Wikipedia talk:Conflict of interest limit/Archive 2

Archive 1

Archive 2

Some plausible scenarios and how the vagaries of this proposal might play out

(All references to people or entities are fictional in these four scenarios but represent real-world conditions. Assume all relevant article topics are otherwise notable.)

• An elected member of a Native American tribal council receives remuneration from the council, but is also an employee of the local tribal library receiving a salary. This individual is a Wikipedia editor (editor Bob). Can editor Bob contribute to an article on the Tribal Library (by this proposal—article’s subject—NO). Can this editor Bob contribute to an article on the Tribal Council (by this proposal—article’s subject—NO). Now, can editor Bob contribute to other WP articles that further the aims and interests of the tribe such as other tribal organizations, treaties, history, events, etc? (the vague nature of the phrase article’s subject leave this up to inquisitional interpretation.)
• An employee of Acme Donut company is a Wikipedia editor, editor Jane. Clearly editor Jane cannot edit an article on Acme Donut under this proposal. But Acme Donut is independently operated, but owned by Acme Brands, a company that owns five different fast food brands. Acme Brands is a portfolio company of Acme Capital which owns companies in a variety of market sectors. Some of the profits of Acme Donut flow to Acme Capital. An executive at Acme Capital learns that editor Jane is an experienced Wikipedian and asks her to write (without additional remuneration) articles about Acme Capital portfolio companies. If editor Jane complies, will editor Jane be violating this policy. (the vague language, “article’s subject” leave this up to inquisitional interpretation.
• Editor Bill is a PhD working for a non-profit organization that is charged with promoting and supporting research to assist various government agencies in meeting their missions. The organization is sponsored with both funding and infrastructure by the local university because the university derives a lot of revenue from related government projects and grants. Editor Bill also assists the university with some class room instruction in related curriculum, but is paid solely by the non-profit organization. Can Editor Bill work on an article on the university? Can editor Bill work on other articles closely associated with the aims and interests of the university? Can editor Bill work on articles related to the government project his non-profit is supporting?
• Editor Edith is an employee of the Potato company, a large global technology company that has a great many products, technology related services and either operates or sponsors a lot of industry related associations, commissions, learning and teaching organizations. Clearly editor Edith cannot edit the WP article about the Potato company, but what about all those products and services, associations, commissions, and other entities closely related to and supported by the Potato company?

The phrase Article’s Subject is very vague in establishing a relationship between an editor getting paid by some entity (most of us get paid by someone) and articles the editor might not be allowed to edit. If we chose to interpret Article’s Subject literally, then it is not prohibited for an Apple employee to contribute to an article on an Apply product or Service, as Apple (the company is the Article’s Subject in relation to whose paying the Editor). It that the goal here?

The phrase engaged in competition [with] is equally vague. Is Burger King in competition with McDonalds—Yes. Is Burger King in completion with Darden restaurants (Maybe. Is a QSR restaurant chain in competition with casual dining chains? Is Burger King in competition with Flemings, a fine dining restaurant? Are Montana State University and the University of Montana in engaged in competition for students, grants, government funding, etc. Indeed they are as are all universities, as are all participants in any given market sector. Who decides the competition, which inevitably exists and will continue to exist, rises to the level where an editor working for one competitor can be sanctioned for contributing to articles about other competitors in the market sector. The only way it will happen is through energy sapping inquisitional interpretation. --Mike Cline (talk) 14:15, 25 November 2013 (UTC)

Whilst I appreciate the concern regarding the ambiguities of any conflict of interest policy, I think your sample scenarios are still subject to interpretation with your proposed wording: Editors who are paid or otherwise compensated by any institution, organization or government agency may not edit articles that further the interests and aims of the institution, organization or government agency. If you do believe a broader wording would help clarify these cases, can you expand on this? Is it a matter of trying to find a balanced set of descriptive phrases that will identify as many clear-cut cases as possible? isaacl (talk) 16:49, 25 November 2013 (UTC)

I think my "proposed wording" has been taken out of context. I would agree with you that the "proposed wording" in the context of these scenarios would still leave a lot up to interpretation. The real intent of the wording was to suggest that if the goal was to disallow all paid editing, then you could write that policy in an unequivocal way. The wording was not a real proposal for language in this proposed policy, but merely a sample of unequivocal language that would eliminate the need to create explanations and exemptions. I really don't believe there is suitable and concise wording that would clarify a COI policy whose intent was to disallow edits from one class of editor while allowing the same type of edits from another class of editor based on the relationship of the editor to the topic being edited. Life, commerce, government, etc. are very complex enterprises. Any given individual's role in those enterprises is equally complex. Drunk driving is bad and there are policies against it. Those policies are clear, concise, measurable and desirable. More importantly, they are essentially unequivocal, thus easy to enforce. Still doesn't stop drunk driving. COI policy for a global, volunteer enterprise that functions in and across all aspects of human enterprise is extraordinarily difficult to promulgate in a clear, concise, measurable and desirable way. Why are we trying? What is the real goal here? --Mike Cline (talk) 19:22, 25 November 2013 (UTC)

Hi Mike. Cannot quite figure out where you are coming from. From my perspective, the goal is to get a COI policy in place, which we lack. COI policies are basic good governance that regulates access to or use of "X" for private gain. In the case of companies (for-profit and nonprofit alike), "X" = corporate property and resources. In our case "X" is the privileges we grant to anyone to edit Wikipedia. There are editors who use their access, to edit for private gain. Because we lack a policy, the community is in turmoil, and doesn't know how to deal with issues when they arise, and we remain in situations where the door is wide open to paid advocates (which are the concern, and are a common-sense disinguishable subset of paid editors) and we are subject to scandals like the Wiki-PR debacle, which leave us further in turmoil and which damage our reputation with the public. I grant that it is difficult to create useful, actionable language; difficult is not the same as impossible, if we are committed to the goal and agree to allow some wiggle room for common sense to handle boundary cases.Jytdog (talk) 19:59, 25 November 2013 (UTC)

Above you said one option was to remain with "... current COI guidelines which in my view are sufficient." Naturally no guideline or policy is going to provide complete clarity, but perhaps there are aspects of the current guidelines, which you seem at least in part to agree with, that seem reasonable to use as a basis for a policy? isaacl (talk) 20:01, 25 November 2013 (UTC)

Plain and simple where am I coming from: What is or will be the goal of the COI policy you are proposing? If the goal is just to have a policy, that's a lame goal and tells us nothing about the purpose or results of actually executing the policy. Once the policy is in place, how do you see WP changed or improved? What is the measurable and desirable consequence of the COI policy that the community can agree on? Once you can decide and get community agreement on the goal, then the job of creating appropriate language to achieve the goal will be much easier. So my advice in moving forward would be to seek community consensus on what a "COI Policy" is trying to achieve in clear, concise, measurable and desirable terms. After that the details become easy. --Mike Cline (talk) 22:53, 25 November 2013 (UTC)

I haven't personally proposed any policies, and so I can't speak for those who have been putting forth proposals. From a general perspective, though, organizational conflict of interest policies have the following objectives:

• Identify what situations in which someone has a conflict of interest.
• Provide guidance on what someone should do when they have a conflict of interest.

By laying out guidelines in advance, it is easier to make decisions on what to do when a situation occurs, as thought has been invested in considering multiple factors, and more equitable treatment should ensue of those affected. So the net effect should be to reduce questions on what constitutes a conflict of interest and how to deal with them. Of course discussions may still happen, but they gain a head start from having the policy in place, versus having to start each discussion from scratch each time.

I suspect, though, you are thinking of what are the desired goals of any actions taken in response to someone breaching the conflict of interest policy. Based on these discussions, my suggestion is the following:

• Reverse any harm caused by conflict of interest editing.
• Seek to prevent further harm.
• Guide editors with a conflict of interest on how to collaborate with the Wikipedia community.

Given that these ideas are embodied within the current conflict of interest guidelines, I believe there should be common ground for all interested parties, upon which progress can be made. isaacl (talk) 23:53, 25 November 2013 (UTC)

Sorry, there is also the question of what are the goals for the guidance on what editors should do in conflict of interest situations. The ones that I think have general support are the following:

• Permit a narrowly defined set of non-controversial edits that can be made to articles in Wikipedia's mainspace where a conflict of interest situation exists.
• Disallow any other edits to articles in Wikipedia mainspace where a conflict of interest situation exists.

I believe there are many who support a degree of engagement on discussion pages, including proposing text, but there are some who are against significant engagement. isaacl (talk) 00:04, 26 November 2013 (UTC)

Hey Isaalc looks like we are more or less on the same page. Hard part of Mike's questions was metrics... how do you tell if the actual policy is meeting your goals? You need to something to measure. They teach this way of thinking in MBA and MPA programs - think about outcomes in terms of something you can measure, and set up ways to measure it, so you can tell if you are doing well or screwing up. Jytdog (talk) 00:09, 26 November 2013 (UTC)

I think most corporate employees will have heard of SMART goals: specific, measurable, attainable, relevant, and time-bound. For now, I'm just looking for some agreement on the general direction. isaacl (talk) 00:20, 26 November 2013 (UTC)

my apologies for telling meyou what you already knew. what do you think of my response below? Jytdog (talk) 00:55, 26 November 2013 (UTC) (corrected pronoun Jytdog (talk) 01:29, 26 November 2013 (UTC))

I apologize for my tone coming out wrong; I just meant to say that the concept has been popularized beyond master degree programs. I like how you've tried to figure out specific metrics that can be monitored, so trends can be examined. isaacl (talk) 01:16, 26 November 2013 (UTC)

it's quite alright! i am a bit intense about avoiding miscommunication which can lead me to state the obvious. sorry again. Thanks for reviewing my response - happy we are on same page then. thanks.Jytdog (talk) 01:29, 26 November 2013 (UTC)

(edit conflict) Hi Mike OK, thanks for making it more clear. I hear you now and understand what you are asking - these are the appropriate questions to ask of a policy in formation. Clear goals, measurable outcomes. Yes. I am an idiot for not framing it this way earlier, even in my own thinking about it. So thanks for posing them. Here is a very quick answer. I apologize that I am producing this on the fly, in trust that you understand this is draft-y and we can talk about it and work together to improve it. (If that is a bad assumption and not the kind of conversation you are interested in, please let me know and I will take a different stance) But what do you think?

• Goal 1: Provide guidance to editors and the community about
  • a) what constitutes a conflict of interest on Wikipedia,
  • b) how editors with a COI should behave, and
  • c) how editors who suspect another editor of making conflicted edits should behave
• Measurable outcomes:
  • a) improve understanding of COI, as measured by surveys (e.g. there is existing literature showing that PR professionals have a lack of understanding of COI in Wikipedia, which we could use as a baseline)
  • b) Super hard to quantify outcomes here. Could look at things like (i) number of editors disclosing COI; (ii) number of complaints brought to COIN; (iii) reduction in tendentious editing (NB: Very hard to come up with metrics of what improved outcomes with regard to tendentious editing or paid-advocacy-editing would look like, as there is not good data now on the extent of COI editing or tendentious editing in Wikipedia to use as a baseline - I opened a thread on Jimbo Wales Talk page asking if anyone knew of data and was told there is none and it is basically impossible to get, although there were a few promising replies).
  • (c), there are two things I could think to measure (i) number of complaints brought to COIN; (ii) reduction (hopefully) in complaints about COI or tendentious editing at ANI; (iii) very hopefully, reduction in complaints about hounding accusations of COI-editing at COIN (a COI policy with clear procedural guidance should reduce the length and severity of "inquisitions" and should eliminate hounding, as those specific behaviors would be sanctionable).
• Goal 2: Improve trust of the public in the integrity of our articles, and trust of donors/grantors in our governance
• Measurable outcomes: Level of trust as measured by surveys
• Goal 3: Drive PR houses that promise to do reputation management via Wikipedia out of business (controversial goal, right?)
• Measurable outcomes: measure job postings at ELANCE and number of PR houses like Wiki-PR and WikiExperts; I would expect to see a reduction in their numbers or size as such activity would be against policy and I cannot imagine that legit businesses would generally pay for illegitimate activity.

There is a quick outline, very open to suggestions! And as noted I recognize that Goal 3 is probably controversial. Jytdog (talk) 23:59, 25 November 2013 (UTC)

Jtydog - If I was coaching someone on how to rephrase these goals, the intent of which is clear and with which I agree, they would probably read something like this.

• Goal 1: The COI policy ensures editors have a clear understanding of community consensus as to what does and what does not constitute a sanctionable conflict of interest on Wikipedia. The COI policy ensures editors understand acceptable editing behavior when they have a sanctionable conflict of interest with any given article topic. The COI policy ensures editors who suspect other editors of sanctionable conflicts of interest understand how such conflicts of interest should be dealt with.

