Jump to content

Wikipedia talk:Conflict of interest/Archive 13

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 10Archive 11Archive 12Archive 13Archive 14Archive 15Archive 20

Proposal: drafts marked as user pages

Because of the potential for WP:MIRRORs to present CoI-editor drafts to the less-well-informed as though they were legitimate WP pages, I would like to require that all CoI-editor or paid-editor drafts display the {{user page}} template. -- [ UseTheCommandLine ~/talk ] # _ 17:06, 29 March 2013 (UTC)

There really is no need to get overly prescriptive. We should assume that we're dealing with reasonable people until it's shown that we aren't (which has not happened yet). All we need to do is remind people that all edits are their own responsibility and when a user with a declared COI makes a suggestion, they should check it carefully for subtle bias - just as they would an external source such as the subject's website. We do this every day, after all, it's not as if this is some radical new departure. Guy (Help!) 16:45, 30 March 2013 (UTC)
Isn't it already the case that any draft in userspace should be appropriately tagged with {userspace draft} to avoid such confusion? Ocaasi t | c 19:55, 30 March 2013 (UTC)
Is it? I don't make a habit of looking at others' userspace drafts. From WP:UP#SUB:

Your new page will come up, and you can start using it right away, or if you are using the new subpage for an article draft you may want to add the template {{userspace draft}} at the top of the page.

The difference here is that for editors with a declared CoI, I am suggesting that policy should be that this template should be mandatory rather than suggested. -- [ UseTheCommandLine ~/talk ] # _ 23:04, 30 March 2013 (UTC)
We shouldn't be encouraging large companies or governments to suggest words for articles about themselves, their products or their legislation. Fixing errors and pointing out omissions is helpful, but they shouldn't be supplying drafts of entire articles or sections. They also shouldn't be taking over talk pages and swaying the discussion in certain directions. The greater the public interest in an issue, the more important it is to keep the source/subject at arm's length. SlimVirgin (talk) 22:03, 30 March 2013 (UTC)
This should not be construed as justifying the posting of userspace drafts. editors with a declared CoI can do all sorts of useful things in userspace. Whatever they use it for, I think it should be tagged as a user page.
And while I tend to agree with you re: arm's length, that also seems like a non-starter from a political standpoint. -- [ UseTheCommandLine ~/talk ] # _ 23:10, 30 March 2013 (UTC)

BP and large company editing in general

There are some significant issues raised by the editing of User:Arturo at BP on the BP article talk page. I don't want this discussion to be about Arturo's behavior, but rather on the issues raised. Indeed it may very well be the case that he is editing according to the letter of our rules, but not in the spirit that any non-paid Wikipedia editor would recognize.

The issue - BP, through Arturo, is providing the content for the BP article, especially in regards to its environmental performance, by posting a notice on the talk page, which is followed by Wikipedia editors putting the content "as is" into the article. Notice he is not editing the article directly himself. Arturo is not exercising independent judgement during this process. He states on his talk page: "The information I present from news sources is verified by the various subject matter experts within the company. I am not an expert myself on all of the topics..." In other words, the material is cleared by higher-ups first. Probably the most worrisome problem here is that our editors don't appear to be exercising independent judgement either, just posting the material "as is."

It should be clear to everybody that having BP write up the BP environmental record for Wikipedia in a way that is not transparent to our readers is completely unacceptable for Wikipedia, and it should be completely unacceptable for BP as well. Somebody over there hasn't thought this through very well. BP has every opportunity to present its side of the story as often as they want in adverts, press conferences, etc. Putting material in Wikipedia (where the author can not be easily disclosed to the public) is going to be viewed as trying to pull the wool over people's eyes. There definitely is a conflict of interest - and not just in the way Wikipedia defines it - because of things like all the environmental lawsuits BP is involved in.

BP may also be skirting the law in the US that prohibits undisclosed advertising (this was the law long before the internet and has been confirmed in the case of using the internet - google "astroturfing" if you're not familiar with this). BP is certainly violating the codes of ethics of the main US and UK PR associations codes of ethics.

I can think of 3 ways to handle this, though there must be more:

  • Appealing to BP's common sense to just withdraw - they must know that this will come out and how bad they will look.
  • Just use WP:Ignore all rules to remove the offending content - this rule - Wikipedia's 1st rule - was designed for these cases where the right thing to do is so totally obvious that little discussion is really needed.
  • Change the rules so that no large companies (say Fortune 500 sized) can edit even on talk pages.

I'll inform various parties. Thanks for any feedback. Smallbones(smalltalk) 02:07, 19 March 2013 (UTC)

