Wikipedia talk:Conflict of interest limit

Paid Editing Proposals

In November 2013, there were three main discussions and votes on paid editing:

No paid advocacy (talk) (closed: opposed)

Paid editing policy proposal (talk) (closed: opposed)

Conflict of interest limit (talk) (closed: opposed)

Questionnaire

I can't tell where people stand on various questions that have been raised about COI, here and on other pages, present and past. How about if I post a questionnaire, and anyone can change the points or agree or disagree with any points? - Dank (push to talk) 17:49, 3 December 2013 (UTC)

I think that's a good idea, although the RfC at the top of this page does provide a general survey of people's views. Coretheapple (talk) 17:53, 3 December 2013 (UTC)

That's true, but many potentially useful points have been made at one point or another that have been addressed by few or none of the voters, at least so far. There's a chance that we may have made real progress here, but without more data, all I can do is guess. - Dank (push to talk) 18:03, 3 December 2013 (UTC)

True (though I wouldn't hold my breath about real progress). Coretheapple (talk) 19:18, 3 December 2013 (UTC)

Followup: my understanding of my job here is to say something when we seem stalled or when a deadline is about to run. We're certainly not stalled today, so I won't post my ideas for a questionnaire, but if any of you have questions that you think would clarify where everyone stands, go for it. I'm not going to shut things down on the 9th if we're in the middle of anything. (I can't speak for any other potential closers, though I wouldn't hold my breath waiting for any, it's hard to get closers for COI RfCs.) - Dank (push to talk) 19:26, 4 December 2013 (UTC)

I guess I would ask whether there is sentiment for beefing up notability requirements for companies and organizations, as much paid advocacy centers in that area, and concerns companies and organizations of marginal notability under current rules. Someone suggested that once in one of the other discussions but it hasn't come up here. Coretheapple (talk) 20:50, 4 December 2013 (UTC)

I think a questionnaire is a great idea to parse out points of agreement and points of disagreement, super clearly. Great idea.Jytdog (talk) 15:53, 5 December 2013 (UTC)

I agree, a questionnaire is the best way forward. With all of the different proposals and comments it would be good to find areas of agreement and possibilities for consensus. Adding to the confusion is this failed proposal and all of the changes that were made after most people had already commented. First Light (talk) 22:58, 8 December 2013 (UTC)

Okay, at some point we need a questionnaire. The 30 days for this RfC will run tomorrow, so unless there are objections, I'll close the discussion tomorrow and say the obvious: that we don't have the levels of support in this RfC, or in the previous six simultaneous conflict-of-interest votes, that would be needed for a new policy page along any of the lines discussed here. But I understand that this is not an acceptable result for the supporters, that "no change" may not be good enough to deal with threats like this one ... and they may be right about that. So I think the next step is to do some research on how big the threat is, and survey opinion on acceptable countermeasures. I'll ask Wikipedians I know, and I'd be happy to get any input along these lines. I'll report anything significant back on this page. - Dank (push to talk) 01:51, 9 December 2013 (UTC)

Thanks very much. I agree that is the most productive way forward and appreciate your not only laying the plans but offering to execute them. Jytdog (talk) 02:23, 9 December 2013 (UTC)

Yes, that's a good idea. Perhaps te Questionnaire would best be posted on the WP:COI Talk page in order to seek wider input there before deciding on the next (if any) step.--Ubikwit ^連絡 _{見学/迷惑} 05:25, 9 December 2013 (UTC)

• Dank, where would you like to accept suggestions for your questionnaire? Here or on your talk page? Coretheapple (talk) 19:49, 9 December 2013 (UTC)

Here is good. - Dank (push to talk) 20:01, 9 December 2013 (UTC)

OK, a couple of areas come to mind that haven't been previously discussed very much. Perhaps you could ask about: 1) Tightening the standards of notability for corporations, as that is a center of much of the COI and paid editing that we see; 2) Expressly prohibiting cash payments on the Reward Board; there's a proposal to that effect on the talk page of the reward board but it hasn't received much input; 3) Creating a Code of Ethics that would have the force of policy and go beyond the Conflict of Interest guidelines, expressly prohibiting paid editing but also setting forth ethical standards for Wikipedia editors in general. The point of the COE would be that Wikipedia editing is a public service, and that selfish motives, such as advancing agendas and editing for pay, are contrary to Wikipedia's mission, reduce the project's credibility, create a burden on other editors, and hamper editor participation and retention. Coretheapple (talk) 20:36, 10 December 2013 (UTC)