• Goal 2: The COI policy ensures that public trust in the integrity of our articles and the trust of donors/grantors in our governance is not damaged due to serious conflict of interest issues.

• Goal 3: The COI policy ensures that "reputation management" editing by Wikipedia editors, paid or otherwise is either strongly discouraged or outright prohibited.

You'll note a common characteristic which each of these goals. The burden is on the policy (and assumed associated processes) to achieve desirable results. When a goal does start with a verb it merely describes activity, not desirable, measurable end states. If I was satisfied that WP COI policy language would achieve these goals as written I would most likely support it as long as the associated processes were not so onerous as to cause net harm to the encyclopedia. --Mike Cline (talk) 00:01, 27 November 2013 (UTC)

If @Jytdog and @Issacl were clients or students in my strategy seminars, you'd be getting good grades. You both grasped my concerns well and the initial ideas around goals and measurement surrounding a COI policy are not without merit and some of the best I've seen on WP. I want to digest further before making any detailed comments. But, that said, my gut says something is missing. Not quite sure how to express it as a goal, but COI should NEVER trump NPOV, RS supported contributions. If an editor's contributions are 100% consistent with our content policies, then the COI should be a secondary matter, maybe called attention too, but not punished. This is a big challenge. Clearly, serious COI impacts the ability or the impression of the ability to contribute NPOV, RS supported content. The ideal COI policy, in my view, would create an environment that promotes NPOV, RS supported contributions, regardless of the status of the contributor. The ultimate goal here is to build a better encyclopedia, that has to be captured in some form in this policy. --Mike Cline (talk) 01:54, 26 November 2013 (UTC)

Thanks for the praise! :) Do you all find the current COI guideline reasonable -- that if you have a COI for a given chunk of content, that you should a) disclose it and b) not directly edit or create content in the article, but rather provide drafts in a sandbox or on the Talk page and then request that the draft be reviewed for NPOV and RS by unconflicted editors and directly added by unconflicted editors if the draft passes review? I find it a reasonable way to manage COI. Jytdog (talk) 02:14, 26 November 2013 (UTC)

I really like what you said Mike, b/c it echoes what many opposers of a COI policy have said, that content is king. I think a lot of those opposers might shift some, if the intro stated clearly that the mission of wikipedia policies is to build NPOV, RS content, and that the goal of this specific COI policy is to manage COI in furtherance of wikipedia's mission. Framing matters, I think.... Jytdog (talk) 02:19, 26 November 2013 (UTC)

The perceived reliability of an edit is also a concern; if an editor has a conflict of interest, this can compromise the degree of trust readers will place in the corresponding edits. In the interests of a clear policy, which is a concern you have raised, it may be simplest to stay within the bounds of the current conflict of interest guidelines which do not have an exception (outside of non-controversial edits) along the lines you suggest. Since the conflict of interest guidelines have a wide amount of support, including your own, I think it can serve as an initial starting point. As more experience is gained, further refinements can be entertained. isaacl (talk) 02:26, 26 November 2013 (UTC)

@ Jytdog, the current COI guideline in my view is sufficient to mitigate the impact of COI on Wikipedia. It is obvious that many ignore its advice and I suspect many, especially newer editors don’t even know it exists. Although it doesn’t set unequivocal policy, its advice is pretty clear. I do think it has a couple of weaknesses. The guideline says: COI allegations should not be used as a "trump card" in disputes over article content. but that is what happens in a great many cases. First, it is an easy route to win a content dispute with a new uninformed editor and second, making the allegation, even if done in good faith, diverts energy from real evaluation of the content to the legitimacy of the editor. One only has to go back and look at a lot of COIN discussions over the years to see that they are inherently inquisitional in nature, especially for new, uninformed editors. The contributor may be arguing for the righteousness of the content, while the community COI police are saying we know you are guilty of COI so there’s no way this content is any good. The second flaw in my view is that the COI guideline in general enforces a community-wide bias against content about private commercial enterprise—the evil corporation so to speak. There is no doubt in my mind that there are probably 1000s of commercial enterprises/processes/events, etc. that would easily make the notability cut in WP, especially when compared to the very low notability hurdles that some other topic areas favored by the community seem to enjoy. Yet, content availability and independent references for these types of enterprise is generally known only to individuals with some close association to the topic (a COI if you will). The way the community handles COI in general then becomes a significant blocker to improving the encyclopedia in these types of topics.

The one COI minefield that you all have included in the proposed policy—competition—is not referenced in the guideline. That’s a good thing. Almost every enterprise is in competition with other enterprises in some way. Establishing a policy based level of competition that would trigger a COI allegation would be impossible and render the idea of using “common sense” moot. --Mike Cline (talk) 12:17, 26 November 2013 (UTC)

Hi MIke, I just reviewed some of the links you provided in earlier comments you made -- I see that you got burned by COI allegations when you were a newbie and understand your concern about a COI label trumping NPOV and RS (although I reviewed the prometheus principle case -- albeit briefly - and it ~seemed~ that your sourcing was pretty weak. You were new then!) But I do see what you mean. Thinking about that experience... thanks again. Jytdog (talk) 12:56, 26 November 2013 (UTC)

Jytdog - The Prometheus Process experience was very instructive for me on many levels. At the time, I knew literally nothing as to how Wikipedia worked and had a lot of misconceptions about the "authority" of individual editors who collectively at the time were behaving like well coordinated pack dogs in attacking my COI. Although at the time, I freely admitted my COI, even such admission was seen as a ploy to avoid community sanctions. The sourcing at the time may have been weak, but no one even cared about actually seeing if notability could be established. I was tarred an evil spammer from the onset. I survived and had the fortitude to find other ways to be productive on WP, but to me at the time the lesson was clear. COI trumps NPOV, notability and RS. I think is still does when one looks at a lot of the COIN discussions.. It is interesting today to think about whether a Prometheus Process article would meet the notability test. I think it would as my experience has taught me how to ensure that. But I wouldn't even try, nor would I ask anyone that would have any ability to actually write an article about the process to do so. Why, because anyone with the knowledge and ability to access and understand the sources would inherently have a COI. Thus in a era of free knowledge, the knowledge about a strategy process used by the WMF, a number of prominent large corporations in a variety of market sectors, and a number of prominent foreign governments is not in the encyclopedia. That's the kind of harm that over zealous COI enforcement causes. --Mike Cline (talk) 00:27, 27 November 2013 (UTC)

The current conflict of interest guideline broadly defines a potential conflict of interest existing "when advancing outside interests is more important to an editor than advancing the aims of Wikipedia". It further says "Adding material that appears to advance the interests or promote the visibility of an article's author, the author's family, employer, clients, associates or business, places the author in a conflict of interest." This covers editing articles of competitors. Given your opinion that the current guideline is sufficient to mitigate conflict of interest concerns, a view that many have expressed, then perhaps we should proceed forward with proposing it as a policy. isaacl (talk) 13:30, 26 November 2013 (UTC)

Issacl, I would characterize my position about the COI guideline this way. It seems to be sufficiently comprehensive enough to mitigate most COI that WP might be concerned with. I don’t think it should become policy without considerable revision and really continue to question the need for a COI policy in the first place. I am working on comments re the goals of a COI policy and hope to get them up later today—am bouncing around my client’s time and my own for the rest of day. --Mike Cline (talk) 18:13, 26 November 2013 (UTC)

Wikipedia’s reputation argument a Red herring

This type of statement: The perceived reliability of an edit is also a concern; if an editor has a conflict of interest, this can compromise the degree of trust readers will place in the corresponding edits. I believe is a distraction and as the section header says is a Red herring without empirical evidence to back it up. In theory, real COIs may indeed result in some content that is not consistent with WP norms. But, there’s no evidence that I can find that’s there’s been any widespread decline in WP global readership, even though we’ve had some high-profile cases of COI. So the idea that we must stamp out COI to improve trust in WP sounds good, but the need to do so is unsupported by any real evidence that trust in WP is waning or declining because of COI under current guidelines.--Mike Cline (talk) 12:35, 26 November 2013 (UTC)

Sorry to do this, but do you have any data to support your position (readership, surveys, etc?) I recognize that I need data to support the assumption that publicity about COI editing scandals damages our reputation; I will look too. But we both should have support for our conclusions. Am taking travelling today for the holiday but will try to find time to do this - it is important. Jytdog (talk) 12:53, 26 November 2013 (UTC)

Well WP is currently the 6th most visited website in the world. (Follow links at Special:Statistics to Alexa). I'd have to do some digging, but I've seen recent statistics that say ~350M readers look at WP everyday and its been that way for a long time. If our reliability or credibility is really being hurt by COI, I would think those numbers would be in decline. Most importantly, our audience is so large, that generalizing any specific type behavior to "readers" doesn't carry much weight. --Mike Cline (talk) 13:06, 26 November 2013 (UTC)

Readership statistics is one index, but there would be a lag time, due to the influence of search engines which will continue to rank a site high as long as it has a lot of incoming links. Furthermore, without a comparable site of similar breadth as an alternative, unless the issue becomes pervasive, Wikipedia will likely remain a popular link destination. Editor participation stats are another index, since the prevalence of biased editing can discourage good-faith editors, but the effect can be difficult to discern from the many other forces that can influence participation. A survey of perception by the academic press and general media might be instructive. isaacl (talk) 13:30, 26 November 2013 (UTC)

I am surprised by this suggestion, Mike. In my experience, it is an axiom in journalism, academic publishing, and civic reporting that COI contributions and paying people to write about you invalidates the value of the resulting work. I don't think we need to wait for a catastrophe of reputation to recognize the potential problem. Anecdotally: whereas normally at public talks about WP people ask me general questions about how Wikipedia works and how people become editors, in the past few months many people have asked whether some articles are paid for or written by PR firms. That may take years to translate to a drop in readership (!), but it already influences some teachers to ask students not to use WP as a reference. – SJ + 22:28, 26 November 2013 (UTC)

You can't seriously be suggesting that students should cite Wikipedia when our own WP:TERTIARY guideline says Wikipedia shouldn't be used as a source. Anyway, if you are experienced in the axioms of COI, why did you choose to sign your post as SJ rather than as SJ (WMF)? - Pointillist (talk) 22:52, 26 November 2013 (UTC)

Hello Pointillist, Wikipedia is a reference work. It is designed to be used as a reference. That is not the same as citing it as a source: tertiary reference works are best used as a way to understand the broad picture, and to find individual sources, which are then directly cited. (If a writer is being thorough, they should also note any tertiary works used. However the WP style guide does not yet go into that level of detail, nor do our citation templates support such subtleties as intermediate-refs, such as via/ᔥ and HT/↬ ). If educators and other community leaders discourage others from using Wikipedia at all (not simply from citing it in papers; something that as you note we also discourage), that is negative feedback worth considering. And unlike readership stats, that is a leading rather than a trailing indicator.

I don't see a WMF-specific meta-COI here, and my edits on en:wp are made in my personal and not Board capacity; but I would be interested to read more about what conflict you have in mind. – SJ + 23:09, 26 November 2013 (UTC)

Sorry, I misunderstood your 22:28 post. Yes, in general Wikipedia should be reliable as a reference for schools. The areas where COI is most likely to occur – such as popular culture, companies, brands, technology frameworks, products, services – don't belong in an academic curriculum anyway. As for your WMF affiliation, when you comment about policy it is your responsibility to make your affiliation clear without needing to be prompted. This is a widely accepted axiom in journalism, academic publishing, and civic reporting nowadays. This is not the place to discuss the reputation of the WMF, and I don't want to single you out personally in any way, but as a long-standing trustee you must be aware that statements made by people speaking for the Foundation are not automatically taken as being neutral. - Pointillist (talk) 23:32, 26 November 2013 (UTC)

Thank you for explaining. – SJ + 00:54, 28 November 2013 (UTC)

"If publicists were to overtake the voluntary editors, much of the site would become online billboards (perhaps with deceptively neutral writing). But as it stands now, the protection against biased writing — dogged hounding from voluntary editors, and disorderly debates — is narrowing the field of voluntary editors, and most likely turning off potentially valuable contributors." (Salon 23 October 2013 [1]) The techniques currently used to combat paid advocacy editing (automated tools like Stiki, aggressive enforcement of WP:NPOV and WP:V) are part of what is causing the decline. This was confirmed by Aaron Halfaker's assessment of the current decline in Wikipedia (MIT Technology Review October 22 2013 [2]). "Overall, the site’s community of editors has done a decent job of weeding out the worst self-promotional offenders, or at least moderating their contributions. But according to the Wikimedia Foundation, Wikipedia’s parent entity, the tactics being used to force content onto the site are becoming more advanced and more widespread ... It’s this type of “astroturfing” activity that Wikimedia is struggling to identify and contain." (Digiday 26 November 2013 [3]). With the failure of this proposal, this will be the sixth(?) proposal that has failed to help solve this problem. Many users here who have taken part in conflict of interest editing themselves insist there is no problem, or that more data is needed before any action is taken. WMF executive director Sue Gardner, WMF founder Jimbo Wales and virtually everyone in the technology press who is analyzing this issue seem to disagree. --Atethnekos (Discussion, Contributions) 22:48, 26 November 2013 (UTC)