"Probably the most worrisome problem here is that our editors don't appear to be exercising independent judgement either, just posting the material "as is.""
Wow, you have seriously overstepped yourself. Please remove this personal attack on several editors now. SilverserenC 02:23, 19 March 2013 (UTC)
It's best not to try to make this personal. I certainly haven't mentioned any names. But when I describe the general situation, I'll call 'em as I see 'em. There has been a tremendous lack of independent judgement on this matter. Smallbones(smalltalk) 02:48, 19 March 2013 (UTC)
You don't need to name names, you've already clearly stated that anyone involved in Arturo's userspace drafts or other talk page suggestions is who you're referring to. So, namely, myself, Beagel, Petrarchan47, Rangoon11, Martin Hogbin, and BozMo. I might be missing a few people. Are you seriously accusing all of us of lacking "independent judgement"? SilverserenC 03:12, 19 March 2013 (UTC)
Please add my name to the list of those that have shown a tremendous lack of independent judgement. Gandydancer (talk) 17:22, 19 March 2013 (UTC)
I think another troubling aspect is, assuming my interpretation is correct, that some editors were asking the CoI editor to make changes, rather than making them themselves. This cedes editorial control to the subject being written about, and further makes the editorial process (history, etc) totally opaque. It makes things very easy for the editors doing the integrating, but at what cost? -- [UseTheCommandLine ~/talk] #_ 02:51, 19 March 2013 (UTC)
Because it's being polite to the person's userspace draft. It's generally considered rude to go in and start changing someone's draft page without solicited permission. I don't see how it makes any difference either, as the resulting changes are reviewed just the same. The same suggested sources are used, the wording is scrutinized, what exactly is the difference? SilverserenC 03:13, 19 March 2013 (UTC)
see WP:NOBAN. -- [UseTheCommandLine ~/talk] #_ 03:53, 19 March 2013 (UTC)
Yes, exactly. Your point? SilverserenC 04:03, 19 March 2013 (UTC)
"...it is usual to avoid substantially editing another's user and user talk pages other than where it is likely edits are expected and/or will be helpful." (emphasis added).
So then it seems to me you are implying that edits were not expected to Arturo's work? -- [UseTheCommandLine ~/talk] #_ 04:08, 19 March 2013 (UTC)
I'm saying that most editors feel more comfortable with suggesting changes to a userspace draft and letting the user make the changes. Regardless of expectation, most people feel like they're intruding when they go in to make changes. That's why when drafts, like in Peer reviews, are looked over, changes are suggested by others, rather than the other people just going in and making the changes. SilverserenC 04:13, 19 March 2013 (UTC)
I appreciate that maybe this has been your experience, but it has not been mine. I am wary of your repeated implications that you speak for the community as a whole. I have no quarrel with people making productive edits on my pages, have done so on others without causing drama, and have seen others do the same. WP is a collaborative space, fullstop, and if i can find the sentence i recall seeing stating that even one's userpage or sandbox material may be "mercilessly edited" i would cite that. So thats just, like, your opinion man. -- [UseTheCommandLine ~/talk] #_ 04:20, 19 March 2013 (UTC)
I certainly don't speak for the community as a whole, but I do think I can understand and represent the opinions of a fair amount of people. And I also know the editors in question and it seems more their style that it's better to look over and gauge neutrality and changes to the drafts than to make the changes yourself. Besides, isn't it a far better method to determine if someone is editing neutrally by having them make the suggested changes and then seeing if the changes were written in an appropriate manner? SilverserenC 04:57, 19 March 2013 (UTC)
Silver is right: it is normal to avoid making changes to someone else's sandbox unless (1) you have specifically been invited to do so or (2) it's an emergency. NB that "normal" is not the same thing as "absolutely required".
More pointfully, who cares? If I see a problem in someone's sandbox, and I leave a note saying "Hey, you should consider saying ____ instead in that third paragraph", and the user agrees, then who actually cares which one of us physically made the change in the draft? Really: we care about the final product, not about the process used to achieve it. WhatamIdoing (talk) 15:09, 19 March 2013 (UTC)
I completely agree with you. I was just responding to the question on why other editors, including myself, didn't edit the drafts directly. SilverserenC 22:33, 19 March 2013 (UTC)
I agree with Smallbones that this is problematic; it has the potential to make both Wikipedia and BP look bad. The BP employee, User:Arturo at BP, has said he is here on behalf of BP; he's not just someone who happens to work for the company. He has posted 10 fairly extensive drafts that are being ported over into the article without signalling to the reader that these are BP's words. I believe he has acknowledged conferring with BP's subject-matter experts to approve the text.
The best thing would be for BP to withdraw and for the material to be removed from the article. Alternatively, we can regard the drafts as unpublished primary source material: BP's view of itself, or its interpretation of the news coverage it has attracted. I've asked Arturo at BP (diff) to consider posting his articles to BP's website (that is, if BP will not agree to withdraw). That way, we could use them as published primary sources. It's not unusual to ask article subjects to post material to a website so that we can cite it.
Looking ahead, we should try to come up with words about this for the guideline. No one minds if a small business makes a suggestion for content, but a multinational can't be allowed to rewrite areas where it has been the subject of media criticism and even criminal charges. It's not only a question of checking the text for errors; there are issues of weight to decide, which sources to use, which vocabulary to choose, and what not to include. These are issues that the company really shouldn't be influencing. SlimVirgin (talk) 03:39, 19 March 2013 (UTC)
How is it going to make anyone look bad? BP following Wikipedia's rules and doing things through the talkpage the proper way, as requested by Jimbo, is somehow bad?
And you and I both know that Wikipedians get subject matter experts to look over their work all the time. Heck, there was that proposal at FAC for a while on getting a known list of experts for various subjects together that were willing to look over the article submissions for their subject. It seems extremely proper for Arturo to be conferring with the people who know more about the subject than him and getting their input.
Your WP:PRIMARY suggestion is an utter mockery of the sourcing policy. A userspace draft by an editor is not a source, regardless of how many times you claim that it's "BP's voice", which is just insulting Arturo. He is the one that wrote this material, he is a Wikipedia editor, and you should really stop treating him like a lesser person because of it. SilverserenC 03:45, 19 March 2013 (UTC)

A wrinkle here that I have not seen considered is that if he's writing this in the course of his work for BP, then that seems to me to be a work-for-hire under copyright law. In the US at least, work-for-hire copyrights are owned by the company, rather than the creator. We need some kind of explicit permission from BP for this material to not be a WP:COPYVIO. (If Arturo is an employee, this does in fact mean that this is "BP's voice" -- laws matter.) -- [UseTheCommandLine ~/talk] #_ 03:51, 19 March 2013 (UTC)