I think, for the questionnaire to be effective in helping us find areas of consensus, questions should be narrow and yes/no, so we can get usable data. It might be useful to provide additional space for comments but we need data. Questions like this (very roughly stated): "1) Do you agree that many editors find editing by "paid advocates" (defined as editors who are paid to edit Wikipedia in order to advocate for a client) troubling, especially with regard to Wikipedia's reputation and integrity, and it would be useful to take community-wide action to address these concerns? Yes/no Comment: _______ 2) Do you agree that many editors fear action to address concerns over paid advocacy, especially with regard to our core principles of "content not contributor", our being "an encyclopedia that anyone can edit", and the potential for hounding by anti-corporate activists, and that these concerns must be addressed in any community-wide action under 1)? Yes/no Comment: _______ 3) Many editors perceive that POV-pushing editors cause as much if not more problems as "paid advocates" (there is no data to show whether this is true or false): in any case, should a community-wide effort to address conflict of interest in the form of paid advocacy also attempt to address conflict of interest in the form activism? Yes/no Comment: _______ 4) Would you accept any limits on what "paid advocates" can do on Wikipedia to content related to their client? Yes/no Comment: _______ 4a) Would you find acceptable, an obligation for such editors to disclose their "paid advocate" role? Yes/no Comment: _______ 4b) Would you find acceptable, an obligation for paid advocates to not edit articles directly, except to correct factual errors or violations of BLP? Yes/no Comment: _______ 4c) Would you find acceptable, barring "paid advocates" from any presence on WIkipedia? Yes/no Comment: _______ 4d) Would you accept a "registry" for "paid advocates", participation in which would allow paid advocates to freely edit? Yes/no Comment: _______ 5) If a COI policy were enacted, how should concerns about another editor's potential COI be addressed? i) polite note on their user page calling attention to COI policy; ii) COI notice board; iii) A new noticeboard or function? If so what _________ iv) what limits should be placed on investigations, if any? ________" I realize questions 1 and 2 are pretty leading :) but if we structure the questions right, the questions themselves can help people get consensus-oriented and they can help us identify islands of common ground. i wonder - is there any intra-wikipedia survey function, like what is used to conduct elections for Arbcom, that could be used to gather data? Jytdog (talk) 21:25, 10 December 2013 (UTC)

• I would prefer simplicity of questions for the sake of clarity, especially the following main principles:

1. Should editing by advocates paid by an individual or organization be prohibited?
2. Should editing by advocates paid by an individual or organization be discouraged?
3. Should editing by advocates paid by an individual or organization be allowed on talk pages only?
4. Should we proactively engage with paid advocates by helping them to understand our core policies (WP:RS, WP:COI, WP:NPOV), and allow them to edit articles and/or talk pages?
5. If paid advocacy were prohibited, how should it be investigated and enforced?

First Light (talk) 04:40, 11 December 2013 (UTC)

That's incredibly confusing, since all advocacy is already prohibited. "Proactively engage" is very leading - let's keep that nonsense out of a questionnaire. Smallbones_(smalltalk) 04:59, 11 December 2013 (UTC)

I hear you. One of the key sticking points in this sprawling debate, is that many editors see no meaningful distinction between paid editors and paid advocates, and say exactly that -- that all advocacy is already prohibited. And this has led to a lot of the "oppose" votes. I think this is one of the hardest things to work through if we are to get a COI policy in place. I think defining "paid advocate" very clearly, is going to be crucial getting useful feedback - we need to write questions with that perspective in mind. It may be that we need to drop reference to "paid" altogether, if we are to get consensus. Jytdog (talk) 13:40, 11 December 2013 (UTC)

@smallbones, if all advocacy is prohibited, then any conflict of interest would be prohibited, since anyone with a COI is advocating on behalf of a POV. In practice that is not remotely true. (And smallbones, your attitude confirms many comments in these discussions regarding the "welcome" that advocates will and do receive if they are upfront about who they are. So much for new editor retention for those who do come out of the closet. Honestly, I would continue to encourage paid editors to stay in the closet until certain attitudes and editors change on Wikipedia). First Light (talk) 16:14, 11 December 2013 (UTC)