@SJ I would make this observation about comparing WP to a lot of legacy knowledge forums as mentioned here: I am surprised that you would suggest this, Mike. In my experience, it is an axiom in journalism, academic publishing, and civic reporting that COI contributions and paying people to write about you invalidates the value of the resulting work. Your experience in those fields is undoubtably as you say, but the big difference and one that I think WP will struggle with for the next decade. We are entering a era of free knowlege. Whereas in journalism and academic publishing, there is instrinct value in your work because people pay for the knowledge, it isn't free. When the knowledge is free, as in WP I think the rules will change. All that said, if our content is NPOV, reliably sourced and free, those who consume it will care little about how or by who it was contributed. I believe that as long as we enforce NPOV and RS, no matter how hard that might be, our reputation among the millions of readers who consume our free knowledge won't be harmed a bit. IMHO --Mike Cline (talk) 23:15, 26 November 2013 (UTC)

@Ate this will be the sixth(?) proposal that has failed to help solve this problem. I think the reasons these proposals failed, is not because a lot of editors don't think COI or paid promotional edit isn't a valid issue, they do, They just think that these proposals won't solve the issue and would as written cause more harm than benefit. That's my take and opinion. --Mike Cline (talk) 23:31, 26 November 2013 (UTC)

I believe, in their zeal to influence the discussion, a number of people chose to release their own proposals. While their enthusiasm is great to see, it had the unfortunate effect of diffusing discussion. This has happened with many Wikipedia initiatives; progress is typically best made when there is a concerted effort to converge towards a consensus view. isaacl (talk) 23:48, 26 November 2013 (UTC)

I don't want to make your job any harder, but you really need to fix your "self-edit" exemption to make it consistent with current COI practices. If you do, I can may support this. Figureofnine (talk • contribs) 00:14, 27 November 2013 (UTC)

Overview of comments

Looking at the current comments, it appears that 7 editors support the proposal, 3 tentatively oppose it, and 3 completely oppose it. A number of suggested modifications have been made that I think we should discuss.

(1) Add prohibitions against editing with intent of financial gain, and editing when there are lawsuits or off-Wiki conflicts with the subject. Intent of financial gain means that the motivation behind the editing is financial, which closes many loopholes on paid advocacy without prohibiting good faith edits by people who just may be indirectly connected to the subject. Do we want to add a fifth prohibition that bars editing when they have intent of financial gain, or have been a party to a legal conflict connected to the subject.

(2) The need to define terms or give examples. I think we may get our in trouble if try to define family member or other terms, especially considering that Wikipedia is used around the world. However, do we want to give examples of what is permitted and what is not?

(3) I think that we should treat enforcement of this policy or any paid editing/COI proposal as a separate project. After a policy is endorsed, then we can think about the best way to enforce it. Sockpuppetry was banned before Wikipedia had all of its mechanisms for sockpuppet investigations, and I think the same will be true here. Any thoughts? DavidinNJ (talk) 21:13, 16 November 2013 (UTC)

I think that the intent of this proposal was to focus on the narow scope of paid advocacy (editing), not COI in general, so I've taken the liberty to revert the other points added later not related to paid advocacy.

In order to get this passed, it is necessary to keep it narrowly focused, leaving open all avenues for subsequent expansions to be proposed and discussed in a similar manner. The most effective way to approach this problem is probably an incremental approach, that is to say, one item at a time.--Ubikwit ^連絡 _{見学/迷惑} 21:25, 16 November 2013 (UTC)

Ubikwit, While I understand the desire to make it easier to get approval for this proposal, I think that the removal of the family, significant other, and roommate clause makes this proposal much weaker, and may actually cause a loss of support. Since that clause has been included for the last five days, which is when most of the editors here reviewed the proposal, I have restored it. If a person is allowed to write an article about their spouse, then we might as well not have COI rules. DavidinNJ (talk) 21:34, 16 November 2013 (UTC)

If the policy goes through with the affinity relationship specifications included, then there would be nothing to object to as I agree with the content. This proposal has not received much attention, however, as several other more comprehensive proposals are concurrently under discussion, and none seems to be leading to a tangible result. Since the response has been disappointing it may not matter anyway, but the difference, in retrospect, with this proposal was its narrow focus on commercially motivated advocacy.--Ubikwit ^連絡 _{見学/迷惑} 22:40, 16 November 2013 (UTC)

The job here is bigger than satisfying a handful of people, but I have no objection to how you're proceeding. - Dank (push to talk) 21:45, 16 November 2013 (UTC)

I don't think we should remove the family, significant other, etc. restrictions. In the other discussions, people specifically criticized how only paid editing was restricted. By keeping these provisions, we make it somewhat more comprehensive, while still concise and understandable. I don't think we should make it still broader, though. Superm401 - Talk 01:44, 17 November 2013 (UTC)

I think if you include family you should make it less vague; "immediate family" is clearly defined. – SJ + 02:51, 17 November 2013 (UTC)

As I said earlier, I didn't include the word immediate because is Wikipedia is a global website and the immediate family member does not have the same meaning around the world. I'm going to write a few proposed examples to show what is permitted, and what is not. DavidinNJ (talk) 04:12, 17 November 2013 (UTC)

Proposed examples

In order to eliminate some gray areas that editors have been discussed, I am proposing that we include the following text. Feel free to comment. Keep in mind that no list of examples can be completely exhaustive and deal with every last potential issue. DavidinNJ (talk) 04:44, 17 November 2013 (UTC)

Use common sense in interpreting this policy. Remote connections do not constitute a conflict of interest. The following examples are offered to illustrate what is permitted and what is prohibited.

(A) A person can edit an article about their fifth cousin or long-dead ancestor whom they have never personally met, but cannot write about their brother-in-law who they see for Christmas.

(B) A person who receives Social Security benefits can edit an article about the Social Security Administration (SSA), but an employee of the SSA cannot.

(C) A professional wine reviewer can edit an article about a winery that they written about outside of Wikipedia, but cannot edit an article about the organization for whom they write.

Regarding (C): would it be a form of self-promotion to write about subjects that you had reviewed for your job? Using similar descriptive phrases could raise the search engine ranking of the web site hosting the original review, much like how Wikipedia mirrors show up in search results. isaacl (talk) 14:22, 17 November 2013 (UTC)

When a reviewer writes about a book, movie, winery, etc., they aren't being paid by the subject that they are reviewing, so there's no direct conflict of interest if they edit a Wikipedia article about the same subject. I kind of see your point about indirectly boosting the profile of a non-Wiki article by using similar terminology, but that's beyond the scope of this policy. As stated above, this policy doesn't seek to ban every conceivable conflict of interest, just ones that are serious enough that we have a consensus to prohibit. DavidinNJ (talk) 21:10, 17 November 2013 (UTC)

The potential conflict of interest is with the editors themselves; they can be promoting their own reviews. It comes close to using your own work as a citation. If your work is an undisputed authority on the topic, then there isn't a problem, but with something more subjective, I think it is something to be cautious about, with an uninvolved person determining if the included information is being given its due weight. Perhaps a more clearcut example would be better in this article. isaacl (talk) 23:28, 17 November 2013 (UTC)

I am confused about COI from employment. Could I edit an article about a company I worked for years ago with which I have no continuing relationship?What about if they fired me and I am unhappy about that? Is that worse than editing an article about a company I never worked for which I don't like, or a sports team or politician i don't like, as long as the edits meet guidelines and policies otherwise? Could I write about a company I used to work for from which I retired, and who gives me a pension?(I might be as likely to criticize as to praise, and I would know where to find RS. Could I write about a company I never worked for but whose stock I own, when I own less than one millionth of the equity? Edison (talk) 17:53, 28 November 2013 (UTC)

Flawed wording

As I mentioned in my comment in the RFC section, the fact that this needs so many examples of things that "don't count" means the wording is probably fatally flawed. The biggest problem I see is the phrase "paid by." Other potential edge cases I can see:

• A university professor in the US funded by an NIH grant. Is he allowed to edit the NIH article? The United States Department of Health and Human Services article? Any article about the federal government?
• Would a doctor be allowed to edit the articles on insurance companies that his practice accepts?
• Would people be allowed to edit the articles of companies they own stock in? Even if it's only 1 share? What about bonds (where the interest rate is fixed and the payout is essentially unrelated to how well the company does)?
• Are sub-contractors considered "paid by" all of the companies in the supply chain? If I work for Company A that provides sheet metal to Company B that makes parts for Delphi Automotive, that makes parts for General Motors, obviously I'm paid by Company A, but am I also "paid by" Company B, Delphi, or GM?

Another issue is the "engaged in competition" clause. How direct does the competition have to be?

• Are 2 universities in the same state, or same country "in competition?"
• Would a restaurant be considered to be in competition with every other restaurant that has locations in the same region(s) as them? Just ones in the same price range? Same cuisine?
• Using the supply chain example from above, Delphi and Visteon (which makes parts for Ford) are both in the same general market, but they make parts for different companies. So are they in competition? Since GM and Ford are obviously in competition, would a Delphi employee have a COI if editing the Ford article?
• What about internal divisions within the same company. Does an employee for GE Lighting have a COI on Rolls-Royce Group PLC, because another division at GE makes jet engines and turbines?
• Does a player on a Triple-A baseball team have a COI with other teams in the league? Other players on those teams? Other players on his own team? The coaches and managers of the teams? MLB teams? Lower-level leagues?

These are all things I came up with in around 20 minutes. I can't imagine how many more would come up in actual use. The problem is that this tries to lump paid advocacy in with all other types of COI editing. Paid advocacy is the real problem. There is no reason to treat bookkeepers and IT staff, whose pay is probably not tied in any direct way to the success of the company, the same as their PR people or top management, whose compensation and continued employment may be contingent on the image of the company. Mr.Z-man 20:42, 29 November 2013 (UTC)

BLP issue

I've been away from editing for a while, and just belatedly found this. Although I like the general concept of the proposal, I want to point out that, per WP:BLP, we should not prohibit people from directly editing pages about themselves if they are simply correcting errors or vandalism. --Tryptofish (talk) 22:56, 20 November 2013 (UTC)

I included the relevant language from that policy. Coretheapple (talk) 00:38, 21 November 2013 (UTC)

It seems to me that the first bullet point in the section about "Behavior of editors with a COI" directly contradicts WP:BLPEDIT. I don't mind encouraging caution with respect to such edits, but saying flat-out that such edits "shall not" happen is much too restrictive. --Tryptofish (talk) 23:17, 21 November 2013 (UTC)

Just for the record, language directly quoting the BLP policy was added in response to these BLP concerns but was removed[4]. I see no point in tweaking this proposal further as it seems moot at this time, given the intensity of the opposition. Coretheapple (talk) 22:00, 29 November 2013 (UTC)

COI editors are very likely to argue that they are "Correcting errors or vandalism" when they make self-serving edits. Tommy, who own Tommy's restaurant, might remove a statement in the article about his restaurant that he served food past its expiration date, arguing that the local newspaper reporter who wrote the cited article was a disgruntled former employee who lied and the paper was a worthless rag and not a reliable source. Any criticism of a subject might be seen as errors or vandalism by those who like the subject, whether it be a company, a sports team or a politician. Edison (talk) 18:02, 28 November 2013 (UTC)

Edison, everything you said is correct and I agree with all of it. But, it's just as true that the person who is the subject of a BLP might come along and correct something that is entirely appropriate to correct, such as fixing a date of birth or a date of something else. I don't think that we should want to block them for having done that, instead of making an edit request on the article talk page. It seems to me that the language proposed here would open us up to that: "An editor shall not edit an existing article directly if he or she is: the subject of the article". --Tryptofish (talk) 00:19, 29 November 2013 (UTC)

Another way to mitigate COI impacts

I suspect someone has proposed this before, but if they haven’t I am more than willing to take credit for this brilliant idea. Wikipedia has an very interesting essay: Wikipedia:Ethical Code for Wikipedians. I didn’ know it existed until tonight. The fifth paragraph succinctly discusses to perils of COI. Wikipedia also has an article on Conflict of Interest in which the concluding paragraph contains this language: Codes of ethics help to minimize problems with conflicts of interests because they can spell out the extent to which such conflicts should be avoided, and what the parties should do where such conflicts are permitted by a code of ethics (disclosure, recusal, etc.). Thus, professionals cannot claim that they were unaware that their improper behavior was unethical. Although this would not solve serious COI by currently registered editors, this procedure might prevent new editors from unknowingly falling prey to our COI policy. My suggestion is that all new account registrations be required to acknowledge they’ve read our code of ethics. Now we can’t force them to read them or follow them, but we certainly can record the fact that an account has had the opportunity to read them. As I experienced as a new editor, the community cannot assume editors are aware of COI rules before they violate them. Once that happens, the COI police attack and all is lost. There is a reason they post speed limit signs prominently. --Mike Cline (talk) 02:57, 27 November 2013 (UTC)

Great idea. Many sites have similar "read this and acknowledge" on sign-up, involving code of conduct and terms of use. First Light (talk) 03:37, 27 November 2013 (UTC)

I like this too. One thing I have noticed is that in the motto that greets people, "the free encyclopedia that anyone can edit", the link under "anyone can edit" leads to Wikipedia:Introduction which says nothing about COI, which I think steers people the wrong way, with the best of intentions of being welcoming... Jytdog (talk) 23:04, 30 November 2013 (UTC)

Could professional editing be made safe?