I want to add that several editors on that page have been doing their level best to keep an eye on this and offer opposition. It's unfair to them that they've had to spend their free time, on behalf of Wikipedia, monitoring a multinational that wants to write its own article. I believe Petrarchan47 eventually gave up. Gandydancer and Binksternet are still doing a great job, and there may be others (I'm not familiar with everyone at that article, and have only just started looking at the archives). Clear guidance here for the future will help editors who find themselves in that position. SlimVirgin (talk) 03:56, 19 March 2013 (UTC)
I think that this page is dedicated to the discussion of the guidelines and not the particular article which should be done at the article's talk page. Also, I would ask you to edit your comment above as right now it divides editors on the BP articles into the camps and as such, increases battleground atmosphere. Taking account the heated situation around this article, I suggest to remove the names or give credits to all editors who have done a lot of work on this article (although they may have different view points in different issues), but again, this is here not a place to discuss BP article. Beagel (talk) 05:44, 20 March 2013 (UTC)
"Opposition" to what? Are you accusing Arturo of something? All i'm seeing here is that you assume every action made by him is slanting the article, with no proof on that. He has been completely open and transparent, only using the talk page. We're supposed to be working with companies and individuals, not trying to fight them, as you think we should be. SilverserenC 04:06, 19 March 2013 (UTC)
"We're supposed to be working with companies..."[citation needed] -- [UseTheCommandLine ~/talk] #_ 04:09, 19 March 2013 (UTC)
So, basically, you're of the opinion that if a company or article subject tries to improve their article or points out that there is incorrect information, we're supposed to respond with "screw you"? SilverserenC 04:11, 19 March 2013 (UTC)
I've said nothing of the sort. I have already invited you to point out which of my statements you are interpreting in this way, which you have thus far declined to do. I even asked you to do so on my personal talk page, since you appeared to be accusing me repeatedly in the wrong forum. I don't think this is the appropriate forum either, tbqh, and even though it's not as good as my (or your) talk page it's better than Talk:BP. So have at it, if you must. -- [UseTheCommandLine ~/talk] #_ 04:14, 19 March 2013 (UTC)
If you don't support working with companies, then you oppose working with them, which means not listening to them when they have a concern. This is all pretty straightforward. And, hey, there's a fair number of editors that think we should just say screw you to companies that come requesting changes, so feel free to go and join that camp if you want. SilverserenC 04:28, 19 March 2013 (UTC)
Your tone seems a bit strident to me. That doesn't strike me as a way to identify and address the underlying issues, whatever they are. And I notice you haven't pointed out what specific statements I've made that lead you to make these assertions about my beliefs. Feel free to do so at any point, if the mood strikes you.-- [UseTheCommandLine ~/talk] #_ 04:34, 19 March 2013 (UTC)
That's the thing i'm asking, about your beliefs. What exactly is the issue with Arturo's drafts? What is wrong with them? Are they not neutral? Have you read them? Or is this all just a "they have a COI, so they can never be neutral" thing? Because that's what the whole discussion on the BP talk page seemed like. No one actually looking at the content and instead just going after the editor. SilverserenC 04:40, 19 March 2013 (UTC)

You know, as I stated initially, I just started digging into this today because of some postings on noticeboards. I haven't really had time to do a thorough accounting of the pages; what i did see was subtle issues of word choice and choice of references that seemed like something a PR person would include, and which were accepted not just uncritically, but enthusiastically without substantial revision by a group of editors, or in a few cases the revision was punted back to the COI editor. One of that group of editors actually started to become accusatory in response.

My personal views on companies and their relationship to WP (or any other domain) are complex, and would be better discussed on my talk page, or off WP entirely. -- [UseTheCommandLine ~/talk] #_ 04:55, 19 March 2013 (UTC)

Then, please, go and point out the issues. Focusing on the content would be great and getting those things changed would be even better. But I don't feel that wholesale removing the changes would be beneficial to the article at all. And I don't actually think Arturo is doing those subtle things on purpose. If you point them out to him, then he can work on making sure he doesn't do those things anymore. But if you look at the article as a whole from what it was before, you can easily see that it markedly improved, minor issues aside. Any small subtle wording issues or reference choices can be done after the fact, but if the overall effect is to improve the article, I believe that's something we should be supporting. SilverserenC 05:13, 19 March 2013 (UTC)

Proposed changes to the guideline

I didn't bring start this discussion to rehash the BP talk page. It has been asserted that everything going on at the BP article is within the rules stated here, though I'm not sure I believe that. If that is indeed the case, then we clearly need to change the rules - we clearly can't have companies writing their own histories on controversial matters and have it presented without disclosure on the article page.

Some of the problems I'd like addressed are:

  1. BP is currently involved in court cases on their environmental performance, with billions of dollars at stake.
  2. Arturo at BP is not concentrating on so much on facts, but is rather arguing about interpretation, emphasis and weight. BP has plenty of places where it can present its own interpretation, but these matters on Wikipedia should be left to independent editors
  3. The material is being cleared with others at BP, perhaps even by the legal department. This means effectively that User:Arturo at BP is a joint account, and not an individual taking individual responsibility for his edits.

Clarification of the current rules could be spread through the guideline and could solve this matter, but the most direct way would be a section on the use of talk pages - perhaps something like:

Use of Talk Pages by editors with a COI

Editors with COIs are generally invited to use article talk pages to bring verifiable facts to the attention of Wikipedia editors or to point out sources where the COI editor or his employer have presented their opinions and interpretations of the facts. Talk pages should not be used to argue for an employer's opinions or views on the proper emphasis or weight of material within the article, nor should it be used to argue about matters that are currently before the courts.

The talk page should not be used to present the official views of a corporation. Every editor, including editors who have declared a COI and who only edit on talk pages, must be an individual using his or her own judgement. Corporate and joint user accounts are strictly prohibited. Corporations usually have many outlets for presenting their interpretations and opinions, e.g. "public service advertisements", op-eds in newspapers and trade journals, press releases and press conferences. If a corporation wants to draw the attention of Wikipedia editors to their official views, they may simply post a link to these materials on the talk page with a very brief explanation of the link's relevance.

Well, that might be a start, but I'm not wedded to any particular wording. Comments appreciated. Smallbones(smalltalk) 14:28, 19 March 2013 (UTC)
  1. If a court case is influenced by what Wikipedia says, then the courts have far more serious problems than we could possibly hope to address here.
  2. If Arturo is not editing the page directly, then these matters are being left to "independent editors". Your complaint seems to be "independent editors are agreeing with Arturo instead of agreeing with me".
  3. Talking to other people, even seeking their agreement in advance, does not constitute a "joint account" as far as Wikipedia is concerned. As far as we're concerned, a joint account means different humans physically typing the words into the edit box and pushing the 'save page' button. WhatamIdoing (talk) 15:15, 19 March 2013 (UTC)
Also, you say that "Talk pages should not be used to argue for an employer's opinions or views on the proper emphasis or weight of material within the article", but I wonder why not? Consider the case of some celebrity who had a minor traffic infraction, like a speeding ticket. Imagine that the BLP is wildly unbalanced, with two out of three sentences in the article referring to the speeding ticket. Why shouldn't we welcome the participation of the celebrity's publicist in discussing and correcting that obvious problem about "the proper emphasis or weight of material within the article"? WhatamIdoing (talk) 16:01, 19 March 2013 (UTC)
1. Currently WP:COI says
Legal antagonists

If you are involved in a court case, or you are close to one of the litigants, you should not write about the case, or about a party or law firm associated with the case. Even a minor breach of neutrality in an article that is before the court could cause real-world harm.

