Wikipedia talk:Requests for comment/COI

Project banner
A discussion about downgrading or deleting a guideline is fairly serious, so I suggest getting this RfC listed in the "Watchlist options and notices" box, as well as posted in the Signpost, Village Pump, Jimbo's talk page, and wherever else is kind of a community noticeboard. Cla68 (talk) 01:24, 27 February 2012 (UTC)
 * Jimbo's talk page. Cla68 (talk) 01:26, 27 February 2012 (UTC)
 * Watchlist notice. Cla68 (talk) 02:06, 27 February 2012 (UTC)
 * Added WP:RFC/COI as a shortcut, and added to WP:CENT to promote visibility. Jclemens (talk) 04:43, 27 February 2012 (UTC)
 * For completeness, it was in the Signpost here. --Philosopher Let us reason together. 11:26, 8 March 2012 (UTC)
 * This RFC was not easy to find. It was archived April 24 at Centralized discussion, although it has not been closed. It does not appear to be listed at Requests for comment/All. Was it previously posted there? Edison (talk) 15:22, 27 April 2012 (UTC)

Anything other than harassment?
What is COI used for except harassment? Be— —Critical 15:44, 27 February 2012 (UTC)
 * The term "COI" is quickly easily understood by business people and requires very little explanation. I see it frequently used as a way to remind editors to consider or re-consider the neutrality of their edits. Rklawton (talk) 15:57, 27 February 2012 (UTC)


 * So, anyone mentioning COI to a person is, by definition, harassing, Becritical? --Dirk Beetstra T  C 16:04, 27 February 2012 (UTC)
 * Rule enforcement of any sort involves a certain amount of pressure to conform (i.e. harassment). Rklawton (talk) 16:07, 27 February 2012 (UTC)


 * I am simply saying that I have never seen COI used in a way that was not harassment.  Telling people that a COI can be a problem is one thing, but the current guideline either needs deleting, or it needs a total rewrite to make it totally clear that COI is not the problem, non-NPOV editing is the problem, and harassment is not allowed on the basis of COI.  Cut it down to a paragraph or two that says that COI editing can be a problem, then merge it to NPOV.  Why does it exist except harassment?


 * Telling another editor that they should not edit because of a COI is considered harassment, and editors may be sanctioned if they use this guideline to harass rather than simply to inform editors that COI can lead to Wikipedia policy violations. Or something like that. Be— —Critical  16:13, 27 February 2012 (UTC)
 * But it says that, in the very first paragraph: Where advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest. - if there is no 'advancing outside interests', there is no COI, and the rest does not apply. If it does, the rest does apply.  --Dirk Beetstra T  C 16:27, 27 February 2012 (UTC)
 * Regarding your bolded sentence, Becritical, yes, you are right. But that is not a reason to delete a policy or guideline.  If I start harassing people with WP:COPYRIGHT where that is simply untrue, then that does not mean WP:COPYRIGHT needs deleting, it means that I need to be .. talked to.  And believe me, a lot of people are harassed using WP:COPYRIGHT.  --Dirk Beetstra T  C 16:29, 27 February 2012 (UTC)
 * NPOV is one of Wikipedia's five pillars. COI is one of many possible causes of POV edits. It makes for a good essay, but nothing more. Rklawton (talk) 16:46, 27 February 2012 (UTC)
 * True, Rklawton. This is a matter not of whether the things in COI are true or false, but in how it is used.  If you tell me one instance of legitimate use of COI, I'll tell you ten instances of abuse.  That is a very legitimate argument to drastically rewrite or downgrade the guideline.  Be— —Critical  06:17, 28 February 2012 (UTC)
 * User:Ego records, likely has a conflict of interest with the artists they represent, should take care when editing pages on artists they are representing, and with pages on their own company.
 * User:Onjohn, likely has a conflict of interest with the organisation they represent, should take care when editing subjects they are involved with.
 * User:Orchidinnsb, likely has a conflict of interest with the Bed and Breakfast they own, should take care when editing the page on the bed and breakfast.
 * User:Handheld.culture, likely has a conflict of interest with the company they represent, should take care when editing subjects they are involved with.
 * User:GregMurhy51, likely has a conflict of interest with himself, should take care when editing his own page.
 * I wonder how you would see that notifying these editors that they have a conflict of interest is harassing, I would call it legitimate actually to notify them of that. I'd like to see 50 examples of harassing, please.  --Dirk Beetstra T  C 06:44, 28 February 2012 (UTC)
 * Upping the ante, make it at least a 100:
 * User:Nchenderson, likely has a conflict of interest with the page on himself
 * User:UGMMDI, likely has a conflict of interest with the organisation they represent
 * User:Hyundai finance, likely has a conflict of interest with the companies they represent
 * User:Kristian Donaldson, likely has a conflict of interest with the page on themselves
 * User:Junoreactor2008, likely has a conflict of interest with the page on the organisation (?) they represent
 * User:Wallykennedy, likely has a conflict of interest with the page on themselves
 * User:Folktronica, likely has a conflict of interest with the group (?) they represent
 * User:Overview institute, likely has a conflict of interest with the subjects they represent
 * I understand, you could also notify these editors that they should adhere to WP:NPOV etc., but I do not see how it is harassment if we notify these editors that, because they (likely) have a conflict of interest, that they take due care with their edits. --Dirk Beetstra T  C 06:58, 28 February 2012 (UTC)