Reading @SilkTork's oppose (diff) made me wonder whether there's any possibility that professional edits could be made acceptable, if their editors would consent to sufficiently rigorous conditions. For example:

• Professional editors promise to abide by specific policies about how they edit and interact with other users.
• These policies would include strict rules about notability, verifiability and reliable sources that would apply to all their contributions.^{added 09:45, 4 December 2013 (UTC)}
• If they fail to adhere to these policies, they can be complained about in a specific venue similar to AN/I. They can be admonished, blocked or banned.
• Professional edits may only be made with a special SUL account set up for this purpose. This account is individual, not corporate. It is distinct from any account used for normal "unbiased" editing. The professional account-holder must identify to the foundation (or similar), so any one individual can only have one professional account in their lifetime (or at any one time, perhaps).
• Their usernames must end with a standard string such as "(COI)", "(POV)" or "(PAID)".
• Their user pages have a standard format that lists their affiliation(s). Professional editors don't have discretion about how this is presented.
• Professional editor accounts are technically marked as such: it's a sort of user right, and it cannot co-exist with any of the discretionary rights like autopatrolled, rollbacker etc.
• An edit filter logs their contributions e.g. "(Tag: professional edit: please check for COI)". All their edits are globally opted-in for edit count statistics.
• They can't create a new article directly. Instead, they must use the proposed draft namespace.
• Perhaps they can't edit an existing article that is being watched by very few people (at least 30 watchers probably means any COI will be detected).

...and so on. I'm just putting this up as a thought-starter: I'm not claiming to have considered it deeply yet. The details would require extensive discussion and there are edge cases (e.g. is unpaid professional editing in scope?) that need to be agreed. It just seems to me that since almost all the advice to potential COI editors is "Don't directly edit" (e.g. WP:NOPAY, WP:PUBLICISTS, WP:PLAINSIMPLECOI, WP:BESTCOI) we are pretty much guaranteeing to drive underground something that is probably already happening to some extent. By replacing the muddy middle ground with a distinct class of editor, highly visible and subject to specific policies and sanctions, we could be much clearer about the bright lines around COI editing. Thoughts? - Pointillist (talk) 13:33, 28 November 2013 (UTC)

I think this is much more practical than the proposals to ban it entirely. Mr.Z-man 20:43, 29 November 2013 (UTC)

In some ways I like what you are doing here - allow, with clear boundaries. Freedom, with heightened responsibilities. The downside of this, is that the rest of us are volunteers, and this would seem to put us at risk of opening the door wide to something we don't have the bandwidth to properly manage. Following you out onto this limb.... on the federal level, the way this is handled is to require user fees so that there can be staff on hand to handle the demand (see PDUFA for an example). In our case, user fees could perhaps go to WMF to pay for staff to audit/follow the activity of Registered Professional Editors (RPEs, to risk an acronym). What about something like that? Jytdog (talk) 22:58, 30 November 2013 (UTC)

Thanks for commenting @Jytdog. I should have made it clearer that all such professional editors will still have a duty to make neutral, verifiable contributions. Like the rest of us, their edits must improve wikipedia. If licensing professional editors doesn't add enough value to significantly outweigh the admin overhead of on-boarding them, we shouldn't do it. - Pointillist (talk) 21:38, 1 December 2013 (UTC)

I think that what you're suggesting is a good idea, up to a point. It would take care of the issue of internally identifying professional editors to other users. Current proposals would permit them to continue to make proposals in talk pages, and right now there is no mechanism for identifying who they are, that Editor X is working for the subject. Sometimes they disclose. Sometimes it's part of their username. Sometimes they "forget." Or sometimes there's no disclosure at all; it's just a "big mystery." That would deal with that issue. What they do, how much they can do, is a separate issue. I'm not in favor of them creating articles, in draft space or any space, including Articles for Creation. One point that needs to be considered is whether our notability standards for corporations and organiztions (the main locus of paid editing abuses) need to be tightened up in general. Coretheapple (talk) 22:50, 3 December 2013 (UTC)

I think @Coretheapple is making a useful point about organization/brand/product notability issues, which can waste a lot of time on talk pages, AfD etc. But we could prevent non-notable articles from professional editors because (a) our rules would demand that their edits be built exclusively on reliable sources and cc-by-sa'd media, and (b) they wouldn't be allowed to start new mainspace articles anyway. If their client/employer's organization/brand/product hasn't received mainstream press coverage yet, then it won't be possible for a pro editor to produce an acceptable draft article about them. Underlying all this – frankly blue-sky thinking – is the assumption that pro editors will have a vested interest in behaving properly if they are named, disclosed to the Foundation, checkuser'd, labelled as potentially NPOV, and subject to scrutiny and sanction if they abuse the system. Maybe they'd become collectively self-policing to some extent. If we could get the marketing/PR community to accept a system with explicit rules like this, it would be easier for us to target "amateur" COI editing, wouldn't it? - Pointillist (talk) 09:35, 4 December 2013 (UTC)

Responsible PR people want to do the right thing at Wikipedia, which is to disengage. See this recent article by a PR person] on how the PR industry needs to disengage from Wikipedia. I think that we need to discourage engagement, not encourage engagement. Another useful exercise would be to adopt a Code of Ethics that would cover best practices that would include a ban on paid editing but go beyond it, to include unpaid advocacy editing. Elements of that are scattered throughout Wikipedia policies. Why not centralize them and make them formal policy? Coretheapple (talk) 14:54, 4 December 2013 (UTC)

While responsible PR people may want to do that, they have to strike a balance between what's responsible and what will keep them employed. The fact that an organization like Wiki-PR exists means that there is significant demand by the corporations hiring PR people specifically for editing Wikipedia, which means that PR people who choose to disengage do so at the risk of losing their jobs. Mr.Z-man 15:26, 4 December 2013 (UTC)

That's a problem PR people face in general from their ethical standards. If their clients are wise, they'll abide by their PR counsel and and not engage in activities that could bring them into disrepute. Coretheapple (talk) 19:13, 4 December 2013 (UTC)

Though well intended, this would seem to represent a manifestation of "policy creep" that would be accompanied by bureaucratic bloat, not to mention adding a rather unwieldy strata classification to the editor body politic.

If there is to be a policy, it has to be minimal and aimed at prevention of abuse and preservation of the editing environment as well as conservation of administrative resources, not a cumbersome assignment of new rights.

On the other hand, I would basically agree that the creation of quality content should be facilitated, albeit indirectly through mechanisms that serve to create a buffer zone of sorts.--Ubikwit ^連絡 _{見学/迷惑} 14:56, 4 December 2013 (UTC)

We already have something like 15-20 user groups. How much time was spent in the sockpuppetry investigation of Wiki-PR? Trying to prevent it entirely still has the potential to require significant amounts of resources. We can try to prevent abuse, or we can try to save time, but I don't think we can do both. Mr.Z-man 15:26, 4 December 2013 (UTC)

I think that I basically agree with your points here. If creating "professional editors" would constitute a new "user group", then as described above, I would oppose that.

Regarding the prevention of abuse and saving time and conserving resources, the emphasis I have tried to focus on prevention is related to these points. I don't see this policy as a rallying call to gather bounty hunters gunning for COI editors, but more of an enforcement tool that can help streamline administrative action in the case that a problem occurs and is recognized. As Sue implied below, WMF can take legal action against Wiki-PR types without a community policy, as such companies are already in violation of the Terms of Use, so this policy would more about lower intensity actions at the community level, but the policy would also explicitly proscribe the activities engaged in by Wiki-PR types at the same time.--Ubikwit ^連絡 _{見学/迷惑} 16:29, 4 December 2013 (UTC)

Except in the case of Wiki-PR, the main problem wasn't figuring out what to do with paid editors, it was finding all their socks and meatpuppets. We didn't really need a paid editing policy because they were violating so many other things. If we had a policy that set out clear ways for paid advocates to edit with proper scrutiny, the entire investigation may have been avoidable: Wiki-PR may have chosen the "white hat" route and disclosed their COIs. Or, if there was a set of straightforward rules for PR editors to follow, the entire Wiki-PR business model may have been unnecessary as companies could deal with their Wikipedia articles in-house rather than having to hire contractors. Mr.Z-man 16:56, 4 December 2013 (UTC)

Well, aren't we glossing over the fact that companies pay entities like Wiki-PR because they want to see themselves portrayed in a positive light on Wikipedia even if that means engaging in tendentious POV-pushing, etc., over and against what RS say about them and their activities? Companies would not want to be seen to be engaged in such conduct themselves, as the reverse of the desired effect would result upon exposure.--Ubikwit ^連絡 _{見学/迷惑} 17:04, 4 December 2013 (UTC)

I wonder if it would be worthwhile to reach out to companies that hired WIki-PR to ask them about this.Jytdog (talk) 17:08, 4 December 2013 (UTC)

Comparison with existing COI policies

I would appreciate a list of COI policies in various other industries and networks. The FTC has a policy which CorporateM and others have mentioned; as do many conusmer-advocacy, journalism, marketing, and other groups. Many of the arguments "against" having a COI policy assume that marketers and PR firms would try to avoid them. However I have the sense that most established companies and firms expect such policies and go out of their way to abide by them where they exist. It is simply confusing for those groups when we do not have one.

Having a selection of existing policies to refer to / draw language from would help speed parts of this discussion. – SJ + 22:32, 26 November 2013 (UTC)

I'm going to try to get some. We also may want to look more broadly at codes of ethics elsewhere. Coretheapple (talk) 19:14, 4 December 2013 (UTC)

Thank you for working on this policy

I didn't know this policy was in development: I'm happy to have just now stumbled across it.

As ED of the WMF it's very common for me to get questions from non-Wikipedians, particularly public figures, journalists and PR people, about this issue. Most frequently I'm asked "is it okay for me to edit the article about me," "is it okay for my colleague or student to edit the article about me," "can I correct errors of fact in the article about my [employer/school/whatever]," "is it okay for me to hire [somebody else: a firm or individual] to edit the article about me," and "if I hate the article about me, where can I send corrections, comments or additional information." I've tried to be helpful but it hasn't been easy, because although there is a lot of policy and practice on the enWP it has never (AFAIK) been boiled down to a few sentences that are understandable for people who don't know much about Wikipedia or how editing actually works in practice.

I believe those people would find this policy extremely helpful.

I saw upthread that some editors were expressing concerns about how policy like this could or would be enforced. That's a legitimate conversation, but it focuses solely on bad-faith actors, which IMO misses a large part of the intended audience for this. My experience is that there are many people who generally respect Wikipedia and want to play by its rules, but who don't know what those rules are: they don't know how the projects are actually built in practice, and so they're unsure about what's okay to do and what's not. (This may be particularly the case for the kinds of people who are notable enough to warrant a BLP, and/or company execs who're interested in how their organization is portrayed on Wikipedia: those people skew older and therefore probably generally aren't super-comfortable with online conventions/tropes/norms etc.)

