don't think it should fly for undisclosed COI. I wouldn't mind those exceptions for disclosed COI, but even then, COIU goes too far IMO, insofar as it goes beyond self-reverting, vandalism, and BLP. Levivich (talk) 15:21, 24 April 2024 (UTC)
I understand where this is coming from, but I tend to think that simpler rules are easier to follow. If we have fewer exceptions people know what is expected of them - don't edit articles where you are getting paid; disclose before making edits where you are not paid but do have a COI. The only real emergency situations which I see as exceptions are blatant vandalism and BLP violations. Everything else can wait.
It is why I like 3RR. I know that if I make that third revert I better be convinced that it is a BLP violation or absolutely blantant vandalism. I need to mentally check myself to make sure it is definitly ok, because if I am not sure then it probably isn't an emergency. - Bilby (talk) 06:00, 24 April 2024 (UTC)
True enough. But I think of it as "don't revert unless you are absolutely convinced it is ok". I like that approach. :) And when you look at the list, it is actually pretty sensible - you can revert yourself (ok, that makes sense), you can revert on your own talk page (again makes sense), but otherwise it is serious policy concerns. I think we could use a similar list here. If it isn't a serious policy issue (BLP violation, obvious vandalism, child pornography), do not edit the article with a COI. Makes it easy to follow. - Bilby (talk) 12:37, 24 April 2024 (UTC)
My views are not aligned with most other Wikimedians in Residence so what I am about to say is my individual and personal opinion. I think for most cases the primary problem is not COI versus non-COI, but rather marketing versus non-marketing. If we defined the taboo as being paid to edit biographies, organizations, and products then I think that hits the target of problematic editing 99% of cases. I would like to support, for example, paid public health communication professionals in developing medical articles about medical conditions with advice on the level of government and professional society recommendations. I sponsored development of the general topic of Software maintenance (before, after), currently at Good Article review because we as a research department care about certain aspects of software development. I have biased ideologies here - advocacy for open-source software mostly - but if we wanted a simple rule for preventing most problems, I think it should separate specific commercial interest editing versus broad interest Wikipedia articles and cited sources that are unlikely to give benefit to any specific interest. I especially want to accept paid editing when it improves Wikipedia:Vital articles in such a way that skeptical review would not identify particular bias. A lot of organizations invest a lot of money in trying to communicate general reference information and I think Wikipedia is losing out by not having pathways to accept some of that. I am not sure how to draw the lines but universities are knowledge centers and right now, there are not good paid editing pathways for universities to invest in Wikipedia for public education campaigns. Bluerasberry (talk)18:30, 23 May 2024 (UTC)
How I see it
This is how I see the spirit of existing policy, worded differently and perhaps defined a little more clearly:
If you are paid to directly edit on behalf of a company, or you are paid to promote a company's interests (social media, marketing, advertising) and you edit the article of that company or create edits that mention that company or their products (broadly construed), then you are a paid editor and must fully disclose your relationship, even if your edits are trivial, or are limited to talk pages. How much you are paid is not relevant. Adminship is not consistent with paid editing.
If you own at least 10% of a company, or derive at least 10% of your income from a company (including stocks), but you have have no official role in promoting that company, you have a clear conflict of interest when it comes to that company. It is strongly preferred that you don't directly make significant edits to the content of any article that is directly related to that company, and instead make suggestions on the talk page of those articles. If you are making direct edits to those articles, then disclosure on your user page is required. If you are only making edits to the talk pages of those articles, disclosure is strongly recommended. The 10% threshold isn't a brightline rule, it is a common sense rule of thumb, to say that any company that provides an important source of your income, even if indirectly, should be considered as a conflict when it comes to editing. While direct editing is not forbidden, it is discouraged and you should expect those edits to be closely scrutinized. You are required to be forthcoming in explaining your direct edits if questioned in good faith.
If you work for a company/organization in a non-promotional role, or own a small percentage of a company (either stocks, employee owned businesses or similar), or volunteer in a significant way for any organization, even if you aren't being paid, you might have a simple conflict of interest when it comes to editing articles related to that company. Great care should be taken when editing those articles, or it should be avoided altogether. While disclosure isn't required, your relationship may be questioned if the content of your edits appear to be advocacy. In some cases, you are better off using the article talk pages to request edits instead of direct editing.