Reccomend discussions takes places at RfC not here on talk, unless there's a reason I don't know about. Babakathy (talk) 07:18, 28 February 2012 (UTC)

Organization of responses
Would it be acceptable to group the various views roughly according to the changes proposed? Right now I find it too hard to sort out whether there is any kind of consensus on what ought to be done. Mangoe (talk) 16:23, 2 March 2012 (UTC)
 * I second this proposal. We need to be able to cast one vote to either Support or Oppose the elimination or reduction in scope/power of WP:COI and the associate noticeboard.  Thanks.  Ebikeguy (talk) 16:29, 2 March 2012 (UTC)
 * I think that many of the views are more nuanced than "eliminate", "reduce", "expand" the power of WP:COI and the noticeboard. First Light (talk) 02:24, 4 March 2012 (UTC)
 * I find trying to categorize the current views in a simplistic manner is likely to be counter-productive to the stated goal of the RfC. Why not allow a reasonable period of time for additional views and comments to be weighed, and then examine, say, the ones which have been substantially supported. Collect (talk) 02:27, 4 March 2012 (UTC)
 * Or substantially opposed. ```Buster Seven   Talk  14:29, 8 March 2012 (UTC)
 * (ec) Frankly, I think things are so intertwined that we'll need a second, well-crafted follow on RfC that asks specific questions about the key decision points raised here, and focuses on things highlighted by multiple participants here, to eliminate the discussion on proposals that gain no appreciable traction in this first phase. Jclemens (talk) 02:28, 4 March 2012 (UTC)
 * I question whether anyone can pick out any key points from this, as disorganized as it is. Mangoe (talk) 05:35, 4 March 2012 (UTC)
 * I think that's just as much a function of how many problems there are with the COI guideline and its enforcement mechanisms. It's a tangled mess, along with editor's views. I also think there needs to be another, or parallel, RfC which asks questions such as: what constitutes a COI, how should it be enforced, should declared COI be voluntary, if not then how are investigations to be done, etc. First Light (talk) 07:12, 4 March 2012 (UTC)
 * This is a useful brainstorming session, but I think the variety of responses and their qualifications show that there is not particularly strong general support for demoting COI or closing COI/N. That will help inform the crafting of a future RfC.  That process might include several specific questions: 1) Should WP:COI be demoted to essay status; 2) Should COI/N be closed; 3) Should COI editors be required to declare their conflict of interest; 4) Should paid editors be required to declare their conflict of interest; 5) Should paid editors be allowed to make substantial edits to articles if they have declared their conflict of interest; 6) Should paid editors be allowed to make substantial edits to articles if they seek uninvolved editors to review their changes; 7) Should COI editors be judged solely on the basis of their contributions rather than their background. Ocaasit 15:52, 8 March 2012 (UTC)
 * One more suggestion for the follow up RfC:- Should the COI editors be classified into different categories? -- Anbu121 ( talk me ) 11:50, 10 March 2012 (UTC)

What counts most is the "support" column, as it is quite easy to overweight those who "oppose" a substantial number of views -- then once we list those with substantial support, we should examine where the agreements and disagreements are in order to clearly delineate the issues. Collect (talk) 14:42, 8 March 2012 (UTC)

Follow-up RfC
I think we should start planning a more organized, concrete, and formal RfC to clarify the issues discussed here and bring some of them to consensus. Here's my initial thought for what it should include:
 * 1) Should WP:COI be demoted to essay status
 * 2) Should WP:COI be promoted to policy status
 * 3) Should COI/N be closed
 * 4) Should all COI editors be required to declare their conflict of interest
 * 5) Should paid editors be required to declare their conflict of interest
 * 6) Should paid editors be allowed to make substantial edits to articles if they have declared their conflict of interest
 * 7) Should paid editors be allowed to make substantial edits to articles if they seek uninvolved editors to review their changes
 * 8) Should the COI editors be classified into different categories? (paid, corporate, member, general, etc.)
 * 9) Should COI editors be judged solely on the basis of their contributions rather than their background.