I believe that if enWP makes it clear what it wants people to do, the vast majority will be happy to do it. That's why I think a succinct, accessible policy is really important, and I am very glad to see this one in development. Thanks Sue Gardner (talk) 21:00, 29 November 2013 (UTC)

Good to see you, Sue, and your timing is good too ... I won't close this until the 30 days run on 9 Dec, but we've got enough from this vote/discussion/RfC/whatever (and from the other 6 simultaneous votes, all closed now) to give you guys some preliminary results. What you just said about what you need is clear and helpful. Bottom line: the supporters in these discussions have been unsatisfied that we've got only a COI guideline; they want a new COI policy. In a way, that simplifies the job here: a new policy page would need to be supported by people representing basically all of the significant POVs on the subject among Wikipedians. Taking all 7 discussions together, we've got no better than 25% support for what's currently under discussion, or anything like it. The opposition isn't about the message that COI editing is perilous (which WPians are agreed on), it's about wanting to avoid the question whenever possible of who's a suitable editor ... I think the opposition doesn't come so much from arguments pro and con as it comes from our day-to-day experience of what happens on article talk pages: when people are talking about the quality of the edits, things have a way of turning out okay; when people start talking about who is and isn't suitable to be editing articles directly, things just have a way of going all to hell, no matter how well-intentioned the question. Having said that ... Wikipedians have a tendency to decide that some questions are "too hard", and back away, and I've always had a concern that not being willing to tackle the harder questions leaves us vulnerable ... and also, unable to give Foundation things that it asks for and needs. I hear you saying that the Foundation needs some things here that it doesn't have yet, including basic guidelines on how people can edit "safely", without being attacked for violating some rule they didn't know about, is that right? - Dank (push to talk) 21:31, 29 November 2013 (UTC)

Herostratus made what I thought was a lucid observation on Jimbo's talk page concerning the adoption/non-adoption of a "Bright Line Rule" as formal poliucy: "It's essentially impossible to get significant broad written rules (as opposed to minor tweaks) adopted here anymore. It's essentially impossible because (to elide and simplify a lot) you need a supermajority, and that's hard, quickly approaching impossibility as the number of participants and proliferation of side-issues multiplies. I think that the last big written rule adopted here was WP:BLP, which passed in 2007 I think (and could probably never pass now). The days of making sweeping changes by adopting formal rules on the Wikipedia are over, I think."[5] I think he's basically right, and the opposition to such a rule is so strong, and is especially intense among experienced users and administrators, that it doesn't seem likely at all to be adopted. The only chance of some kind of formal stricture being adopted in the foreseeable future is if the Foundation takes the matters in hand, and makes it part of the Terms of Use, so it's really your call. Coretheapple (talk) 21:49, 29 November 2013 (UTC)

We've got the low-hanging policy fruit in hand already; if there's more policy to be had, it's harder to reach, and Wikipedia has the same problem that every corporation and organization does: people gravitate toward the easy questions and avoid the hard ones. (And I think it's probably time someone did something about that; I'm working on a relevant Signpost column as we speak.) - Dank (push to talk) 22:16, 29 November 2013 (UTC)

Yes, I think that's right. It comes down to how strongly the Foundation feels about the subject, and whether it is willing to live with the status quo since most Wikipedians are not adverse to it. Coretheapple (talk) 22:45, 29 November 2013 (UTC)

Hm. Let me ask some questions and make a couple of comments :-)

First, just reflecting back what I think is happening on this page: am I right that there've been a number of attempts lately to make or formalize policy on paid advocacy editing / COI editing / paid editing, and that all of them have thus far failed, except (so far) this one? And Dank, what you're saying is that thus far this one is likely to fail too, except it won't close until 9 December, and so it's still theoretically possible it would pass? That's what it looks like to me, reading this page, and I just want to check to be sure I'm understanding it right.

A couple of comments.

• I think it's true that it's hard to make new policy on enWP these days, because in the absence of near-consensus we default to the status quo. That said, I think I read Jimmy somewhere saying that policy on enWP has historically derived out of practice -- in other words, policy has successfully been enshrined after the fact, following actual working practice, rather than the reverse. I wonder if that points to a way forward here. Going back to the people I mentioned upthread (the article subjects, etc.) in actual practice their edits have been frowned upon. The trouble is that "frowned upon" isn't very useful for them: they want more clarity than that offers. A clear policy would help them.
• I hear you Dank on enWP generally wanting to focus on edits rather than editors when it comes to policy. I get that and I sympathize with it, particularly given that increasing participation on the projects is the WMF's top goal. OTOH, this policy I think primarily addresses people or institutions in the context of an inherent conflict of interest. It would not be saying someone employed by e.g. the Smithsonian cannot edit Wikipedia, but it would be saying she could not edit the article about the Smithsonian, or about her aunt who happens to be a famous talk show host. That seems reasonable to me.
• On Wikipedia policies are determined by the people who bother to show up to discuss them, and I worry we don't have a large-enough group engaging in these discussions. I worry about that particularly on this topic, because it means we risk people with a vested interest (paid editors, and people who might be accused of paid editing) having an outsized effect on what happens. (I'm not saying that's actually happening: it just strikes me as possible.) Not sure what could or should be done about that.
• Last comment. Please don't see a policy as something the WMF "needs" from editors: we don't. That said, I certainly think, both as ED and as an editor myself, that a policy would be a good idea -- for the projects, and for (as I said above) good-faith article subjects and PR firms and people who would like to edit articles about their employers. Clarity is good, especially for non-Wikipedians who can't be expected to have a sophisticated understanding of how WP works :-)

Thanks Sue Gardner (talk) 06:05, 30 November 2013 (UTC)

Sue, I happen to agree with the points you raise above, but I am in the minority. To be frank, pretty much all your points have been made previously and have been rejected by the majority of editors in discussions. If you peruse the RfC discussion toward the top of the page, and the discussions on Jimbo's page over the past few months, where paid editing comes up regularly[6], as well as the several other paid editing proposals that have been discussed and rejected in recent weeks, you will find the same outcome. I also understand that restrictions on paid editing have been periodically discussed for many years and have always been soundly rejected. Thus the chances of this proposal being adopted are nil.

I think that it's a mistake to think that the opposition is confined to or dominated by paid editors. True, paid editing is not disclosed so we don't know for sure. Also true, on Jimbo's page one of the most vehement opponents (in a now-archived discussion) was an editor who had accepted money in the past to edit an article. But he was also an administrator and a respected one, and that's important to mention. The opposition to paid editing strictures has been dominated by administrators and experienced users who oppose this proposal and similar ones on principle. It's also a mistake to think that the lack of support for these proposals is caused by not enough people discussing it, as this and other paid editing proposals were advertised on every Wikipedian's watch list for weeks.

The point is often made by opponents of this and similar proposals that we're pretty much "done" with this issue and I tend to agree with them. I feel that the ball is in the WMF's court. Until a couple of weeks ago, when I saw that establishing a policy was futile, I had a lengthy section on the evils of paid editing on my user page. But today, if you go to my user page, you can see that that's gone, and it's reduced to a discussion of why paid editing is a reputational issue for the Foundation. My feeling is that people like myself should not get worked up into a tizzy, and waste a lot of time over a futile issue like this if you're not willing to do something about it by making appropriate changes in your Terms of Use to protect your brand name and franchise. Coretheapple (talk) 15:02, 30 November 2013 (UTC)

There is still hope that this proposal can be worked into a more acceptable form to a wider cross section of editors. WMF has already demonstrated that recourse to available legal measures will be made in the case of egregious efforts by large scale PR ventures that threaten to commercialize and undermine Wikipedia.

To that end, I've slightly revised the proposal to more fully address the objective of facilitating good faith contributions in the case of a potential COI.--Ubikwit ^連絡 _{見学/迷惑} 16:32, 30 November 2013 (UTC)

(edit conflict)Frankly, this proposal is not that different from the first one except that it uses an even broader definition. I'm assuming this was an attempt to address some of the concerns that paid advocacy is not the only type of COI. Unfortunately it does that in a fairly ham-handed way that tries to treat all cases of COI exactly the same. So the cousin of a BLP subject is treated the same as someone engaged in a lawsuit against him. A company's janitors would be treated the same as their PR department.

But the biggest unresolved problem is that of enforcement. Any proposal that focuses on the editor rather than their edits is going to have this problem. WP:OUTING, which probably has really strong community support, says that digging up off-wiki info about an editor is bad. Since a paid advocate would have to be a moron to self-disclose when the policy effectively prevents them from doing anything (as this does), enforcement would have to rely heavily on assumptions, which leaves it open to abuse. The proposal here mentions CheckUser, but it's debatable whether confirming suspicions about this would even be allowed by the m:CheckUser policy. It would rely on questionable reasoning: The CU policy allows identifying information to be revealed when necessary to stop disruption, so if we create a policy saying paid editing is automatically considered disruptive, we can reveal personal information of paid editors. This isn't like sockpuppetry where it can be confirmed without revealing any information. If they're editing an article on a company, and CU says they're probably editing for pay, it's pretty easy to put 2 and 2 together to see that they're editing from that company's network. Which means we know their employer and probably their location.

I think the earlier proposals were right to try to separate actual paid advocacy from other types of COI editing. They just did a bad job of defining it. For more mundane types of COI (low-level non-PR employees, BLP subjects, minor stockholders), the current COI guideline is probably sufficient. A new policy needs to just focus on actual paid advocacy (ideally all advocacy, but that may not be practical) - people whose compensation or continued employment are actually contingent on editing Wikipedia. Not people who are being paid by a company and also happen to edit the article about it in their spare time, but people who are being paid by a company to edit their article. Previous proposals have treated these 2 groups exactly the same, which doesn't make any sense. It would be like gun control laws treating a hunting rifle the same as tank. One is probably mundane enough that most people can be left to self-regulate. The other obviously needs some oversight. Mr.Z-man 16:42, 30 November 2013 (UTC)

(ec) This particular page wasn't one of the COI discussions mentioned in the watchlist notice (see this diff), but those pages did point prominently to this page, and this page is currently advertised through RfC and WP:CENT. Otherwise, I basically agree with what Core just said until the last paragraph; I don't think it's a settled question that there's nothing Wikipedians could possibly agree on. Sue, your point is well-taken that we know exactly how the community responds in many cases, and it might be a good idea to write that down ... for instance, if someone winds up making millions of dollars operating a business that advertises that they'll "manage" how companies appear in Wikipedia, I don't think there's any mystery how Wikipedians are going to react to that. I haven't seen a discussion yet that was intended to figure out where everyone is willing to draw the line; all we know so far is that the seven recent proposals (including this one) haven't even been close to the levels of support traditionally needed for a new community-generated policy page. My personal belief is that, compared to past discussions on COI, the job here is simplified by the fact that the various supporters seem to be agreed that nothing less than a new policy will do ... we know what policy looks like, it has to be responsive to all points of view, and we've got a big pile of POVs reflected in the seven recent discussions. All that's needed is for someone here (not me, unless someone else offers to close) to get to work on asking a representative cross-section of the opposition where they personally would draw the lines, and see what we can get a strong consensus on. That could theoretically happen before the close date on this RfC. - Dank (push to talk) 16:54, 30 November 2013 (UTC) (P.S. Z-man and Ubikwit, what you're saying is probably helpful too, but I can't respond pro or con of course. Isaac, thanks for the note on my talk page ... I'm not backing off, I can help, I'm just saying that progress seems unlikely unless someone surveys opinions and pares the page down to something there's wide agreement on.) - Dank (push to talk) 17:28, 30 November 2013 (UTC)

Dank, there's no question that the community has dealt with this issue in a fragmented manner. Still, I think that it's useful to examine what happened to Wikipedia:Paid editing policy proposal. This proposal would have actually permitted paid advocacy as long as it was disclosed. I and other editors who favored strictures on paid advocacy opposed this proposal, because it would have validated the status quo and permitted paid editing. Despite the fact that it would actually facilitate paid editing, it went down to defeat not because it was too weak, but because it was too strong, because it singled out a class of editors, those with COI, with a disclosure requirement. Not even that could be agreed to. So by all means let's try to reach consensus on a proposal; but let's not be under any illusions that what will emerge from such a discussion would in any way resemble what we have here. Given the feelings of the community, in all likelihood what I think we'd get (if anything; I think "no consensus" is more likely) would be a weaker version of Wikipedia:Paid editing policy proposal, with disclosure encouraged but not required. Again, if this is what the owner of the Wikipedia brand is willing to live with, I for one am not going to try to be more Catholic than the pope. It's really up to Sue and the WMF board. If they want a rule banning paid advocacy, they need to act at the Foundation. If they don't act, nothing will be done here. The community has already spoken loud and clear on this issue. Coretheapple (talk) 17:19, 30 November 2013 (UTC)

(ec) As for your hypothetical: There's no question that in any field there are going to be smart players and stupid ones. The smart ones will engage in paid advocacy while paying lip service to Wikipedia policies, and indeed will cynically use Wikipedia policies on behalf of their clients, to spin articles for themselves and against their competitors, and will use those same policies against editors who have problems with the activities of paid advocates. That's been the case in the past, and in fact it is the kind of thing that got me interested in this subject matter. Coretheapple (talk) 17:34, 30 November 2013 (UTC)

@Mr.Z-man It seems to me that almost all of your arguments are strawmen. First and forememost, this policy is aimed at prevention of the egregious COI editing and facilitating good-faith editing in the case of a potential COI. Accordingly, enforcement is not a prominent issue because failure to comply is already a violation--as you indirectly point out, in a somewhat suspect manner, by stating that "a paid advocate would have to be a moron to self-disclose when the policy effectively prevents them from doing anything (as this does)". The policy doesn't "prevent them from doing anything"; it requires them to follow certain specified procedures to contribute to corresponding (i.e., COI) articles. Do you have a problem with that?