If that is not clear enough we need to let BP know that contributing material about a court case they are involved in is not appreciated.
Courts are generally concerned about media coverage of cases before them, e.g. if it might bias a jury. If the media coverage is likely to bias a jury then they will move the case to another county. That remedy is not available however when the coverage is from Wikipedia.
2. I don't think your point 2 above has anything to do with my point 2. All I'm saying is that corporations can present their opinions off-Wiki, then we can actually quote them. "Independent editors" just passing through Arturo's materials from BP into the BP article without quoting BP, are clearly acting improperly.
3. from WP:NOSHARE "Any user account should represent an individual and not a group". From WP:ROLE "Role accounts: Because an account represents your edits as an individual, "role accounts", or accounts shared by multiple people, are as a rule forbidden and blocked." If somebody has to clear their edits through their corporate legal department or just their boss, they are not editing as individuals. People editing with COIs are allowed to edit (even though they are strong discouraged) but they have to be individuals, not corporations.
4. You give a BLP example of why corporations should be able to argue on matters of weight and interpretation - aka spin - on talk pages. Clearly WP:BLP over-rides WP:COI, but there is no WP:BLC Smallbones(smalltalk) 17:07, 19 March 2013 (UTC)
We're basically prohibiting corporate representatives or other advocates from editing directly. That seems to be a growing consensus although it's not explicitly banned, at least not in uncontroversial situations. The result of that is we have more editors coming to the talk on behalf or organizations and individuals. We're encouraging them clearly disclose their COI--in their username, their user talk page or userpage, the article talk page, and the COI noticeboard. Those editors are then seeking review about criticisms, sources, suggestions and drafts. This is what policy currently recommends.
We have a general divide over editors who believe those with a strong COI are inherently biased and should not be trusted or even worked with, and those on the other handwho think that we should judge content not the contributor. This reflects a broader philosophical debate which we're not going to solve here; it's too divided and divisive (see Wikiproject Cooperation and Wikiproject Paid advocacy watch for example). The pragmatic fix we need is to improve the review process for proposed drafts. If a corporate rep suggests a neutral, well-sourced draft, that draft must be reviewed by others. But is that sufficient to ensure neutrality? Smallbones, SlimVirgin, and the other BP watchers at the article think it's not sufficient, that it's parroting a company's views, even to the extent of being a copyright violation or a primary source situation. I think that's a stretch of WP:PRIMARY, but it reflects a broader concern about authorship. Since corporate reps are not editing directly, this is seen as direct editing 'by proxy'.
The judgement call to make is how much review and criticism can ensure that a proposed draft has been independently judged. The Wikipedia editors who move such a draft into an article, after reviewing/improving it, ultimately take responsibility for the draft, as with any edit request. So what do we need to do to make that review/request process more robust? Also, if we're not reviewing suggestions from paid advocates, are we going to drive them underground? What's the right balance to strike?Ocaasi t | c 17:16, 19 March 2013 (UTC)
I don't think the division there is quite so stark as you make it out to be, and I also think that characterizing one side (of an artificial division to begin with) as inherently distrustful is, while understandable, not particularly AGFish.
I think that you are exactly right about the need to improve the review process for drafts. Yes, I think that there is a meaningful concern about work-for-hire and copyvio, but I have also been characterized as a little woo-woo crazypants about copyvio. That's for a legal team to decide, but I think we would be remiss to discount the issue completely.
Perhaps a meaningful solution here is to try and use some kind of proportionality principle akin to WP:UNDUE, whereby the extent of a CoI editor's changes should have proportionally greater or lesser attempts to draw in other eyes to review the edits. The edits in question here are extremely significant, so for instance one might post on a number of noticeboards or make an RfC before incorporating that information directly. I could imagine this having the effect of making small, uncontroversial suggestions such as corrections of fact the least disputed kinds of changes. When we have to consider issues of cherrypicking and due weight, the number of active editors on a specific page may not be enough to fully consider the implications of wide-ranging, substantial changes that are made very easy for them. There might also be some sort of tagging system, whereby certain edits can be mentioned as coming from a CoI editor originally, in order to enable more meaningful review by others after-the-fact. Things can get lost in the history.
I realize that's not much of a solution, exactly, since it brings in even more of these ethereal sorts of considerations, but it's the best I've got. -- [UseTheCommandLine ~/talk] #_ 18:28, 19 March 2013 (UTC)
There's no copyvio problem here. If you're hired to write something and to post it on Wikipedia, then you are implicitly authorized to license it under Wikipedia's terms. You don't give up your copyright when you post to Wikipedia. You keep the copyright and freely license it. A person in a work-for-hire situation assigns the copyright to the employer and still licenses it to the world.
Put another way, if there were a lawsuit about the copyright of the comment I'm typing now, then I personally own the copyright and everyone has a license. If there were a lawsuit about the copyright of Arturo's work on Wikipedia, then BP owns the copyright and everyone still has the same license that they would have if I'd typed it. WhatamIdoing (talk) 19:24, 19 March 2013 (UTC)
I get that. I just think it's odd to allow someone to say "I don't own the copyright to this, so I have no say in the matter, but I assure you it's ok to license under these terms." We don't do that for other classes of copyrighted material unless we're claiming afair use exemption. I think we should have some sort of disclosure from the copyright holder (in this case BP) that Arturo's work is freely licensable under CC-BY-SA. -- [UseTheCommandLine ~/talk] #_ 19:29, 19 March 2013 (UTC)
Okay, Smallbones, let's try this again: You say, ""Independent editors" [scare quotes sic] just passing through Arturo's materials from BP into the BP article without quoting BP, are clearly acting improperly."
I'm telling you that no matter your disdain for the editors involved in that article, it's not at all clear that they've acted improperly. SlimVirgin practically wrote the NPOV policy. If she agrees that a statement complies with that policy, we can safely assume that she's right. There is no evidence whatsoever that our independent editors are acting improperly, or indeed that they are doing anything other than helping out in exactly the way that we regularly beg people to do. WhatamIdoing (talk) 19:20, 19 March 2013 (UTC)
WAID, the problem with this situation is this: (1) there were a very small number of editors involved in agreeing to this; (2) most of the editors on the talk page who agreed to the drafts were probably not in a position to judge whether the drafts were neutral, in the sense of accurately representing the body of literature on each of the issues BP rewrote - informing yourself to the point where you're able to judge that is a lot of work; (3) it's clear that some of the editors felt railroaded; and some were supporting only because they support paid advocacy by corporations; and (4) – and this is the most important point – the tens of thousands of readers who read that article between July 2012 and February 2013 were not told they were reading BP's words.
I'm not familiar with astroturfing and its parameters, but does this not count as a form of it? SlimVirgin (talk) 19:40, 19 March 2013 (UTC)
I'm sure you could find a blogger willing to say so, but normally astroturfing requires actual attempts at secrecy. If you're putting "Hi, I work for this company" on every single available edit, even if someone (i.e., the Mediawiki software, which is outside your control) makes those notices somewhat hard to find, then that doesn't really count as astroturfing. It's certainly less like astroturfing than filing a complaint with ORTS, which is another legal and community-sanctioned way to deal with bias and errors. WhatamIdoing (talk) 03:50, 24 March 2013 (UTC)
I think a principle of "adequate review" is a useful way of thinking about this. One of the criticisms here is that Arturo's edits might not have been adequately vetted. So how about, the more extensive the changes, the bigger the net needs to be cast as far as notifications (e.g. wikiprojects, noticeboards, etc) in order to invite active reviewers to vet the material and make sure that it is not, for example, WP:POVPUSH. The greater the amount of material, the more places there is for this sort of thing to hide, so we need to be correspondingly more vigilant in reviewing it. Uncontroversial edits, such as uncomplicated corrections of statements of fact, would be treated the same as they currently are, but Arturo's edits would invite a much wider range of editors to review the proposed content and kick the tires, so to speak. -- [UseTheCommandLine ~/talk] #_ 00:18, 20 March 2013 (UTC)
During the period from July 2012 up to today, at least five active editors have been around that article all the time. Being around (also at the talk page) they should be aware about these proposals and have had access to drafts to comment and make changes. If they have not done this, it may also mean that they have seen no problems. But in general, I agree that we probably need more precise rules how to disclosure and review proposals made by COI (particularly paid) editors, so maybe a list of relevant forums where it should be notified would be useful. I also think that that kind of drafts should not be placed at the user pages but special drafting subpage should be under the article's talk page to make clear that every editor is invited to make changes and comment. Beagel (talk) 06:24, 20 March 2013 (UTC)