Lets get some strategic questions answered first

I think both Coretheapple’s and Jytdog’s proposed questions, while useful, are essentially merely collecting data to support tactical solutions. They don’t, in my mind, address the real strategic questions here. Which I think might be expressed this way: 1) Does the Wikipedia Community believe that the righteousness (NPOV, RS, NOR, etc.) of the contribution is more important than the status (COI, no COI, new or experienced) of the editor [contributor]? 2) Should a COI (however it is defined) situation involving new or inexperienced editors (a very routine occurrence where one assumes the editor has limited or no knowledge or understanding of Wikipedia’s COI guideline, let alone the myriad of confusing content and MOS guidelines.) license the community to deal with those editors in any way that might discourage them from becoming productive Wikipedia editors? The tactical questions that follow are almost limitless. The one question that I would characterize as semi-strategic/semi-tactical is this: What circumstances (stated as objectively as possible, leaving little room for subjectivity and situational interpretation by individual editors) constitutes an unwanted or undesirable nexus of COI and article editing?

• In my mind Question 1 is the most strategic and fundamental question the community must answer. If the content contributed meets our content policy and guidelines, does it matter what the status of the contributor is?
• Question 2 is almost as fundamental as #1. As a community, do we license ourselves to behave (thru process and discussion) in a manner that is inherently discouraging to new editors when those new editors have a COI/article nexus. (a situation that is routine and has a high probability of continuing as the norm for new editors)? Or do we recognize and understand the inevitable COI situations that new and inexperienced editors routinely find themselves in and make it our priority to encourage and mentor them to become productive members of the community instead of attacking their COI?
• Question 3 is clearly the most difficult because it not only requires unequivocal clarity, but effective conciseness to achieve the desired effect—editors should be able to clearly understand what does and what does not constitute an undesirable nexus of individual COI and article editing from a Wikipedia community perspective.
• Once these three questions are resolved, all the tactical solutions can be put on the table for discussion and resolution. Absent answers to these questions, proposing tactical solutions is pretty much a waste of time. Strategy without tactics is a slow route to victory, tactics without strategy is the noise before defeat—Sun Tzu. I much prefer victory and the strategic approach. --Mike Cline (talk) 00:38, 11 December 2013 (UTC)

Question 1 is ambiguous: Does it mean: 1a) "If the content contributed fully meets our content policy and guidelines, does it matter what the status of the contributor is?" Or does it mean: 1b) "If the content contributed partially meets our content policy and guidelines, does it matter what the status of the contributor is?" If it's 1a, then the answer is obviously no; but the question is also irrelevant and not fundamental at all. This is because no article fully meets the content policies and guidelines, because no article is perfect, and so the answer will simply never apply to any actual case. Question 1b is important, but the answer is obviously yes, because if content is created with a serious COI and lacks adequate coverage, neutrality or verifiability, then any such imperfection can reasonably be suspected as having been caused by the COI. --Atethnekos (Discussion, Contributions) 01:14, 11 December 2013 (UTC)

I don't think #1 is ambiguous at all. First the question deals with the quality of a specific contribution and the status of the editor contributing, not the perfection or imperfection of an article. If an editor creates a new article, then the entirety of that contribution would be the content under consideration. Since NPOV, RS, Notability and OR are in themselves somewhat subjective we should apply those subjectively interpreted policies in the same manner we do to routine contributions--something that's done everyday on WP. If an editor adds content to an existing article, we would evaluate the contributions exactly as we do today. If, in either case, the contributions don't live up to our subjectively interpreted content norms, then we have processes to correct. The causation of failure may indeed be a COI, but in my mind, the evaluation of the contribution, using our standard processes should proceed first, not a concerted effort to deal with the presumed causation prior to evaluating the content.--Mike Cline (talk) 01:35, 11 December 2013 (UTC)

So then, which one does it mean? 1a or 1b? Either is a fine interpretation of the question. --Atethnekos (Discussion, Contributions) 01:42, 11 December 2013 (UTC)

I think Mike Cline's questions are good, and I don't think it's even helpful to split hairs. Edits are usually partially compliant, and the evolution of an article usually involves an evolution of compliance with the main policies. I think "generally compliant" would suffice. First Light (talk) 04:33, 11 December 2013 (UTC)

Well if the answer is partially compliant or generally compliant, then my answer would be yes, it does matter—because a lack of compliance can reasonably be suspected as having been caused by the COI. --Atethnekos (Discussion, Contributions) 05:13, 11 December 2013 (UTC)

I think we could find content in many wikipedia articles, and in all controversial ones, that is only partially compliant. I disagree that it's reasonable to suspect all of it to be caused by COI. First Light (talk) 05:18, 11 December 2013 (UTC)