The point of this policy is to implement the content of the guideline plus reinforcements based on recent developments in a succinct manner that prevents people from attempting to game the system by claiming that Wikipedia only has a guideline for COI, not a policy.
WMF can file lawsuits, but that is an excessive and costly measure against individuals. This policy--which needs work--would effectively solve more than one problem at once.--Ubikwit ^連絡 _{見学/迷惑} 17:59, 30 November 2013 (UTC)

I agree with you that this policy would be helpful, and that it would aid greatly the many legitimate organizations that want to do the right thing, as Sue has pointed out. However, I think it's a question of whether the subject has already been adequately discussed, and the community has already made a decision to not enact a policy. A search of Jimbo's talk page finds 145 hits under the search term "paid editing"[7]. Clearly this has been before the community amply. And then we have the policy proposals that have failed. Coretheapple (talk) 18:24, 30 November 2013 (UTC)

Coretheapple, a little ways up, you referred to an earlier proposal that "would have actually permitted paid advocacy as long as it was disclosed. I and other editors who favored strictures on paid advocacy opposed this proposal, because it would have validated the status quo and permitted paid editing. Despite the fact that it would actually facilitate paid editing, it went down to defeat not because it was too weak, but because it was too strong...". I observed that discussion, and it seemed to me that it went down to defeat because of a combination of opposition due to being "too strong" and opposition due to being "too weak". It seems to me that one of the impediments to success has been the insistence by the various proposals' strongest supporters on either something that is sufficiently "strong" or nothing at all. (See our discussion above, where you pointed to where an edit you made, in response to my concerns about BLP got reverted.) It's a given that there is going to be community opposition from users who are simply opposed to any policy along these lines, but I'm not convinced that there is actually a consensus that the status quo is good enough and that no improvement should be implemented. Please let me suggest that a partial solution may be able to get consensus, if the strongest supporters will "settle" for it, and it may be better than nothing at all. --Tryptofish (talk) 19:56, 30 November 2013 (UTC)

Oh dear Lord no, I beg to differ that the Paid Advocacy proposal was defeated by a combination of "too weak" and "too strong" !votes. Of the 38 opposes[8], I count only six thought it was too weak (me, Robert RE Harvey, Figureofnine, Mongo, Slimvirgin and Cush). Overwhelmingly, the sentiment was that the rule was either too much, unecessary, duplicative or unenforceable. Coretheapple (talk) 20:11, 30 November 2013 (UTC)

Thanks, point taken. But I still think there's a case that 32 opposes would have been better than 38. Let's take it a step further. That proposal ran into opposition (from some of those 32) on the grounds of requiring disclosure. I can imagine proposals that would still center on disclosure, but with some flexibility, some limiting of the requirements. Such proposals would pick up quite a few supports from amongst those 32 remaining opposes; even the closing statement pointed that fact out, in effect. The question then becomes how much opposition would come from the 6, and from previous supporters who sympathize with those 6. I'm suggesting that such opposition is a matter of letting the perfect become the enemy of the good (or good enough). --Tryptofish (talk) 20:40, 30 November 2013 (UTC)

No, I think it's more a question of the meaningless being the enemy of the not meaningless. But as I said, it's really up to the Foundation to articulate what it wants and what is acceptable to it. I think that editors here have better things to do than to argue for a restriction on paid editing that the Foundation, as the party most affected, does not feel is necessary. Coretheapple (talk) 20:52, 30 November 2013 (UTC)

Core, for what it is worth, I have made suggestions similar to Tryptofish's to you... Jytdog (talk) 22:26, 30 November 2013 (UTC)

And it's a pity, really, because no consensus to do anything seems to me to be the epitome of meaninglessness. --Tryptofish (talk) 22:48, 30 November 2013 (UTC)

(ec) What I'm suggesting is that the Foundation has a responsibility here, to itself, its mission and to Wikipedia readers, one that cannot be delegated to volunteer editors who have collectively demonstrated for many years no interest in effectuating a sound conflict of interest policy. Coretheapple (talk) 22:54, 30 November 2013 (UTC)

but see this is part of what you do, with this kind of extreme black and white view and framing of the issue. there has never been "no interest"; that's not accurate or helpful.... Jytdog (talk) 23:01, 30 November 2013 (UTC)

Of course it's accurate and to an extent it is black and white. We don't have a conflict of interest policy, and while there are individual editors who are interested, they are in the minority, and collectively Wikipedia editors do not want one. The proof is that this is not a new issue but an old one, and one has never been adopted. I hate to burst your bubble, and I'm not going to get into another "yes I'm right and no you're wrong" argument of the kind you frequently commence, but that's a fact. Coretheapple (talk) 23:17, 30 November 2013 (UTC)

Ubikwit "this policy is aimed at prevention of the egregious COI editing and facilitating good-faith editing in the case of a potential COI" Until a couple hours ago, it never even mentioned that. And the rest of the proposal still basically says otherwise. It continues to list family members in the same "list of COIs" as lobbyists and PR firms. It does not provide any avenue to facilitate good faith editing. It basically says that any user with a COI, as broadly defined by that list, cannot make any edits other than vandalism reversion and the removal of libel (and then still requires them to state their COI if they even dare to do that). And then says that "intentional violations of this policy may result in an editor being blocked" again with no distinction between good faith editors and PR firms. Anything that automatically treats BLP subjects with equal suspicion to PR firms is not assuming good faith.

"failure to comply is already a violation" - Right, but, the only way to prove that they're not complying (if they don't disclose their COI) is to prove that they have a COI. And if they don't disclose the information, it is effectively impossible to do that within the confines of WP:OUTING.

"in a somewhat suspect manner" - Suspect? What is that supposed to imply?

"Do you have a problem with that" - Obviously. If we want people to disclose their COI, there needs to be some benefit to them to do so. If we have one policy that says "COI editing is basically forbidden" and another policy that says "It's against the rules to dig up off-wiki info about another user (such as where they work or who they are)," then it's kind of in their best interest to keep their COI a secret, since then they can continue editing the article as long as they follow NPOV/RS, etc. But once they've declared their COI, then even if they can demonstrate that they can follow content policies they're still prohibited from directly editing the article.

I have dealt directly with BLP subjects and companies on OTRS. These are usually not "high-profile" people or Fortune 500 companies. They're band members and small businesses that don't have publicists or PR firms working for them and need to deal with problems themselves. Given the sheer volume of complaints, OTRS members don't have time to do all the research and editing themselves for every complaint. Unless it's an easy fix like removing unsourced libel or they say exactly what's wrong and provide a source for the correct information, we usually give them advice on how to work within the community. Our current general advice (it comes up often enough that we have a standard response template) is based on the current COI guideline and states that it we "encourage" them to use the talk page rather than edit directly. I, personally, would be rather uncomfortable telling a BLP subject that it is required, and that doing anything else would get them blocked. Many people try to deal with the issue on their own before contacting us. It's an often-confusing experience with a steep learning curve. If they get blocked, they're just going to be even more confused and infuriated when they contact us and it will be far more difficult for OTRS volunteers to deal with the situation. An overly strict and/or overly broad COI policy has the unintended potential to create an environment that is hostile toward article subjects getting involved in editing. Mr.Z-man 22:03, 30 November 2013 (UTC)

Mr Zman, you seem to be ignoring the language in the Administration section, that lays out a procedure for dealing with concerns about COI editing and makes it clear that going outside that procedure opens the concerned editor to sanctions. Your concern about witch-hunting is shared and has been addressed. Jytdog (talk) 22:26, 30 November 2013 (UTC)

The "Administration" section is rather open-ended and somewhat contradictory with regards to the rest of the proposal. "Take the user to WP:COIN" Then what? Without any evidence, it's unprovable. And unless the user chooses to disclose, there can never be any hard evidence, which again, means under this proposal it's generally never in a user's best interest to disclose their COI. Currently, bringing a dispute to COIN requires that the user in question be doing something else wrong, generally POV editing, or for cases where the user already has disclosed their potential conflicts and it just needs to be decided whether it's an actual COI or not for a specific situation. It's not really designed for investigations. The proposal says (in the administration section) to focus on the edits, not the editor, but it also says (in the intro) users can be blocked for violating the policy, even if (presumably) there is nothing actually wrong with the contents of their edits.

Though really, this being used as a weapon against article subjects is just a corollary of the core problem with this proposal: That it appears to elevate COI above content policies, allowing users to be blocked for making edits that are fully compliant with content policies, just because they have a COI.

And while a user who harasses or outs an established user, who knows how to file a complaint in the right places might get sanctioned, when it comes to a dispute between an established user and a new user, we all know who usually wins. I don't think the current wording is sufficient to be anything more than lip service. Most of these types of disputes are not going to be on highly-watched pages. They're going to be on low-traffic talk pages between the article's subject and 1-2 established users. While most policies can be abused, one that says users can be blocked for making edits that are otherwise unobjectionable and non-disruptive, so I think it is especially open to abuse. Mr.Z-man 00:40, 1 December 2013 (UTC)

OK, the policy needs some work, but the goal of facilitating good faith edits has not been adequately articulated, and the policy is not intended to promote some sort of vigilantism or witch hunt. Most people that aren't paid advocacy editors would likely simply follow the procedural guidelines set out in the policy. You are probably correct that paid advocacy editors and the like aiming to circumvent policy are still going to attempt to do that, but that is not a reason not to implement a policy that explicitly prohibits such activity on Wikipedia.

I don't know enough about the outing policy and don't have time to put into that, so hopefully you and others can further suggest ways to address those concerns. --Ubikwit ^連絡 _{見学/迷惑} 03:14, 1 December 2013 (UTC)

If there's no significant progress in finding which elements have the kind of broad support necessary for a new policy page in the next couple of days, then I'll suggest we get to work on a collaborative closing statement, and hopefully finish by the 9th. It seems to me there are some noncontroversial and potentially effective things we can offer the supporters so that they don't walk away empty-handed. - Dank (push to talk) 15:46, 1 December 2013 (UTC)

I think there are a few key aspects that a policy can address:

• How to identify if you have a potential or apparent conflict of interest
• How to manage your potential conflict of interest, so it doesn't become an actual conflict of interest
• How to manage your apparent conflict of interest, so it isn't unduly disruptive to either the editing community or Wikipedia readers
• What should be done to manage actual conflicts of interest that arise

I think there is a reasonable amount of common ground on the first three aspects, as evidenced by the degree of support enjoyed by Wikipedia's guidelines on conflicts of interest (WP:COI), whose first four sections deal with these points. Thus I think it would be possible to pursue further discussions on developing these portions of WP:COI as desired, with the view of making them into a policy. isaacl (talk) 00:24, 2 December 2013 (UTC)

Thanks, Sue. Also, thanks Isaacl, Dank, etc. for thoughtfully responding to the discussion here and elsewhere. I'm sure there is a lot of common ground from which something can be constructed. My original idea (at the top) was indeed to implement a minimal policy. By minimal, I meant: what most of us can accept. I believe all issues are intelligible. That is, if a proposal rules out too much, then it can be rewritten to avoid such negative consequences. If we can identify problems with proposals with our minds, then our minds are also capable of rewriting such proposals in order to make them avoid the problems. This aspect of intelligibility just seems like a truism to me. E.g., if almost all of us agree that editing articles when one has certain serious conflicts of interest should be discouraged and prevented by policy, then we can write such a policy which captures just enough and not too much.

From my side: Sue Gardner's point about enforceability is important. An argument that a policy is unenforceable does not seem to be an objection at all. This proposal (as almost all are) is independent: It does not prevent any other policies from being enacted. If the policy is unenforceable, then the worst that can be said about it on that point is that it has no positive effect, not that it has a negative effect. But Sue rightly identified a positive effect: Most users are good-faith editors and will follow policy to the best of their ability, even if they could get away with not following it. And editors who may otherwise want to flout community standards would have no excuse with which to justify their actions in the public eye. Editors have no policy which can be cited to inform users who may otherwise take part in paid advocacy. This is not a merely academic point. The person who ran WikiExperts, a paid advovacy firm, explained his not following the WP:COI guideline explicitly: It is just a guideline, and those are just suggestions that do not have to be followed, because they are not policy [9]. This is also the excuse that paid advocacy editor Michael Wood of Legalmorning uses [10]. And the infamous Wiki-PR says the same thing [11]. If there was a policy, then this excuse cannot be made. They need such excuses in order to legitimize and sell their services to clients, the vast majority of which want nothing to do with black hat practices that take advantage and infringe the policies of a charitable organization.