BP's rewrite of the article about itself

To be clear about the extent of this, I've compiled links to BP's drafts, listed how long they were, which ones were posted, and links to any discussion; see BP's drafts.

It appears that between July 2012 and February 2013 BP rewrote around 44 percent of the (currently) 9,215-word article. SlimVirgin (talk) 19:32, 19 March 2013 (UTC)

Once again, BP did not wrote the article. In general, these are not new drafts but based on the existing sections/part of the articles. Saying that BP (that means Arturo) is a sole author of these texts is incorrect. Also, it was discussed several times that all these draft where disclosed and open for editing/discussion by other editors. If they used or not this opportunity, is up to any individual editor but that time at least five editors were permanently active on that article and thus were aware of these drafts. No draft was used for posting which was objected by other editors (e.g. in that happened in December as there was no consensus among editors). It was also said above (and also during discussion at the BP's talk page), if any editor put the draft to the article, s/he bear responsibility about that text. I would kindly ask you to go through these posts made by me and to say what exactly is not n line with the WP policies. Otherwise, I kindly ask you to stop accusing editors who in god faith have spent a lot of time improving that article. I would line also say that accounting words may give a wrong impression as since edit conflict which statrted in spring 2012 between editors Rangoon11 and Petrarchan47, which was discussed in different venues and which bring a lot of editors to this article) the article has gone through a serious changes and a lot of improvements. Beagel (talk) 06:11, 20 March 2013 (UTC)
(ec)

People seem to be missing the fact that putting BP's draft into the article is a form of quotation or paraphrase that must be cited according to our rules. Thus I present the following revision. Others are invited to suggest their own revisions.

Use of Talk Pages by editors with a COI (revision 1)

Editors with COIs are generally invited to use article talk pages to bring verifiable facts to the attention of Wikipedia editors or to point out sources where the COI editor or his employer have presented their opinions and interpretations of the facts. Talk pages should not be used to argue for an employer's opinions or views on the proper emphasis or weight of material within the article, nor should it be used to argue about matters that are currently before the courts.

The talk page should not be used to present the official views of a corporation. Every editor, including editors who have declared a COI and who only edit on talk pages, must be an individual using his or her own judgement. Corporate and joint user accounts are strictly prohibited. Corporations usually have many outlets for presenting their interpretations and opinions, e.g. public service advertisements, op-eds in newspapers and trade journals, press releases and press conferences, or just on the company website. If a corporation wants to draw the attention of Wikipedia editors to their official views, they may simply post a link to these materials on the talk page with a very brief explanation of the link's relevance.

If an editor with a declared COI offers a draft of a section or of a complete article on the talk page, Wikipedia can not include that draft into an article without a complete rewriting. We cannot assume that the proposer of the draft is actually the representative of his employer. We cannot include the views of an interested party in the article without citation and we cannot cite the quote or paraphrase if it hasn't been published. Thus a corporation must publish its views - if only on its corporate website - if it wishes to make its views known to Wikipedians.

I think that should make it clear that uncited quotes and paraphrases are not allowed. Smallbones(smalltalk) 19:43, 19 March 2013 (UTC)
Could it be summarized like this?

Company employees, particularly those working in corporate communications, may set up user accounts on behalf of their employer, and proceed to write drafts for insertion by others into the article about that company, rather than engaging in direct editing themselves. These drafts should be regarded as unpublished primary sources. The company employee should be asked instead to submit his material to a publisher, so that Wikipedia can cite it as a source. The company's website may be an acceptable publisher; this would allow the material to be used as a published primary source, with the usual caution.