Agreed. I'm saying that when a COI does exist, a lack of compliance can reasonably be suspected as having been caused by the COI. Take for example the article on Dennis Lo. This article was, supposedly, written by a person paid by a business associate of Dennis Lo to do so. The article does not mention that Lo's process of fetal genetic screening has been criticized many times as a form of eugenics, even though there are many reliable sources documenting this. Is the reason for this lack of neutrality because of the COI? No one can know for sure. But can one reasonably suspect that it was caused by the COI? Absolutely; that's not an unreasonable judgement at all. --Atethnekos (Discussion, Contributions) 05:31, 11 December 2013 (UTC)

In that handpicked case, perhaps. In all cases, no. I agree with Mike Cline below. First Light (talk) 16:06, 11 December 2013 (UTC)

@Ate – although an apparent COI (however it’s defined) may lead one to “suspect” the COI to be the cause of poor content, I think it is a huge mistake to presume that. The presumption that a COI has caused the poor contribution amounts to “guilty until proven innocent”, which goes against our WP:AGF policy. In my view, WP has four classes of editor:

• The Righteous Editor: They understand the community norms, they follow them, and depending on their level of writing and research skills, they make quality contributions to the encyclopedia. A Righteous Editor may indeed have a COI/article nexus, but however they deal with that nexus, it is consistent with community norms.
• The Clueless Editor: They don’t understand or even are aware of the community norms. They may be very subject matter knowledgeable and skilled in writing and research or they may not. Their contributions may fall short of our community expectations and norms, but the reason they fail short is that the contributor is essentially contributing blindly, without benefit of understanding the environment they are working in. The WP community, in my view, must assume good faith from the clueless editor until such time, as that good faith is proven wrong. Even if the Clueless editor has a COI/article nexus, they community should never presume the editor is a POV editor without first mentoring and counseling the editor appropriately. As a community, our job is to turn clueless editors into righteous editors as quickly as possible.
• The POV Editor: They understand the community norms. They are aware of the policies and guidelines. They just choose to ignore and push their POV. They are easy to spot and I think the community does a reasonable job dealing with the POV editor. The POV editor may or may not have a COI/article nexus, but if they do, it is certainly evidence against them in any POV dispute. The POV editor, knowing the rules, tries to find clever ways to circumvent them, and hide their POV pushing from the community. We already have effective ways to deal with such circumventions. The difference between the POV editor and the Clueless editor is clear. The POV editor is consciously ignoring and circumventing the rules for whatever is motivating them, whereas the Clueless editor, when caught in a POV situation is only doing so because they don’t understand the norms.
• The Vandal Editor: They may or may not know the rules, but their motivation is clear—disrupt, trash, vandalize, or other harass the Wikipedia community and its encyclopedia. We have ways to deal with that.

The real challenge here is to create COI related guidelines that don’t impede the process of turning the Clueless editor into a Righteous Editor. It is important that if COI is the root cause of POV editing, then our COI related guidelines are clear and useful in dealing with such motivations, but not at the expense of Clueless and otherwise good faith editors. --Mike Cline (talk) 12:16, 11 December 2013 (UTC)

Mike, with all due respect, I think you're focusing far too narrowly, and in a totally speculative way, on the impact of COI on editors and not taking into consideration the reason why COI policy is required, which is the impact of COI on the content and on the reader. You're also losing sight of the impact of COI editors on editors who don't have a COI, who are demoralized and sometimes driven away from the project entirely by editors with both declared and undeclared COIs. I've personally seen it happen and that is why I am so passionate on this issue. If we are going to be drawn into a debate on how to rescue, counsel and rehabilitate editors who have ulterior motives, quite simply nothing will be done. It may indeed be one of the reasons that nothing has been done; that is, our priorities are misplaced. What's needed is to give editors, when they sign up, an explanation of best practices. I think that agreeing to a carefully drafted Code of Ethics, encompassing bans on both paid editing and unpaid POV pushing, and explaining the purpose of the project in case they're not familiar with it, would be a good way of helping everybody: the COI editor and the non-COI editor. Somebody once suggested this "sign up" requirement; I forget who unfortunately. Coretheapple (talk) 23:37, 11 December 2013 (UTC)

Mike I am hearing you. Would you be more comfortable if a COI policy would make explicit reference to WP:BITE and would note that new editors need some training and need to be given the opportunity to comply?Jytdog (talk) 13:34, 11 December 2013 (UTC)