More from my side, back to minimal: Compromise is important for the creation of consensus. The position that no limits at all should be placed on serious conflicts of interest editing (because we should look at the "edits, not the editors") is not a compromise position, it is an extreme. Here's an example of a compromise position which I believe everyone on my side would support: "Do not edit an article if you are paid by the subject of the article to do so, unless you are correcting vandalism or a BLP violation." This captures only a selection of paid advocacy edits. Yet it seems that many on the other side do not even accept this most minimal limit, which I take to mean that they accept no compromise whatsoever, but maybe I'm wrong. --Atethnekos (Discussion, Contributions) 01:54, 2 December 2013 (UTC)

(edit conflict)@ Ate – Here’s why such language when used in policy is problematic and why many not “on my [your] side” object to it: Here's an example of a compromise position which I believe everyone on my side would support: "Do not edit an article if you are paid by the subject of the article to do so, unless you are correcting vandalism or a BLP violation." “If you are paid” – this is very vague language, what does it mean? People with jobs get paid. Does the payment have to come directly from the subject of the article? or a better question would be: How far removed from the payer does the payee have to be to not “be paid?” “The subject of the article”. The only subject of articles that can “pay” someone are living humans and organizations, yet the subject of an article can be all types of things—organizations, living or dead humans, events, biological organisms, diseases, products, et al. etc. etc., none of which except organizations or living humans can pay anyone. Who is “paid” and what is a relevant “subject of the article” is still very vague and open to COI inquisitions under such language. --Mike Cline (talk) 02:18, 2 December 2013 (UTC)

Take it under the most minimal interpretations: Paid by the subject of the article means paid directly by the subject, receiving money or kind. Yes, indeed, the actual subject of the article. The subject of the article Plato is Plato. The subject of the article Stephen Hawking is Stephen Hawking. The subject of the article British Petroleum is BP. If the subject of the article cannot pay the editor, then obviously such a minimal rule would not apply, because no editor would be paid by such a subject. --Atethnekos (Discussion, Contributions) 02:42, 2 December 2013 (UTC)

So under the above interpretation, it would be perfectly legitimate for BP to pay a Wikipedia editor to edit the article Deepwater Horizon explosion, since this is an event, (the subject of the article in this case would not be the payer--BP). Am I correct in this logic? --Mike Cline (talk) 03:10, 2 December 2013 (UTC)

It wouldn't say that anything is perfectly legitimate, because it says nothing about legitimacy at all. The rule would simply not apply to that case, because it doesn't affirm the antecedent. --Atethnekos (Discussion, Contributions) 03:25, 2 December 2013 (UTC)

I suspect the COI police would take an extremely liberal interpretation of the language and absent any clear, unequivocal agreement and interpretation of what "paid" and "subject of the article" really means, it is just a license to attack any perceived COI on the basis of a vague policy. My two cents. --Mike Cline (talk) 03:39, 2 December 2013 (UTC)

An object-oriented programming approach would seem to be capable of putting the "subject" in focus. The Deepwater Horizon explosion, for example, would be considered a subclass categorized under Business activities of BP (incidents), for example.

It is obvious that BP has an interest in portraying its activities in a positive light for business purposes, so the presence of a COI is clear. It doesn't seem that there are many legitimate semantics problems that would prevent the implementation of a minimally worded policy. The objective of the policy is preventative, so leaving the scope for interpretation broad is desirable, as most of these questions are easy to flesh out in terms of the presence or absence of a COI.--Ubikwit ^連絡 _{見学/迷惑} 15:56, 2 December 2013 (UTC)

As I pointed out in the section above, using "subclasses" does not solve the problem. There's still potential issues with companies that are "affiliated with" but not actually operated or owned by another company and companies along a supply chain. Then there's the issue with government agencies. Would an employee of the US Department of Agriculture have a COI on the US State Department article? They are, for most intents and purposes separate organizations, but the money for both comes from the same place. Would a doctor have a COI on articles about the insurance companies his practice accepts? He is being paid pretty directly by them, but he doesn't really have any sort of stake in the companies' future. In general, I agree with Mike that leaving it vague leaves it open to abuse. And, it still has the same problem that I've mentioned several times about treating all employees the same. A contracted PR firm, marketing department, or partner/owner has their continued employment and/or compensation directly based on improving the image of the company. They may even be specifically hired to use Wikipedia to do that. Contrast this to a "typical" employee like a secretary or a bookkeeper, who, unless the company goes out of business, probably makes the same amount of money regardless of how well the company does. Obviously the former has a much bigger COI than the latter, but most of the proposed policies would treat them exactly the same. Less than a third of Americans even like their jobs, so why assume that everyone is going to want to promote the company they work for? Mr.Z-man 15:07, 3 December 2013 (UTC)

You're right about the supply chain etc. In a situation like Deepwater Horizon explosion#Disposition of financial obligation, each company may have a reason to brief against the other, and their competitors can gain commercial advantage from discrediting them. - Pointillist (talk) 16:57, 3 December 2013 (UTC)

I don't want to open a can of worms here, and perhaps the subject has already been raised (and I apologize if that is so), but you all do realize that BP does have a PR person weighing in on but not directly editing, this and other BP-related articles? This is a situation that is permitted by current rules as well as by this policy proposal, and is troubling enough that it has received widespread negative publicity. Indeed, that is how I became involved in this paid-editing maelstrom in the first place. Coretheapple (talk) 17:51, 3 December 2013 (UTC)

@Coretheapple, I'd be interested to hear your comments on Could professional editing be made safe?, above. - Pointillist (talk) 22:41, 3 December 2013 (UTC)

If one accepts the premise that one of two primary objectives of a COI policy would be to preserve the editing environment of Wikipedia by preventing COI editing (i.e., its tendentious form), then a minimal approach based on the concept of negative liberty would seem to be the only feasible option, and would offer the advantage that it becomes unnecessary to define every potential case of COI because that would be analyzed on a case-by-case basis at AN/I or the like when a problem arises, with recourse to the more detailed guideline.

What the policy would produce is a enforcement measure, which the guideline doesn't. Whether that will streamline the process of dealing with COI editing is open to question. The reason being is that the manner in which to deal with COI editors intent on violating the policy by concealing their COI and directly editing articles probably doesn't change much. In that respect, the policy would primarily serve to help good-faith editors avoid COI problems. If it were to hel prevent the launching of Arbcom cases--such as the BP related case--then the advantages of having a policy are manifold.

Regarding the BP case discussed above, one would presume it obvious that all of the companies in the "supply chain" would have a COI in one form or another (e.g., items 4 and 5 on the "Behavior of editors with a COI" list), and should therefore not be editing related articles directly. Accordingly, basing the wording of the policy provisions on a the basis of a "freedom from...(tendentious COI editing)" concept facilitates the analysis of such COI by not constraining the scope. The present wording seems to facilitate the above even considering the use of the term "subject", though it will likely be improved upon in due time. Maybe introducing the concept of "agency" linked with "subject" would improve the text.--Ubikwit ^連絡 _{見学/迷惑} 00:58, 4 December 2013 (UTC)

Sue, can you clarify for everyone that you are commenting here as an ordinary member of the community - your opening comments that "As ED of the WMF ..." might make it appear that you are speaking on behalf of the Foundation. If the Foundation do wish to enter into a dialogue with the community on the matter of COI editing, that needs to be made explicit and much more public, and a new discussion started with appropriate community notification. SilkTork ^{✔Tea time} 16:30, 4 December 2013 (UTC)

Yes, that's a good idea. I was under the impression she was stopping by as WMF director, and she seemed to be saying that. Coretheapple (talk) 19:16, 4 December 2013 (UTC)

Hi SilkTork and Coretheapple. Yes, I am commenting here as ED, and what I've said here is informed by my experiences both as ED and as an editor. That said, I'm just speaking informally here -- if I were going to announce a decision or a WMF action or anything like that, I wouldn't do that casually like this. Thanks Sue Gardner (talk) 19:41, 5 December 2013 (UTC)

Sure, I think you were pretty clear on that. Hope you don't feel you were beaten up too much in the process; I'm sure that everyone was glad to hear from you. Coretheapple (talk) 21:38, 5 December 2013 (UTC)

Sue, I thank you, and am moved to put more cards on the table than I usually do. We met and spoke occasionally when I was a volunteer at the WMF offices in San Francisco a few years ago working with Cary Bass, who interviewed me for the post with Mike Godwin. I truly enjoyed my year as a part-time intern at the WMF, and only the fact of my move across the country prevented my continuing.

I say this not for my self-aggrandizement but to establish that I have a perspective few can claim: someone who has nearly 58,000 edits as a Master Editor, as well as spending nearly a year deep in the confidence of the WMF with access to IRC and OTRS. I think I can honestly say I have a fairly good overview of the totality of the project.

It is my strongly-held belief that the greatest threat to Wikipedia is paid COI editing. There is a growing and arguably justified perception that the encyclopedic values here are under attack by moneyed interests of various types. Coretheapple, who has spent much more time on this than I have, is completely in the right of it. The community cannot make this decision - we do not own the website. The WMF is the only body that can enact comprehensive policy, with strict enforcement provisions. Allow me to add that this policy should begin at home: the WMF itself should not accept donations over a cap of, say, 1,000 dollars. Money, power, and information access are intertwined in corrupting tangle, and the WMF needs to cut the Gordian Knot in the coming year. All other issues are secondary here, and indeed mostly radiate from this one. Thanks for your obvious concern and great work over the years. Jusdafax 22:02, 5 December 2013 (UTC)

Hi Coretheapple and Jusdafax. No, I don't feel beaten up at all. I've been reading this discussion with a lot of interest -- I'm super-busy lately but have been paying attention here. And I totally hear you, Jusdafax, and I can't say that I disagree. I don't know if paid COI editing is the single biggest threat to Wikipedia, but I do think it's a very serious threat. I say that because I was a journalist for 17 years, and so I know that for an information resource, credibility is everything. When Wikipedia first launched people were dismissive of it, and at the time that was reasonable -- it was new and unproven, and in the early days obviously not very comprehensive or high-quality. In my six years running the WMF though, I've seen people's trust in us grow as Wikipedia began to prove itself as a responsible source of unbiased information. We are not always correct, and people do sometimes see vandalism or errors, but they have an enormous amount of love for Wikipedia anyway, and that's because they trust it to always be trying to tell them the truth, untainted by whitewashing and commercial interests. Everywhere else in their lives, people are trying to spin and propagandize and sell to them, and in its neutrality, Wikipedia is really unusual and refreshing.

We *do* risk losing that public trust, IMO. When people see headlines like Wikipedia's Paid edits: How To Make Money, the WikiWay, Is the PR Industry Buying Influence Over Wikipedia? and Wikipedia being spun by spin doctors -- yeah, I worry this will damage people's confidence in us, perhaps irreparably.

So I thank you for sharing your views here. I believe it's the community's responsibility to make editorial policy, and that is what I would like to see happen. But at the same time, I recognize that the community is having difficulty here, and I realize (or at least I believe) that it'd be near-impossible for the community to come together and develop a cross-language policy, which may actually be what's needed. I am glad to see SJ here too: I think it's important for the WMF Board of Trustees to be paying attention to this, as well as me. Thanks Sue Gardner (talk) 03:54, 6 December 2013 (UTC)

I'm no expert on Wikipedia, just a relative newcomer, but here's an analogy that occurred to me. I was watching on DVR a History Channel special about the Kennedy assassination from a few days ago. Seems that a ridiculous percentage of the American people believe that a conspiracy claimed Kennedy's life. Now, imagine if you crowdsourced the Kennedy assassination, to find out who did it. You'd get everything but the truth, which people don't like, but which is that Oswald did it. You examine the evidence, and that's where it brings you. I think there's a rough analogy here with the effort to formulate a policy on paid editing. If you crowdsource it to the Wikipedia "general public," you'll find that opinion is all over the lot, that only a minority favor banning paid advocacy editing, just as only a small minority of the American people believe that Kennedy was killed by Oswald. The general median of the positions expressed, the "consensus," is that "we don't need a policy." That's the crowd's answer, but you have to decide if that is the right answer. Similarly, it's been shown in the past that when you poll the general public you find that people aren't so hot on the Bill of Rights. Some things just can't be crowdsourced - like your good name. Coretheapple (talk) 04:25, 6 December 2013 (UTC)

Distinguishing two issues

I like all of Mr.Z-man's comments on this page, and wanted to call out / amplify two of them:

A policy is needed for narrowly-defined paid advocacy

A new policy needs to just focus on actual paid advocacy ... people whose compensation or continued employment are actually contingent on editing Wikipedia. Not people who are being paid by a company and also happen to edit the article about it in their spare time, but people who are being paid by a company to edit their article [which] obviously needs some oversight.