This is in line with the advice Jimbo gave Microsoft when they were caught paying someone to rewrite bits of their article. He suggested Microsoft publish that material somewhere independently, so that we could cite it as a source like any other; see "Microsoft Violates Wikipedia's Sacred Rule". SlimVirgin (talk) 20:42, 19 March 2013 (UTC)
How do we verify that the editor represents the company that he claims to represent? It is not a trivial task, and spoofing (e.g. by competitors or just people who want to make them look bad, could be a problem. Publishing on their own website avoids these problems completely. Smallbones(smalltalk) 22:09, 19 March 2013 (UTC)
Right, it solves everything. It means we know who the author is, we can cite the source, quote it, and the reader can see where the perspective has come from. It also means we're not giving one source privileged access to the article over any other. SlimVirgin (talk) 22:44, 19 March 2013 (UTC)
(to sv) I think it's a slippery slope and slightly circular logic to characterize any editor's contribution as a primary source. Wikipedia isn't a source in relation to Wikipedia, to me that's just incoherent logic. Unsourced and biased potentially, but I wouldn't call it a primary source. Gigs (talk) 20:50, 19 March 2013 (UTC)
It's material about the company written by the company; that makes it a primary source. It has been posted on Wikipedia, rather than on their own website; that makes it an unpublished primary source. To resolve this, they need to convert it to a published source so that we can use it. The point here is that BP is a source of information on BP. It is not an editor, not directly and not by proxy. We can't choose one of our sources out of the dozens we use, and give it privileged access by saying "you don't need to publish and be cited; instead, we will add your unpublished perspective directly to the article." SlimVirgin (talk) 21:07, 19 March 2013 (UTC)
No, it doesn't. First, if it's published on Wikipedia, then it's published, which is the opposite of unpublished. Second, writing about your employer makes it "affiliated" or "non-independent", but it could still be secondary. WP:Secondary does not mean independent. If an employee read a bunch of news reports about the employer and summarized them, then the result is secondary, not primary. WhatamIdoing (talk) 03:54, 24 March 2013 (UTC)
(e/c in reply to smallbones) I don't like it. Why shouldn't they present the official views of the corporation here? They are pretty much bound by their job to do that. That's like asking the scorpion not to be a scorpion. So we'll just wind up with scorpions in sheep's clothing. Not that sheep ever wear clothes. What were we talking about again?
On the complete rewrite, that's also no good, since some COI editors produce decent work, and a lot of editors don't blindly approve their drafts anyway. Take a look at AfC sometime. Are we going to require AfC to reject or completely rewrite every article that clearly came from someone with a COI, i.e. nearly all of them?.
So I'd say this is an idealistic proposal. We need something a little more concrete and workable. Gigs (talk) 20:45, 19 March 2013 (UTC)
We can't actually assume that anybody here is the official representative of the firm (see above). If we want to use material that is claimed to be from the firm, then we have to cite it, which means it has to be published. Note that I'm not saying that anybody with a COI can't write anything, just that people who claim to represent the firm (a big difference) have to publish it off wiki. Smallbones(smalltalk) 22:17, 19 March 2013 (UTC)
Hello, I have replied on the BP Talk page but also wanted to leave a note here to clarify some information. The drafts that SlimVirgin mentions above were added into the article by volunteer editors after they had been reviewed. The majority of the drafts that I proposed focused on the company's operations and provided new information. These were not "rewrites" but drafts providing entirely new information on the company's activities that had not previously been included in the article. In mid-2012, there was virtually no information on the organization of the company or its various activities. The drafts I provided for the UK, US and worldwide operations addressed this lack of information. These drafts provided factual information, not the company's views.
As a side note: I believe Smallbones may have misinterpreted a comment I made about subject matter experts, what I was saying is that I rely on available sources (primarily third party sources such as news articles) to write the drafts but that I check with subject matter experts within the business that the facts in my drafts are correct (for instance, production figures or the chronology of events). This is to ensure that the information I present on Wikipedia is accurate. I am the author of the drafted material and by posting it to Wikipedia discussion spaces, I am releasing the material under Creative Commons just as any other contributor to Wikipedia does.
I believe that SlimVirgin's estimate of how much of the current article is based on material I proposed is incorrect: some material was edited, shortened or later replaced. Indeed, I should note that the first draft that I presented was an overview of all the operations, which was largely replaced with the more detailed information about the operations by location. Not all my proposed drafts were added, and one was significantly reduced following feedback from other editors.
As I have said on the BP Talk page, I would be open to a review of the material in the BP article to ensure that it is neutral. I would like to remind editors that edits have been made by other editors to this material since it was added to the article, so please bear this in mind when reviewing. Arturo at BP (talk) 20:52, 19 March 2013 (UTC)

We should look at the context here. While this addition of material was going on (Nov 15, 2012) BP admitted to and settled charges with the Department of Justice of "obstruction of justice" - lying to Congress - effectively perjury, as well as misleading its investors (only a $525 million fine on that one). "The company initially tried to cover up the severity of the spill, misleading both Congress and investors about how quickly oil was leaking from the runaway well, according to the settlement and related charges."BP Will Plead Guilty and Pay Over $4 Billion, NYTimes Passing off BP statements as fact in the article about BP, without labeling them as being from BP is just nuts. I do not blame Arturo - presumably he is just doing what his bosses tell him to do and his livelihood depends on that. I blame a poorly written guideline that some editors interpret as saying that this is ok. It is not ok. It is against all of our fundamental rules: V, NPOV, NOR, you name it and a straightforward reading of our rules prohibits it. It is a violation of our readers' trust. Any proposal (or non-proposal) that will let anybody believe that anything like this is ok is simply a cop-out. We have to very clearly say "this is not ok" or no reader should ever trust anything in Wikipedia ever again. As a practical matter, we have to say this is against the rules in the clearest possible language. Smallbones(smalltalk) 22:02, 19 March 2013 (UTC)