Feel free to make suggestions for other questions or clarifications to this list, or change it yourself :) Ocaasit 14:25, 10 March 2012 (UTC)


 * My suggested additions;
 * Should WP:COI be promoted to policy status
 * Should paid political operatives be required by policy to self-identify whenever editing a political article
 * ```Buster Seven   Talk  16:41, 10 March 2012 (UTC)
 * I added the first one. The 4th and 5th cover disclosure.  Can you briefly explain why you think paid political operatives should get a separate question?  And are there any other categories besides 'all COI editors', 'paid editors', and 'paid political operatives' (for example: paid public relations professionals, owners or employees of companies, freelance paid article writers, etc.)? Ocaasit 19:20, 10 March 2012 (UTC)

Except in cases where the editors are paid by Wikipedia or through some sort of acceptable collaboration with Wikipedia, paid editors shouldn't be allowed at all, and it's horrifying that you're acting as if that wasn't already policy when framing your questions. We are not a place for PR firms to advertise their clients, and the moment we say we are, we may as well shut Wikipedia down, as it's completely lost its basic principles. 86.** IP (talk) 23:38, 27 April 2012 (UTC)

List of identified problems with COI

 * 1) Can easily be used to harass.
 * 2) Has only one circumstance where complaints about COI editing are legitimate: actual non-NPOV edits.
 * 3) Discourages dishonest editors from revealing a COI
 * 4) Encourages honest editors to reveal a COI, which may then expose them to harassment.
 * 5) Discourages COI editors from editing at all ("COI editing is strongly discouraged"), when in fact their contributions are usually good for the encyclopedia.
 * 6) Tends to discourage experts from editing.
 * 7) Makes a big deal about COI, when advocacy is a much bigger problem with the very same symptoms.
 * 8) Generally "criminalizes" COI, partly by use of language ("Disadvantages of COI editing on the sly") rather than advising, cultivating and encouraging COI editors, which would be more in the vaunted spirit of Wikipedia.

Please add to the above list. Be— —Critical 20:13, 10 March 2012 (UTC)
 * I'm not sure what your goal is in writing this list. Is this your third "View By?" Perhaps it would be better placed on the RfC for discussion, rather than creating another location for discussion here. --TeaDrinker (talk) 20:42, 10 March 2012 (UTC)


 * Where do you want to summarize the problems found in the project page? My goal is clear thinking about a problem in Wikipedia.  Be— —Critical  20:49, 10 March 2012 (UTC)


 * I am puzzled by your proximate goal in placing this here, and not on the main page. I am not at all puzzled by your ultimate goal of fruitful discussion--a goal, incidentally, I share.   It may be important to distinguish between your conclusions and what other people think.  I appreciate your attempt to clarify your position, of course, but I just think discussion should be centralized and consistent in format.  --TeaDrinker (talk) 21:01, 10 March 2012 (UTC)
 * I want people to help make a list of problems we've found with the COI guideline. Be— —Critical  21:08, 10 March 2012 (UTC)
 * If your intending to synthesize from the discussion, I suggest my first move would to replace your list with "COI is lacks sufficiently strong wording." But that's my view and not yours. Maybe we should stick with the usual discussion forum for the time being, to resolve these differences.  --TeaDrinker (talk) 21:11, 10 March 2012 (UTC)
 * I agree that this is a useful list, but I don't know what it's doing on this page. Please put it on the main discussion page where more people will see it, and in the proper context. Ocaasit 12:38, 11 March 2012 (UTC)

I strongly disagree: Paid editing means that spotting NPOV can be difficult. If I source something to an obscure 1976 magazine, and I'm a paid editor, I can be horribly POV - and good luck proving it. 86.** IP (talk) 01:14, 28 April 2012 (UTC)

Time to close?
So is it time to ask at WP:AN for an uninvolved admin to close this? Jclemens (talk) 01:11, 21 April 2012 (UTC)
 * Would you object if I hired someone to do it? ;) --TeaDrinker (talk) 01:41, 21 April 2012 (UTC)
 * If you hired someone to do it, I suspect the RfC would collapse under the weight of the irony. :-) Jclemens (talk) 01:44, 21 April 2012 (UTC)
 * I think a closure would be appropriate. Two months seems a sufficient amount of time to gauge the consensus. Regards, Whenaxis talk (contribs) DR goes to Wikimania! 23:50, 24 April 2012 (UTC)

✅. Asked on WP:AN. Whenaxis talk (contribs) DR goes to Wikimania! 23:56, 24 April 2012 (UTC)
 * Leave it open for a couple of weeks. It was not very prominently made known, and some renewed prominence was granted to it by recent dramah at WP:ANI. Let there be no haste to close this very complex RFC. Edison (talk) 04:08, 26 April 2012 (UTC)
 * Or close it, without claiming that there is a clear consensus, and launch a more focussed followup RFC, with better defined alternatives from the positions which garnered the most support. Edison (talk) 15:26, 27 April 2012 (UTC)