A good proposal in this area can admit up front that enforcement is difficult: this is for paid advocates who want to do the right thing (of which there are many examples). Puppeteers will be covered by existing no-puppet policies. This would only cover how people whose compensation or employment are contingent on contributing to Wikipedia for their own company or for a client - and want to follow policy - should contribute.

On the supply side, such a policy would cover any groups that are found to be advertising their WP-editing services. On the demand side, it would cover any groups that are paying or advertising for contractors to update Wikipedia's content about themselves. When we do identify such contractors or employers, we should be able to point them to a clear policy and they should know whether they are in compliance or not.

I think it's possible to reach some sensible simple policy here mainly because some of the supporters of the idea are active paid editors and CREWE members, and because a majority of the opposition to recent proposals (such as Z-man's) have been for reasons related to gray areas of COI and policy creep.

Any COI-related policy should be a subset of the current COI guideline

A COI policy should briefly describe what is already accepted practice. It shouldn't try to alter the COI guideline. The guideline should by definition be stronger - while following the policy is almost always appropriate, guidelines may at times be ignored. – SJ + 01:39, 3 December 2013 (UTC)

I like the supply/demand sides distinction, because it is important that companies considering hiring PR firms on their behalf understand the parameters.

I think there are problems with two other points, though.

First, if you constrain the scope strictly to "paid advocacy editing", then you have to define all of the forms thereof. As I've mentioned above, that is generally cumbersome and ineffectual, which is one appealing aspect of using the "freedom from..." approach to define policy. It would also appear that the degree of efficacy with respect to prevention would be adversely affected by adopting such an approach. Is there any reason why requiring a company employee 'editing their company's article in their free time' to do so indirectly would be problematic?

Second, since policy per se is a superordinate category to guideline, the subset classification is inoperative. If, through working on the policy, aspects of best practice are elucidated in a manner not reflected in the guideline at present, it's easy to adjust the guideline. I don't think that the issue has been raised that the policy under development is overstepping the accepted bounds of current common practice. It is trying to redefine them in manner that facilitates their promulgation as policy. --Ubikwit ^連絡 _{見学/迷惑} 16:36, 6 December 2013 (UTC)

I think that's a good idea. However, it needs to be broad enough to cover corporate employees who are dispatched to Wikipedia for the purpose of editing their articles or creating articles. But I think the the general idea of zeroing in on a particular type of abuse is very worthwhile. Also I was not aware that SJ is on the WMF board. Thanks for coming here and participating. Coretheapple (talk) 16:44, 6 December 2013 (UTC)

I could be mistaken, but wasn't this idea of targeting paid advocacy editing only already run through the mill once with the failed "No Paid Advocacy" RfC? Since you seem to shifted from a position the either WMF promulgates a policy or Wikipedia is doomed to supporting a community generated policy, why not try to contribute something to the text of the proposed policy.

Part of the motivation for this minimal COI:limit" proposal was undertaken in an effort to overcome the problems with the other RfCs. The multiple discussions have been too much for me--and others, no doubt--to follow in detail, but we have to avoid going around in circles with this.

I was in favor of the "No Paid Advocacy" proposal, and would obviously support another attempt if this failed, but it would have to represent progress with respect to the specification of the provisions of the policy. Seeing more effort being put into actual text of provisions corresponding to any of the above-described suggestions might help ground the discussion.--Ubikwit ^連絡 _{見学/迷惑} 23:58, 6 December 2013 (UTC)

That's right. Such a proposal was rejected by the community. I don't see the harm of another attempt. That way WMF people can participate if they wish, or at least observe the group dynamics, and determine if it is satisfied with what is done or not done here and decide whether to proceed further. Coretheapple (talk) 02:01, 7 December 2013 (UTC)

There is one aspect being overlooked here, and that is the desirability of having a policy that facilitates good-faith contributions by editors with a potential COI, as expressed by Ms. Gardner in her WMF capacity.

As a contrast class, I've just added (to the bottom of the current text) a contracted version of the (copy edited) text limiting the scope to Paid advocacy editing. --Ubikwit ^連絡 _{見学/迷惑} 08:28, 7 December 2013 (UTC)

The problem with the "No paid advocacy proposal" was that it used such a vague, broad definition ("close financial relationship") that it could be construed to apply to almost any employee of a company and family members of BLPs. That, and it had the issue of focusing on the editor rather than the edits by prohibiting a class of editors from editing articles directly without regard to their ability to follow content policies. Mr.Z-man 14:55, 7 December 2013 (UTC)

I don't recall the wording of that proposal and don't have the time to check it, but I believe that the wording of this one, now that I've copy edited it some more, is quite tight and approaching viability with respect to adopting the broader scope that includes the dual provisions for facilitating good-faith COI editing and preventing pernicious paid advocacy editing.--Ubikwit ^連絡 _{見学/迷惑} 15:39, 7 December 2013 (UTC)

The RfC at the top of the page does not seem to favor this wording, however. I'm not sure what's to be done about that. The closer (very wisely, I think) decided to keep it open for a while longer. But overall, reading all the comments, one doesn't come away with a feeling that this approach has a consensus. Coretheapple (talk) 16:35, 7 December 2013 (UTC)

Well, you're right about that, but the expanded definition was originally added by DavidinNJ, I believe, and he restored it after I once reverted the addition (on November 17). That was quite some time ago, and before Sue initiated this section. I suggest you review the evolution of the text under discussion as well as the evolution of the discussion. Since you contributed some text after that series of edits, what approach, exactly, is it that you would be interested in seeing passed. I'm sort of getting tired of the abstractions here by people that are beginning to sound more like pundits than editors.--Ubikwit ^連絡 _{見学/迷惑} 17:31, 7 December 2013 (UTC)

Changes to criteria

Regarding this edit: I believe the modifications do not have broad support in the discussions that have been held so far, as they are overly expansive. If there is agreement on keeping this change, though, then I propose the following wording to make it easier to follow:

An editor shall not edit an existing article directly if he or she is:

1. acting on behalf of the subject of an article;
2. acting on behalf of a business partner of the subject of an article;
3. engaged in competition, litigation, or lobbying for or against the subject of an article or an individual;
4. acting on behalf of anyone or any organization associated with the subject of an article; or
5. is paid by anyone in the above categories.

My personal preference would be something more along the lines of my previous suggestion, based on the definition within the current conflict of interest guideline. Rather than trying to enumerate every situation, I believe the criterion should focus on the key aspect: does the editor have competing interests that result in a potential or apparent conflict of interest? A list of examples can be given afterwards. isaacl (talk) 18:20, 7 December 2013 (UTC)

Re No. 4, I'd suggest tweaking to read something like: "acting on behalf of anyone or any organization associated with or in opposition to the subject of an article". The same tweak should be made to the alternate language recently placed there, if it is to be kept. I think that what's outlined here seems to be equivalent in purpose while being simpler in construction. Coretheapple (talk) 18:54, 7 December 2013 (UTC)

So if it's an article about an abortion activist, the prohibition would extent to people working for the "People For Choice" and the "People Against Choice." Coretheapple (talk) 18:55, 7 December 2013 (UTC)

(edit conflict)I would suggest simply reverting back to something like what people actually saw when they commented on the RFC. Since it looks highly unlikely that this proposal in anything close to its current form will get consensus, I think it's more important that the page reflect what people actually commented on, as a reference for any future proposals. Trying to make changes to the wording at this point seems more like rearranging the deck chairs. We should be working toward a compromise proposal, not trying to salvage this one. Mr.Z-man 19:02, 7 December 2013 (UTC)

@isaacl The new wording was inserted in light of concerns discussed above with relation to the term "subject", with an analogy being made with respect to BP and the Deepwater Horizon explosion. It is not expansive beyond including such events that can be attributed to a subject under the category of subject. I also eliminated one item on the list to simplify it.

The wording of the original proposal was substantially changed early on to include non-paid COI so I don't see why we should stop working on this now, though it would seem necessary to call a re-vote or launch the current (or revised text) in a new proposal.

The current text would seem to represent progress toward a compromise agreement, and after working on the text a little more and further adjustment of the scope, a new proposal based on it might be in order.--Ubikwit ^連絡 _{見学/迷惑} 02:27, 8 December 2013 (UTC)

The issue is that it's not a good idea to be reactive to the concerns of each individual editor, one at a time. Just because one editor says something doesn't mean the proposal text should be directly amended to suit, since this may not align with the overall consensus (as I do not believe is the case here), and can lead to a patchwork policy, which erodes overall support (as has already happened in each of these proposals). In addition, as stated above, it muddles things regarding knowing which version of the proposal was being discussed in each comment.

I think it would be better to start with the current conflict of interest guideline and have a discussion broken out by section, where improvements can be put forth and debated. Although I appreciate the inconvenience of changing venue once again, it would seem most logical to hold this discussion on the talk page for the conflict of interest guideline. Closers can be identified in advance to evaluate the consensus for changes to the conflict of interest guideline, and once an agreement has been reached, changes can be made to the guideline. If at that point there seems to be sufficient support for making the guideline (or portions thereof) into a policy, then an RFC can be opened. isaacl (talk) 03:14, 8 December 2013 (UTC)

It's not a bad idea to work on the guideline in conjunction with assessing whether there is consensus for the promulgation of a policy. After several proposals, with this the latest appearing to fail, the procedural approach you outlined might be the most efficient course forward. But I would think that the goal should be the establishment of a policy from the start. Otherwise there would appear to be little incentive to undertake the effort. Personally, I have little--if any-- more time to contribute to this undertaking.--Ubikwit ^連絡 _{見学/迷惑} 07:45, 8 December 2013 (UTC)

As has been discussed by Dank, policies generally arise from guidelines that enjoy broad support. They are in effect draft policies, so making the conflict of interest guideline ready to progress to a policy aligns well with the intentions of those who would like a policy to be put in place. For better or worse, discussion amongst a large group of people takes time. (It would be a lot easier if everyone would just agree with person X's proposal, but sadly there's no agreement on who person X should be.) isaacl (talk) 08:12, 8 December 2013 (UTC)

What seems to be an important point related to the pending indefinite result here is that the proposal itself turned out to be a work in progress, and the initial changes made to it were of a scope transforming level.

Without a re-vote or something, this proposal is effectively finished, but the work in progress can be moved to the guideline talk page and built on there. Some of the discussion and text produced here is decent 'raw material'.--Ubikwit ^連絡 _{見学/迷惑} 09:10, 8 December 2013 (UTC)

I don't see the harm of tweaking the proposal insubstantially here and there. For instance, I'm dead set against letting COI editors create articles, but there's no point in my pushing that at the moment. Whatever is done on this, as I said, please fix it so that organizations opposed to the subject are included. Unless someone objects I'll add that language myself. Coretheapple (talk) 15:57, 8 December 2013 (UTC)

Item no. 4 on that list has been eliminated as redundant, so there is nothing to change. Also, it seems that it's down to me to point out that your well-intended efforts to prevent COI from creating articles runs against the momentum favoring the facilitation of good-faith positive contributions to the encyclopedia through the promulgation of a COI policy. That has been one of two objectives that have become focal points.--Ubikwit ^連絡 _{見学/迷惑} 16:30, 8 December 2013 (UTC)

The changes are broadening the scope for a potential or apparent conflict of interest from just the subject of an article to any organization associated with the subject of an article, which is a substantial change from the previous wording. I think it would be better to restore the earlier text until a more definitive agreement is reached.

Regarding your proposed change, I think it is adequately covered by item 3. isaacl (talk) 19:19, 8 December 2013 (UTC)

Looking at the wording - which I agree right now is much too complex - I think the main gap is with No. 2. If I'm the business partner of the subject of an article, I can't write about him. If I'm the business partner of somebody litigating against or lobbying against the subject, I can. We need to be sure that this applies to a conflict of interest in both a positive and a negative sense, to be fair. Coretheapple (talk) 20:40, 8 December 2013 (UTC)

If the business is suing or lobbying against the subject, then the potential conflict of interest is covered under item 3. I'm not sure if there is general agreement on a potential conflict of interest when your business partner is engaging in personal legal action or lobbying against a subject; this would mean you would have to be aware of all of your partner's personal activities. isaacl (talk) 23:22, 8 December 2013 (UTC)

If my partner is lobbying against a particular defense contract, but I don't happen to be, I shouldn't be editing the Wiki article on that subject. I can see such a situation arising. Coretheapple (talk)

I don't think you should be responsible for asking your business partners all about their personal lives and activities (are you lobbying for gay marriage? for health coverage initiatives? for a higher minimum wage? and so forth); it is intruding on their privacy. isaacl (talk) 23:58, 8 December 2013 (UTC)