Anything that helps protect the new, clueless editor, from inquisitional COI processes, is good. But your question prompted another thought. What if, instead of focusing on the very complex and difficult to define COI/Article nexus that the community deems undesirable, we focus on POV editing policy. That puts the focus on the righteousness of the contribution versus the motivations of the contributor. Clearly, POV editing can be motivated by COI, but focusing on the POV not the COI allow negates the need to clearly and unequivocally define undesirable versus allowable COI/article nexus. The great majority of the opposes in this RFC and others have said in some way "focus on the contribution not the contributor". A POV policy, vice a COI policy would do that and accomplish the same goal--improve and sustain the quality of the encyclopedia as laid out in NPOV, V, RS, etc. --Mike Cline (talk) 14:15, 11 December 2013 (UTC)

Thanks for replying. But I am asking if you would be open to dropping your objections if a COI policy explicitly dealt with newbie issues. Would you? With respect to shifting focus away from COI per se... I really believe we need a COI policy and that it is irresponsible to out public and ourselves for Wikipedia to lack one. So I am unwilling to give up the focus on COI per se. I am sorry you have not heard me on that. I am pretty flexible about the specifics of the policy. I just want clarity on COI. Jytdog (talk) 14:31, 11 December 2013 (UTC)

Actually, I have no objections to the idea that WP has a COI policy. Such a policy might be very useful and I will fully support (participate in) its development. My objections (if you want to call them that) stem from the absence of a clear community position on the questions I posed above. That's why I characterize them as strategic. Once we understand where the Wikipedia community comes down on these questions, then a COI policy can be crafted that respects and implements these positions. I don't mean to be beating a dead horse here, but as I've said before a number of times above, clearly and unequivocally articulating what does and does not constitute an undesirable COI/article nexus from the Wikipedia community perspective is a huge challenge. It has to apply to ~125,000 editors +. But I think it is a critical first step in outlining a COI policy. Such a definition will not be useful if it allows for whimsical, situation interpretation anytime someone has an ax to grind with another editor (new or experienced). I would much rather see an RFC that attempted to resolve this one question before any attempt was made to regulate what happens when an undesirable COI/article nexus occurs. --Mike Cline (talk) 15:52, 11 December 2013 (UTC)

It doesn't fail to assume good faith at all, nor does it presume someone is guilty. First of all, there is no requirement that readers assume good faith. But it is they whose trust must be kept. However, if a reasonable reader leaves an article with a misconception of the facts of the matter (due to lack of neutrality, verification, etc.), and then she later finds out the article was written with a potential COI, and that this imperfection could serve the conflicting interest, then she can reasonably suspect the COI caused the imperfection, and her trust is betrayed. Academic publishers/editors don't employ policies which target authors with potential COIs because they fail to assume good faith, they do so in order not to betray the trust of readers. Indeed, as the ICMJE says "the potential for conflict of interest can exist regardless of whether an individual believes that the relationship affects his or her scientific judgment" [1] (see [2] for a list of journals which follow ICMJE recommendations). This position has been upheld by numerous academic studies, see [3] for example: " Our results suggest that problems of conflict of interest may be more profound than is commonly assumed. It is not enough to consciously counteract potentially biasing influences on judgment; people might not be able to adequately correct for biasing partisan influence. Eliminating partisan allegiances may be the only way to eliminate conflict of interest." Even if users are judging that the conflict of interest caused the lack of neutrality etc., that's not a failure to assume bad faith, that's just following the consensus of reliable sources on the issue. --Atethnekos (Discussion, Contributions) 21:23, 11 December 2013 (UTC)

I don't think it's necessary to answer any "strategic" questions first. It's not unusual for Wikipedia rules to overlap and contradict themselves. If we're going to view "question the content, not the contributor" as supreme and sancrosanct, then our hands are tied and it gets kicked over to the Foundation, which is probably where it belongs anyway. Coretheapple (talk) 02:40, 11 December 2013 (UTC)

To maintain the assumption of good faith, I think it is important to distinguish (as the current conflict of interest guidelines do) between potential, apparent, and actual conflicts of interest. Questions about the relative importance of a well-sourced piece of information versus the provenance of the editor who included it are related to the problems of apparent conflicts of interest as well as actual conflicts of interest. Depending on which one has come into play, different responses may be warranted. isaacl (talk) 16:16, 11 December 2013 (UTC)