This all makes sense. I would add that, for the benefit of the encyclopedia, we want people to edit what they're interested in. We need subject matter experts. We want doctors to edit about medicine, lawyers to edit about law, biologists to edit about biology, mathematicians to edit about mathematics. Our COI policies need to allow this kind of editing. Clearly our policy should discourage lawyers from trying to create articles about their own law firms and doctors from trying to promote the niche surgery they happen to specialize in. But we'd be shooting ourselves in the foot if we only allowed people to edit articles they have absolutely zero affiliation with. ~Awilley (talk)00:05, 24 April 2024 (UTC)
Good idea. It seeks to address a very problematic area regarding this behavioral guideline. I could quibble that it seeks to define it is a "yes/no" situation in an area which is really matters of degree, but such is the reality of trying to move forward. North8000 (talk) 00:40, 24 April 2024 (UTC)
Our COI policies do allow this kind of editing, quiteexplicitly. Literally the only time it comes up is when anyone suggests maybe making this 4000 word guideline a bit clearer – then it's suddenly crucial that we discuss WiRs and academics and Dennis 'AT&T' Brown and all sorts of other pointless whataboutisms that bear absolutely no relation to the day-to-day efforts of volunteers trying to stop the project being overrun by commissioned articles, PR departments, and self-serving autobiographies. – Joe (talk) 08:18, 24 April 2024 (UTC)
you just can't resist can you? You have to insult when others don't feel like it's as pointless as you do. It's not enough to say you disagree, you have to try to discredit those that disagree with you. Farmer Brown - 2¢ (alt: Dennis Brown) 01:03, 30 April 2024 (UTC)
pointless whataboutisms that bear absolutely no relation to the day-to-day efforts of volunteers trying to stop the project being overrun by commissioned articles [etc]. Except they are not "pointless" nor "whataboutisms", indeed they are directly related to the policies and guidelines that are the basis of the activities you describe - we are attempting to make sure it is fit for purpose and excludes from the encyclopaedia those things that should, by consensus, be excluded but doesn't exclude those things that should, by consensus, not be excluded. It's equally harmful to the encyclopaedia for it to be overrun by non-notable articles and for it exclude notable articles for fear of being overrun by non-notable articles. Thryduulf (talk) 02:27, 30 April 2024 (UTC)
I have been following this thread, and I fully agree with what is being said here. In my experience, the COI tag has been thrown around to "discredit" an authors edits to an article, even if those edits are in no way promotional or against policies. I do not see the benefit in putting a COI tag on an article (unless it is a clear paid editing) unless the editor is adding unnecessary promotional material or editing against policies.
I just noticed that the proposal used the term "this policy". Unless that's typo or I missed something, proposing making this whole bundle into a policy would make me a strong "oppose" This has such a wide range of what could be considered a COI, and is so open to abuse and often abused to get the upper hand in a dispute (including McCarthy-esque drilling of people) that upgrading it to a policy would be harmful. Sincerely, North8000 (talk) 01:50, 29 April 2024 (UTC)
Extended content
I think this was the spirit of policy c. 2012. Over the years consensus has shifted towards a broader definition of paid editing. The clearest way this is evidenced is the introduction of the term "financial conflict of interest", which refers to something more serious than a "simple conflict of interest" but isn't as narrow as "paid advocacy" (the ToU's term), and this was present by 2015.[10] It is notably absent from your summary here.
The difference between owning 5% of a company and 15% of a company may be obvious to the kind of person that invests in stocks, but I think expecting it to be self-evident to most people is a failure to recognise your own privileged viewpoint. Similarly, 10% of the median income where I live is USD$2000. Are we really prepared to say that sums less than that are insigificant? For comparison, the going rate for creating a Wikipedia page on freelancer sites is about $500-1000. Or another way to say "10% of your income", for working people, is "more than a month's salary". – Joe (talk) 08:13, 24 April 2024 (UTC)
The average US wage is $53,383, median is $46,625, not $20,000 (source: Wikipedia), so where you live (your example) is much lower than the US average, yet 3 to 5x higher than many places. It's all relative, which is why percentages work and dollar figures don't. The going rate for someone to edit a Wikipedia page is not $500-1000, it is significantly less because most of it is coming from Nigeria, the Philippines, Pakistan, and India, places that some of the highest numbers of English speakers. It is driven by both poverty and greed, and it makes the bulk of paid editing because it is the cheapest, $50 or less. But my opinions above weren't really about paid editing, they concerned the different types of lesser issues, conflict of interest without being a paid editor.
One example would be someone who tries to edit the Bitcoin article to make Bitcoin look more attractive, because they want to dump some coin. This has happened. This poses a serious threat to article neutrality. Another example is the Starbucks manager, who edits to reflect that the newest, most popular drink new drink is some new flavor of latte. This is still a conflict, but doesn't really affect neutrality and is more benign. Both should be reverted, but they have different effects on the encyclopedia and are inherently different.
Claiming this view is "2012" is a bit silly, Joe. Relatively little paid editing was taking place then. I know because I spent most of 2013 and 2013 working paid cases as an SPI clerk, so I saw it every day. The amount of COI and paid editing dwarfs what took place then, as does the breadth of it. It is clearly more of a problem now, which is why most people recognize there are different types of conflict that need to be dealt with differently. This includes recognizing COI at the lowest level and addressing it, and encouraging people with COI to do what is best for the encyclopedia, instead of slapping them down and pushing them to the shadows, or simply running off editors who do mostly good work but also have a COI.
As for 5% or 10% or 15%, you have to draw a line somewhere. If I was earning 20% of my income in dividends, I would instictively not edit the article, or declare I have a conflict of interest. If I am earning a few percent, there is no need because it is not enough to financially benefit me, relatively speaking. 10% is strict enough, even if arbitrary. You are welcome to suggest a different line in the sand.