See, this is the problem. You are exactly violating "comment on the content, not on the contributor". What is wrong with the content? Is the content biased? How is it biased? Give exact examples. You have presented no evidence whatsoever that the material in Arturo's drafts aren't neutral. All we care about is our content. Is the content as presented neutral or not. The reason we focus on the content is that, practically all of the time, we don't know the background of the contributor. While Arturo is working above level and openly in his affiliation with BP, he didn't have to. He could have just been a random editor that presented these drafts for inclusion. That is why we focus on the content. If the content he is presenting isn't neutral, then that is indeed an issue. But without the evidence of it not being neutral, you have no argument, since the article is no different than it would be if anyone else wrote it neutrally, because that's what neutral is meant to be. SilverserenC 22:42, 19 March 2013 (UTC)
Frankly I'm tired of the accusation that I'm commenting on the contributor rather than on the content. Specifics on the content should be discussed at the BP talk page, but I'll give an overview of the general problem here (at the risk of repeating myself). When the group of editors takes on the responsibility of inserting content from a COI editor they have taken on the responsibility of insuring that the content of the article is NPOV. That group has failed spectacularly in doing this. The major problem is that BP, a party to multi-billion dollar lawsuits on the matter being written about, has had its words used directly in the article without quotation or even a reliable source and has not been identified in any way as the source of the wording. There is the matter of Arturo writing approx "BP has a mixed environmental record" - not even close. Thankfully, I don't think this had a chance to be added to the article before the s hit the fan. There are little things in the article that are clearly not NPOV, e.g. BP pleaded guilty to "Obstruction of Congress" - that is of lying to Congress and though "Obstruction of Justice" is mentioned twice, there is no explanation of what it means and no mention that BP itself pleaded guilty. Similarly BP's guilty plea to lying to its investors and the $525 million fine it paid for this is not mentioned at all. Thus, from the NPOV standpoint, the content of the article is simply terrible. Smallbones(smalltalk) 03:46, 20 March 2013 (UTC)
You're conflating a discussion about the article itself with a discussion about guidelines. Guidelines have to come from somewhere. -- [UseTheCommandLine ~/talk] #_ 00:06, 20 March 2013 (UTC)
Except this is just a proposal for rule creep to rectify an issue that has yet to actually be expressed on why it's an issue. Like I said, it's about the content and there has yet to be specific reasons expressed on why this content is any different than what any other editor would write. I don't think starting off prejudiced against an editor is the best way to go. Especially since we already have these editors following the post on talk pages only guideline, even though that isn't an actual rule. They've already gone out of their way to make extra restrictions on their editing that isn't even required of them. And now this discussion is trying to restrict them from even being able to do the minimum they put themselves out for. On no basis whatsoever. SilverserenC 00:54, 20 March 2013 (UTC)
all i can conclude from that is you're either not reading carefully, or being deliberately obtuse. -- [UseTheCommandLine ~/talk] #_ 00:57, 20 March 2013 (UTC)
I read it perfectly fine. There is absolutely no reason to not allow editors to work on draft versions to be submitted for independent review before implementation. COI is utterly irrelevant, as the neutrality of the content is the only important part. That's why our COI guideline says time and again that COI editors are still perfectly allowed to edit, so long as they do so neutrally, the same as anyone else.
And i'm not even going to address the userspace drafts are primary sources thing, because that's just utter nonsense and a perversion of our sourcing guidelines. SilverserenC 01:21, 20 March 2013 (UTC)
You sound pretty hostile to the idea of coming to any sort of consensus then. -- [UseTheCommandLine ~/talk] #_ 01:25, 20 March 2013 (UTC)
A consensus on how to make different classes of editors? Yeah, I guess I am hostile to that. I don't believe in telling someone that they're not as equal as someone else, that their edits aren't as worthwhile, when they haven't done anything wrong. I don't believe in segregation. SilverserenC 04:01, 20 March 2013 (UTC)

Proposal

I'll try again - certainly my proposal can be made simpler

Wide community input would be wonderful. I've ask to get this listed by the Signpost in their discussions section. I'm not sure that this page has any lower standing than an RfC, but if you think listing it as an RfC will help, please do it. Smallbones(smalltalk) 16:57, 20 March 2013 (UTC)
Use of Talk Pages by editors with a COI (revision 2)

Editors with COIs are generally invited to use article talk pages

  1. to bring verifiable facts to the attention of Wikipedia editors
  2. to present evidence that a purported fact in the article is false, or
  3. to point out published sources where the COI editor or his employer have presented their opinions and interpretations of the facts.

Talk pages should not be used by Editors with COIs

  1. to argue about matters that are currently before the courts
  2. to argue on the proper emphasis or weight of material within the article
  3. to present the employer's opinions or interpretations of the facts, other than giving a link to a published source and briefly explaining its relevance
  4. to present a proposed draft of the article or sections of the article, other than pointing out material that the employer or others have published.

It's simple and straightforward and should solve this problem.

  • This looks good, let's do this. (I added "by Editors with COIs" to the heading for the second section, I hope this is OK.) Herostratus (talk) 04:51, 20 March 2013 (UTC)
  • Support Yes sounds reasonable. Our readers expect our content to be written by people independent of the subject matter at hand. If it isn't this will hurt our reputation and our brand. Doc James (talk · contribs · email) (if I write on your page reply on mine) 07:37, 20 March 2013 (UTC)
  • This is better, but the last point (currently #4) is a sharp divergence from current practice. Before adopting such a policy, we need to invite much wider comment, since our policies are supposed to be primarily descriptive of practice, not prescriptive. I wouldn't be opposed to such a change in practice if it is the best solution that we can come up with. Gigs (talk) 12:42, 20 March 2013 (UTC)
  • Agree on 1 & 2, Oppose 3 & 4 However, the way this is worded, "arguing" is very vague. Does that mean pointing out that something is incorrect in such a section or to present reliable sources with no added commentary is "arguing"? And I completely disagree with 3 and 4. COI editors are perfectly allowed to suggest changes to the article in general, whether that is just with sources, rewording, or section re-writes. Furthermore, you seem to be confusing COI editors with paid editors with your use of "employer". This statement makes absolutely no sense in the context of COI. Additionally, this would also apply to POV pushers trying to negatively slant an article (and thus have a COI) as well, correct? SilverserenC 15:35, 20 March 2013 (UTC)
  • I generally support this proposal. I do wonder also whether we have any duty to notify legal authorities when there are attempts by CoI editors to change material that is the subject of a legal dispute? The possibility for this to affect the progress of a court case is small, but not insignificant. Not saying this particular case (BP/Arturo) qualifies, just to be clear. -- [UseTheCommandLine ~/talk] #_ 15:51, 20 March 2013 (UTC)
I don't think we have any legal responsibility to report violations or enforce the (multiple countries') laws ourselves, but we do have moral responsibility to make sure that our rules are generally consistent with the law, so people can be fairly sure that if they follow our rules in good faith they are not obviously breaking the law. Smallbones(smalltalk) 16:57, 20 March 2013 (UTC)
I believe that we should strongly support Wikipedians however who wish to do this sort of work. Let say we have a major device manufacturer that was trying to "improve" the Wikipedia coverage of their product to improve the chance that medicare would continue paying a billion plus for the procedure in the US when the best available evidence does not show benefit? While the procedure itself is fairly neutral in effect, the fact that this billion is not spent on other measures could result in overall harm. Oh wait this happened already a couple of month ago. Doc James (talk · contribs · email) (if I write on your page reply on mine) 01:51, 21 March 2013 (UTC)
  • Support. This would work; points 3 and 4 are particularly important. Volunteer editors are being worn down by the extended arguments from PR people, and the repeated requests to add their drafts directly to the article. The PR employees are being paid to do this, so they can keep going as long as they need to, which means editors get burned out and stop arguing, or leave Wikipedia entirely. Instead, people representing companies or other groups should refer editors to sources, which can include source material published by that company. Then we can cite those sources as we do any other. SlimVirgin (talk) 17:14, 20 March 2013 (UTC)
"Volunteer editors are being worn down by the extended arguments from PR people". Where? IRWolfie- (talk) 21:18, 20 March 2013 (UTC)
  • For the most part this is aimed at PR people working for large companies. I don't know whether Smallbones wants to keep it more general. The proposal does talk about employers, so perhaps that just needs to be made clearer.