Although the arguments being made are rational, they're not going to solve anything, because neither side in the seven recent votes gave much of a response to the arguments of the other side. (That is, the arguments are good, but I don't see much in the recent votes that encourages me that the bulk of the votes can be swayed all the way over to the other side by a clever argument, in either direction.) If anyone can see a compromise position, something that feels "halfway" between the two sides, please either post it or email it to me (using the link on my userpage). - Dank (push to talk) 03:09, 12 December 2013 (UTC)

amen brother. Jytdog (talk) 03:26, 12 December 2013 (UTC)

Agreed. Dank, I think you know what's going on here well enough by now to craft a questionnaire if you still think that's the best way forward. The same main disputants can't even agree on the questions. A questionnaire might draw new blood into the discussion, and would at least give the community a big picture view of the main issues and how editors feel about them. First Light (talk) 03:56, 12 December 2013 (UTC)

I think WP:COI already has broad support, both through those who stated that it was sufficient, and through its citation in many discussions. Thus I believe it can act as a starting basis for a policy. isaacl (talk) 04:35, 12 December 2013 (UTC)

A poorly crafted COI policy will be a threat to new editor retention

One of the difficulties in trying to craft a COI policy such as this is creating language that is crystal clear in preventing abuse against new and inexperienced editors by veteran editors. Wikipedia wouldn’t need projects like this: Wikipedia:Snuggle and like this: Wikipedia:WikiProject Editor Retention if life as a new or inexperienced editor was easy. One of the supporters above lamented: This policy does a great job of providing clear guidelines that are already pretty well adopted by the Wikipedia community. As for enforceability, our three core content policies are often unenforceable. We have thousands of articles that are not notable, are unverified, and not neutral. But shouldn't these still be policies? I think so. My first thought when I read that was that there are thousands of potential new editors who might provide the verifiable content that would improve those articles, but they would probably get hammered by the COI police before they ever had a chance to contribute productively. If we create a COI policy, the language must be such that it not only prevents what we need to prevent, but that it will be understood and enforced in a manner that is not unwelcoming to new and inexperienced editors. Big challenge, but of utmost importance for the future of WP. --Mike Cline (talk) 14:59, 5 December 2013 (UTC)

I really hear your own experience as a new editor coming through, Mike. But even though you had that experience, you stuck it out, because you do care about the encyclopedia. If somebody's only interest in being here is a COI-driven desire to grind an ax or polish an image, and a COI policy would run them off, I think that is great, as it saves us the work of cleaning up after them. (I have had a strange little run of COI editors showing up at articles I work on, strange... and it has just been a huge time suck for me.) We would need to administer the policy wisely of course, so as not to lead to bruises like the ones you received, that still appear to smart even today. In some ways a clear COI policy probably would have saved you some of the pain you went through, as it would have been very clear to everybody involved what the boundaries were. Jytdog (talk) 15:48, 5 December 2013 (UTC)

My position here is much less about my experiences years ago, than it is about creating policies (and resultant community behavior) that drive away potential new editors that have access to verifiable content and the ability to improve scope and quality in the encyclopedia. This language in WP:COIN Once COIN declares that an editor has a COI for a specific article, COIN (or a variety of other noticeboards) may be used to determine whether an edit by a COIN declared COI editor does not meet a requirement of the Wikipedia:Conflict of interest guideline. makes it clear to me that the COIN process trumps NPOV and RS. Additionally, this one COIN example is a classic case of a COI inquisition. Wikipedia:COIN#Nevron_Open_Vision. Without a doubt, the suspect new editor probably has a COI related to the product, but not one word of encouragement or advice has been given, merely accusations of bad editing behavior.[4] Even without a thorough examination of the edits, which I assume weren't properly sourced, etc. (not unusual behavior for a new editor), its clear that the product Nevron Open Vision, is potentially notable in its own right, let alone being referenced in related articles. [5]. Yet, in the world of COI inquisitions, NPOV, RS, or welcoming, encouraging and mentoring new and inexperienced editors is of secondary importance. That's my concern. --Mike Cline (talk) 16:37, 5 December 2013 (UTC)

There's a good case for re-thinking our approach to editor recruitment and induction. Historically, the emphasis has been on encouraging new editors to contribute/correct content. Nowadays, we should be encouraging potential editors to begin by contributing references instead. It's a much safer way to learn about how Wikipedia works: once a newbie is comfortable with WP:RS there's a comfortable path to WP:N and WP:V before getting into the more difficult issues of WP:BLP, WP:NPOV and WP:COI that a new editor needs to grok before adding content. In the Nevron Open Vision case, I've some sympathy with Christo Bahchevanov, because it isn't made clear that any item added to List of platform-independent GUI libraries should have its own article. Having checked out your Google search, their website and a few other places I'm afraid it's not very likely he could find an independent source that would convince me their product is notable, though. - Pointillist (talk) 18:30, 5 December 2013 (UTC)