And saying "failure to recognise your own privileged viewpoint" is bordering on ad hominem. I live in a developing country (Davao, Philippines), working remotely part time, with water and electricty that doesn't always work. I don't own a home or a car and either walk or take jeepneys or tricycles everywhere I go. I'm not complaining, but you know nothing of my situation or lifestyle, and likely nothing about my education or background. You are not in a position to judge how "privileged" I am, nor should you be judging anyone here as "privileged".
I've laid my cards on the table, opened my ideas for debate, spelled out in plain English. It seems obvious that you and I are fundamentally different types of people. and I am not afraid of having my ideas, my history, or my conduct examined. If you have a better idea, then by all means, present it and lets discuss it. But lets stop the baseless AN reports, baseless quotes, the "2012 thinking", the "privileged" claims and other failed attempts at intimidation, because it doesn't work with me. How about we actually talk about concrete ideas. Dennis Brown - 2¢09:21, 24 April 2024 (UTC)
Right... I don't live in the US. What's your point? Although you're right I did get the wrong figure. The median annual income in Denmark is actually USD$50000. Which rather underlines my point that, whether you live here or in the US, 10% of your income is an extremely high threshold to propose triggering a "close financial relationship", one which we apparently both agree would exclude the vast majority of real, documented articles written for pay. In my opinion, any "line in the sand" would be too high: if there's any possibility that you'll receive money as a result of your editing, you shouldn't be doing it, full stop.
I've probably investigated hundreds of paid editing cases. The going rate is certainly not $50 – maybe as an hourly rate. To verify this, make an account on upwork.com, search for freelancers with the keyword "Wikipedia", and look at their completed job history. Prices are variable but it's ~$200 minimum for a new article and anyone who's good enough at socking to actually be able to do it is in the $500-1000 range.
I think we're well past the point of diminishing returns in this conversation. You seem convinced that "how you see it" is very important, but you working SPI over a decade ago is the first I've heard of you actively working in this area (please correct me if I'm wrong), and your supposedly genuine concern that overzealous COI enforcement would see you persecuted for holding AT&T stock turned out to be complete bullshit, because you've never actually done any editing that would make it relevant. You also seem convinced that I'm out to get you, even though I don't recall us having a single conversation before yesterday. If you're "not afraid of having my ideas, my history, or my conduct examined", why has the mere prospect of that happening caused you to fly off the handle and start seeing conspiracies to silence and intimidate you?
I don't care where you live or how many wheels you use to get about and I can't imagine why you think anybody would. I'm sorry you take exception to the implication that owning a significant amount of stock in large corporations is a form of privilege, but you've also just said "a few percent" of your income (so $460-$1500 annually, in terms of the figures above) is "not enough to financially benefit me", which kind of does say something. If you are upset about comments "bordering on ad hominem", please take the time to review the comments I've listed above, which are a long way away from any borders. – Joe (talk) 12:09, 24 April 2024 (UTC)
You've done everything except propose a better solution, Joe. Are you "after me"? You made a veiled threat to block me, you dragged me to WP:AN, and you continue to be insulting, your arrogance is dripping in how your experience trumps the consensus of the community. You keep holding to the fantasy that policy outlaws paid editing when it doesn't. I've offered an interpretation with substance, in detail. You've only tried to intimidate me, but you can't. You keep trying to paint me as privileged when it isn't your job or anyone else's to judge me. And you see nothing wrong with this. I stand by what I said in the failed WP:AN report, you are not fit to be admin because of your actions in these discussions. Any admin that (multiple times) tries to intimidate any editor into not participating isn't fit for the bit. So put up or shut up, and offer a an solution instead of trying to bully people. Dennis Brown - 2¢12:29, 24 April 2024 (UTC)
And I didn't say a few percent of my income was derived. I said if that is what it is, then I wouldn't consider it enough to trigger a required disclosure, just as 20% obviously would trigger it. For someone who investigates at COIN, your reading comprehension seems lacking, or simply lazy. So quit mischaracterizing what I am saying. Dennis Brown - 2¢12:34, 24 April 2024 (UTC)
Mate, you asked me why I hadn't blocked you. Answering that question honestly is not a threat, veiled or not. In contrast, we can now add a third call for me to be desysopped and a second and third baseless accusation of intimidation to your tally over the last 30 hours. And I'm the "bully"? – Joe (talk) 12:47, 24 April 2024 (UTC)
No, you've used your position and the processes here to derail a valid discussion and intimidate myself and others. I haven't called for your bit, I just pointed out the fact that you're unfit to have it and have given specific examples as to why. Your gaslighting is very transparent. Farmer Brown - 2¢ (alt: Dennis Brown) 22:47, 24 April 2024 (UTC)
Visibility of external advertising accounts
About two weeks ago, I submitted this proposal on Meta-Wiki seeking consensus on additional requirements. It has been observed in multiple cases that, after disclosing their off-wiki profile link, paid editors then protect their profile from public view, making it challenging to review their contributions. I believe the main purpose of the existing requirements is to facilitate the review process, so if a profile isn't publicly accessible, the disclosure is useless. Therefore, I invite everyone to share their thoughts. Thank you. GSS💬04:59, 22 June 2024 (UTC)