 * Leave open AND follow up with a more focused RFC afterwards as Edison suggested. Why can't we have both? The cake doesn't taste as good unless you eat it too! This issue has long been confusing to many and the learning curve on the issue may be improved drastically. I think that alone is worth the discussion!--Amadscientist (talk) 22:27, 28 April 2012 (UTC)
 * Congrats to those who stuck it out. Still watching but not reading. What's the status? Any relevant changes on WP:COI from the actual discussion. Just wondering. Don't want to get go over right now and end up in another discussion don't have time for :-( CarolMooreDC 19:59, 12 May 2012 (UTC)

Hello? Just wondered if anything is likely to happen as a result of all of this? Any prospect of closure, next stage, etc.? Victor Yus (talk) 08:19, 8 June 2012 (UTC)
 * No answer after a month, so I did it myself. No objection if people want to undo my close and get an admin to do it instead, as long as it doesn't then sit around gathering dust for another unknown number of months. Victor Yus (talk) 12:35, 3 July 2012 (UTC)


 * I concur with the general message of Victor Yus's closure. Nothing has changed, the rules stand as they did before the Rfc. I might go a bit further in saying that there were various consensuses that the rules should be kept as they are, but that these consensuses might be interpreted as contradicting each other.  The arbitration committee asked for advice on interpreting the various rules involved, and they have at least some answer: the rules are unclear but there is no consensus to change them.  My interpretation of this is that the Arb Committee must continue to enforce WP:COI in the same way that they enforce any other guideline. If they want further advice, they should probably state their own interpretation as clearly as possible, and then organize a more organized RfC.  I would also suggest that Arbs not participate directly in any new RfC, if they really want the community's advice, but rather state their own interpretation up-front, make sure the Rfc is on a limited, well-defined, set of questions, and then stand back and pay attention to what is being said.  Smallbones (talk) 13:39, 3 July 2012 (UTC)
 * I agree with that; if you want meaningful answers, you have to ask meaningful questions. Victor Yus (talk) 08:08, 4 July 2012 (UTC)

This may sound like I'm conflicting with myself but I'm not. About 95% of the material in the policy guideline is good. But it has very severe problems. It is logically/structurally a disaster. It starts out with an excellent definition of a COI, and then the rest of the guideline is structured to be in conflict with that definition. Also the other 5% sounds like a combination of severe threats and a witch hunt. If those severe problems were fixed, (in particular, bringing the article in line with the definition in the lead) you would have a much more usable foundation to resolve the issues of the moment. It would be understood that most of the situations that folks are calling "COI" are actually potential COI situations, or situations which present a high risk of COI editing. Once the immediate mistaken pejorative /scarlet letter branding of everybody with a potential COI as "COI" is ceased, it would be much more reasonable & doable to ask for and receive disclosure. And, with disclosure, the natural Wikipedian processes would 95% take care of COI editing. North8000 (talk) 16:05, 4 July 2012 (UTC)
 * I presume when you say policy, you mean guideline? I did point all this out on the guideline's talk page at one time, but the eternal ongoingness of the RfC gave people an excuse to reject without discussion any suggested changes. However, if the result of the RfC is now accepted by everyone, then it might be possible (among other things) to get back to rewriting the guideline. One option would be to bring the whole page into line with the definition in the lead, as you say (though first the lead should be rewritten to replace the three conflicting definitions that it currently consists of with just one, saying what we really mean); the second option that has been suggested is that we first bring our definition of COI into line with the real world's definition (which is said to be essentially different from ours) and then bring the guideline into line with that; a third option might be to express the content of the page without using the more-trouble-than-it's-worth phrase "conflict of interest" at all. Victor Yus (talk) 08:01, 5 July 2012 (UTC)
 * Yes, I should have said guideline, not policy. And good points. And, just clarifying, I meant that the bolded definition in the lead is the good one.  North8000 (talk) 10:55, 5 July 2012 (UTC)
 * So do you think, then, that we should be using the phrase COI to mean, effectively, any kind of manifestly biased editing, regardless of whether there's a personal affiliation involved? (Incidentally, I'm informed on my talk page that an admin review of my closure has been requested - that's fine with me, I just mention it here in case anyone else is thinking of doing something similar.) Victor Yus (talk) 07:15, 6 July 2012 (UTC)