Discouraging COI editing from new editors is desirable. Editors with a COI will be discouraged from COI editing. Those without will be happy that COIs are prohibited. Coretheapple (talk) 19:03, 7 December 2013 (UTC)

I think that you put it the wrong way, in focusing on the COI and not on the new editors. Most new editors make COIs, I believe, because the subjects they have COIs in are the subjects they are most interested in. We should be *encouraging* new editors to turn to non-COI subjects. We should never be discouraging good-faith new editors at all. —David Eppstein (talk) 18:56, 9 December 2013 (UTC)

I think a Code of Ethics would go a long way toward giving that encouragement, if drafted correctly. The idea would be to provide a positive, uplifting message, not a "keep off the grass" sign.Coretheapple (talk) 19:38, 9 December 2013 (UTC)

Everyone working on this COI thing ought to read this

The Decline of Wikipedia

It provides an interesting perspective that ought to inform whatever we choose as our "COI Policy". --Mike Cline (talk) 16:16, 12 December 2013 (UTC)

What's your point, @Mike Cline? Do you want feedback on that article or on Aaron's "productive newcomer" model? Personally, I don't believe the currently available statistics are a satisfactory basis for reasoning. What I'd like to see is a timeline of sources being added, because that's the heart of the editing process. If you counted one point for each reference added to article space and zero for everything else, I expect you'd see a very different pattern of editor involvement and productivity. This has the great advantage that it can be measured directly and doesn't trip over all the problems with edit counts: if an editor prepares a big edit offline, adding five sources in a single edit, that will score five points; quick edits tidying up grammar and punctuation score no points, and so on. - Pointillist (talk) 17:51, 12 December 2013 (UTC)

My point is that it is an interesting perspective that shouldn't be ignored as we pursue this. No more, no less. --Mike Cline (talk) 17:58, 12 December 2013 (UTC)

I wish people could read that article with an open mind, without having a knee-jerk defensive reaction and attacking Mike for posting the link. "....90 percent male, operates a crushing bureaucracy with an often abrasive atmosphere that deters newcomers who might increase participation in Wikipedia and broaden its coverage." That statement could be applied to many of the edits in this discussion ("abrasive") and I wouldn't be surprised if the combatants here were 100% male. Any paid or COI editors who see this discussion will continue to remain in the closet, out of fear of running afoul of some of the very same abrasive editors who have been trying to form a new policy here. First Light (talk) 18:24, 12 December 2013 (UTC)

I'm not attacking the link. My mind is still wide open. But as that article describes multiple perspectives, none of them specifically about COI, it's not unreasonable to ask which aspect Mike Cline finds "interesting". If the point is that a poorly crafted COI policy will be a threat to new editor retention, we already have a section above to discuss that! On the other hand, if Mike Cline is saying that in order to broaden Wikipedia's coverage we need to attract people who have a COI (and then presumably channel their efforts) then that's more along the lines of could professional editing be made safe? Either way, I don't think the existing statistics are going to be very helpful. - Pointillist (talk) 21:56, 12 December 2013 (UTC)

Indeed, that's why I've been citing the article both above (Wikipedia_talk:Conflict_of_interest_limit#Wikipedia.E2.80.99s_reputation_argument_a_Red_herring), and at Jimbo's talk page. --Atethnekos (Discussion, Contributions) 18:37, 12 December 2013 (UTC)

evaluating threat size and countermeasures

Following the statement in the close: "I think the next step is to do some research on how big the threat is, and survey opinion on acceptable countermeasures." -- I've started a list below with a few that I've heard others discuss off the top of my head; feel free to edit, add, remove. – SJ + 05:07, 28 December 2013 (UTC)

Threat size and classification

Many commented that the biggest bias threats come from non-financial COI: things handled in the COI guideline but generally not in the 'paid advocacy' discussions. This is just classification and data about threats from PR firms and other organized paid-editing institutions:

• Major PR firm cases: most recently wiki-PR in EN, wikiExperts in EN and ES; previously, Bell Pottinger in EN; another firm in ES a year ago (? : link needed). All in all, a few hundred+ accounts. Were these caught as soon as they were noticed? Where was there uncertainty about how to handle those cases? (WikiExperts has been debated as possibly "ok", or "ok if they commit to abiding by COI guidelines", or "ok if they abide by guidelines and list all past and present contractors")
• Major corporate clients of PR firms: a few have made the news (after their contractors were blocked for policy-gaming PR practices). Is this an active problem for certain classes of firms?
• Although this is just one example (there are hundreds), wouldn’t employees of this organization Wild Salmon Center be considered “paid advocates” if they edited articles related to the “advocacy” mission of the organization? The mission of Wild Salmon Center is to promote the conservation and sustainable use of wild salmon ecosystems across the Pacific Rim.[6]. Unlike, commercial organizations whose typical “mission” is the production of products and services (not advocacy of a POV), a great many non-profit organizations have a specific “advocacy” or POV promotion mission. In general, wouldn’t employees of these types of organizations pose a greater NPOV threat to Wikipedia? --Mike Cline (talk) 17:59, 28 December 2013 (UTC)
  • No, in short. That logic skirts the entire public interest issue vis-a-vis Wikipedia's own status as a "non-profit", basically.

While it is true that, overall, POV pushing not related to paid advocacy editing is a bigger problem to Wikipedia, that would appear to be a problem that Wikipedia has lived with for a long time. While I agree that the problem needs to be solved--through more stringent content related policies--that sort of POV pushing is not as easy to deal with as paid advocacy editing, policy wise, and does not pose a threat to the non-profit status, which is fundamental to the information provision ethos of Wikipedia. If Wikipedia becomes commercialized, it becomes a Frankenstein creature of the sponsors, not an open-sourced source of reliably-sourced information (hahaha).--Ubikwit ^連絡 _{見学/迷惑} 19:09, 28 December 2013 (UTC)

I think the idea of "exempting" non-profit organizations from the "paid advocacy" label not only goes contrary to the definition of an organization in WP:CORP but demonstrates an extreme bias that has nothing to do with "public interest". In the example I cite (as I said one of hundreds I could cite), this organization in 2011 generated revenues of ~$9M, had salary expenses of ~$2.7M (paid advocates) and generated a gross profit of ~$3.7M.[7] They profit from their advocacy and they pay employees to advocate for their POV. They just don't pay income taxes. Why should we rely on the problem needs to be solved--through more stringent content related policies--that sort of POV pushing is not as easy to deal with as paid advocacy editing for this type of organization which profits from its advocacy and not for "for profit" entities? If we want to create a double standard then we ought to be very upfront about it. --Mike Cline (talk) 19:35, 28 December 2013 (UTC)

Right, I didn't mean to grant an outright exemption to non-profits, but the money trail and the difference in orientation makes them a different beast from PR orgs. There is a lot os corruption in the NPO/NGO sphere, but they are still more single purpose organizations more readily identifiable on a given article related to their field.--Ubikwit ^連絡 _{見学/迷惑} 19:50, 28 December 2013 (UTC)

I don't think Wikipedia's status as a non-profit should be the concern here. If that was a significant issue, I think this would be being handled by the WMF. Just as they don't ask volunteers to run their fundraising and defend them from legal challenges, they wouldn't ask volunteers to form a policy if their status as a non-profit depended on it. And this isn't a global policy, whatever we decide on here will only affect one of the hundreds of WMF sites (though enwiki policies are often used as a framework for other projects). I think we should be focused on concerns local to the English Wikipedia - NPOV, disruption, and our reputation. Mr.Z-man 21:32, 28 December 2013 (UTC)

Proposed countermeasures

What countermeasures have been proposed, and which were acceptable to whom?

COI-focused

• More active implementation of the COI guideline
• Making the COI guideline policy (periodically opposed)
• Implementing a policy themed on the guideline (per the 6+ now-closed discussions similar to this one: no consensus to do this; but much participation)

Topic- and edit-focused

• Better tools for flagging edits / topics that may be being gamed(?)
• Better ways to flag topics for extra COi scrutiny, based on past editing patterns(?)

Transparency-focused

• Increasing incentives for self-identification of conflicts. [extra lenience/support for self-id'ed editors; extra penalties/reversion for socking, or for those found not to have self-id'ed (?)]
• Using a draft namespace to allow direct proposals of drafts and draft changes
• ...

Just acknowledging I'm still watching; input on any of this would be appreciated. - Dank (push to talk) 17:35, 28 December 2013 (UTC)