 * I think that the bolded definition at the beginning of the policy says it more perfectly than anything I could say here:  "When advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest". And we should say that the ultimate goal of that guideline and others is to avoid that. The immense range of challenges, complexities, quandaries etc. we encounter regarding recognizing, evaluating and implementing the above (starting with "we can't see inside of an editor's head") should be recognized as being merely such, and not as a reason to change the ultimate definition and goal.  This includes recognizing that "COI" may not be the best moniker  & method for dealing with many situations where there is no (known or unknown) specific personal affiliation involved. The operative clauses of wp:coi would continue to relate more to situations involving specific affiliations.  An example of this is biased editing (only) to pursue an ideology.   The definition would still define that as a COI, the stated goals would still be to avoid that, but the operative clauses and day-to-day implementation of wp:coi would not go there. North8000 (talk) 10:33, 6 July 2012 (UTC)
 * I'm not quite sure what you mean. Why would we define COI to encompass some class of situations, but then make the guideline called "WP:COI" explain how to deal with only a subclass of those situations? (Or is that not it?) (Also it has been pointed out that a COI as defined here is not necessarily something to avoid per se; an editor might have quite different motivations from "our" motivations, but their actions might still turn out to be desirable from our point of view just as much as from theirs.) Victor Yus (talk) 11:14, 6 July 2012 (UTC)
 * I tend to divide policies/guidelines into two types of statements. One covers goals and intentions.  The other is the operative clauses.  The former sets goals and guidance for high-minded people and has some general influence, but when you get to specific enforcement, the only "teeth" of the guideline, and then only things that do anything are the operative clauses. What I'm proposing (which, by the way, leaves the current guideline, 95% as-is) is to leave the main definition as-is, add a goal statement which is to avoid that, and then have operative clauses which cover only situations of specific affiliations. North8000 (talk) 11:48, 6 July 2012 (UTC)


 * On another point, there is a flaw in asserting that the "an editor might have quite different motivations from "our" motivations, but their actions might still turn out to be desirable from our point of view just as much as from theirs" situation refutes the broad definition and proposed goal. The definition says conflicting "with the aims of Wikipedia" ; the example given is essentially a group of editors overreaching to assert that "their" view of the "aims of Wikipedia" is THE view and attempting an enforcement action based only on that view.  Such is a faulty example at several levels and thus does not support that argument. North8000 (talk) 11:58, 6 July 2012 (UTC)
 * I don't think so; what I had in mind was something like this (and this is just filling out my understanding of what someone else said elsewhere). Editor A wants to advance the career of Singer Y, who is notable by WP's standards (but has no article). Editor A knows what a tenable WP article ought to be like, and creates one about Singer Y. Editor A doesn't give a fig about "the aims of Wikipedia" (whatever they may be), and so, according to your definition, is in a conflict of interest situation. And yet, Editor A has just made a positive contribution to Wikipedia. So what's to avoid? (I know this situation is probably not typical, and some large percentage of the time Editor A will just do a crap promotional speel and then waste everyone's time fighting to retain it, but the existing consensus seems to be that the problem only arises when the editor does do something problematic, not when he is merely known to have a motivation that might cause him to do something problematic.) Victor Yus (talk) 11:03, 7 July 2012 (UTC)
 * To me, the answer to that is easy. Using that main definition, nothing in the situation just described establishes a COI.  The situation is such that, based on what we know so far, it's a medium-high risk situation for a COI, but so far nothing you described constitutes an actual COI. North8000 (talk) 11:27, 7 July 2012 (UTC)
 * But according to the boldface definition that you say you support, it is a COI (the example is constructed specifically so that it does) because advancing outside interests (those of Singer Y's career) is more important to that editor than advancing the aims of Wikipedia. To fit current practice, we must either change the definition of COI in such a way that what it refers to is always going to be a bad thing, or acknowledge that a COI is not always a bad thing. (Or not use the term at all.) Victor Yus (talk) 12:05, 7 July 2012 (UTC)