    Actually thinking about it, it would apply to everyone. We don't want (e.g.) individual Scientologists explaining issues at length in drafts in userspace or on talk pages; we want all editors to provide published sources that we can cite. So yes, although this has been prompted by PR intervention, really it's applicable to everyone with a COI. SlimVirgin (talk) 18:58, 20 March 2013 (UTC)

  • I couldn't care less about scientologists and other similar mass movements/organisations. What concerns me is possibility of this rule suddenly popping up then subject of BLP (or relative of one) tries to argue about "emphasis or weight of material" on biography's talk page.--Staberinde (talk) 19:12, 20 March 2013 (UTC)
  • We could make clear that option 2 doesn't apply to BLPs; BLPs are exceptions to most of these issues, in the sense that, particularly for borderline BLPs, most editors are willing to accommodate any reasonable input. SlimVirgin (talk) 19:21, 20 March 2013 (UTC)
  • Oppose This whole proposal is founded on a non-issue. That a worker at BP proposed changes to a talk page, which were reviewed by independent editors who used their judgement, and then incorporated into the article largely without issue, is indicative that the material was well written in the first place. Who cares how much of his edits got in, if they were good edits, that's good. We don't need arbitrary policy about something that isn't an actual issue. IRWolfie- (talk) 21:14, 20 March 2013 (UTC)
  • That misses the point about the work involved in judging whether 4,000 words have included all the key facts, focused only on the key sources, and given due weight to everything. That's a huge amount of work. You're arguing that the British Chiropractic Association should be allowed to rewrite the article on chiropractic, so long as they do it via drafts on a user page that are reviewed. But what if the reviewers are largely sympathetic to chiropracty, or aren't knowledgeable, or get driven away because it's such a time sink? The PR reps are being paid, so they can continue it for as long as they need to. SlimVirgin (talk) 21:44, 20 March 2013 (UTC)
  • Are you saying that we can't review things properly? You already know that policy says that when implementing such material the due weight of it and all the rest, from the reviewing, is with those that did the review. If you have a problem with the material and think myself or other editors reviewed it wrong, then please point out specific things that are wrong. Passive-aggressively saying that we can't review properly without giving any proof is not going anywhere. SilverserenC 23:37, 20 March 2013 (UTC)
Then whoever disagree with the edits reverts, and then BRD begins. PR reps can continue as long as they want on a talk page, but if noone is listening then it's not an issue, IRWolfie- (talk) 18:21, 21 March 2013 (UTC)
  • Oppose - editors with a COI have the same rights and responsibilities as any other editor. They can edit articles, and they can use talk pages. And, as with any other editor, they need to follow the same rules of verifiability, reliable sources, and neutrality as everyone else. Indeed, we can and should treat them per WP:Expert editors. That is to say, we have no means to verify that they are who they say they are, and if they wish to add to an article, they should have no difficulty finding reliable sources to support their proposed changes. Rklawton (talk) 22:12, 20 March 2013 (UTC)
  • Comment Consider the impact this would have: We're trying to encourage COI disclosure, right? But under this guideline, as soon as someone does that they can't propose a draft. No disclosure, draft proposal ok. Yes disclosure, draft proposal not ok. Which is going to encourage more disclosure? I think people need to think about the incentives we create when we make rules like this. As in any body, you don't just get what the rule says, you get all the spillover effects of people avoiding the rules they don't like. I've outlined my preferred alternative below, which is to require proportionate scrutiny under robust review. If the community decides that the above, however, is the best way forward, I'll adjust my guidance accordingly. Ocaasi t | c 23:28, 20 March 2013 (UTC)
    • Well spotted! Regardless of the merits of the proposal, this is an absolutely fundamental problem in practice - it creates strong perverse incentives against disclosure. Andrew Gray (talk) 15:12, 21 March 2013 (UTC)
  • Oppose The whole idea of bringing it to the talk page is so that non-COI editors can review and approve changes, making revisions where they feel it is necessary.--The Devil's Advocate tlk. cntrb. 00:34, 21 March 2013 (UTC)
  • Oppose per The Devil's Advocate. We've long (always?) told people with COIs that talk page proposals (including drafts) are the way to go. Why would this possibly be a bad idea? I'm not convinced by your reasoning; the only situations in which I can see it being bad are those in which the COI editors are being disruptive, e.g. posting copyvios, arguing, or attacking other editors. We already prohibit arguing (in the sense of WP:HORSE or WP:TE), because arguing isn't helpful for building consensus; we should continue to tell COI editors that they're not welcome to argue but that they're welcome to present their drafts in a calm and collegial manner. Nyttend (talk) 02:35, 21 March 2013 (UTC)
  • Strong Oppose - You don't tell people who are motivated by money that not only can't they edit the page directly, they may not even suggest positive changes. Not only does this proposal weaken our goal to provide a balanced and neutral point of view to the public, it would just force paid writers to implement the changes themselves (which they already do under the current system). Jimmy Wales said he would personally address any concerns by companies just so they wouldn't edit their own articles. Why would you not allow willing volunteers to help bring an article to a more neutral state? *Hurls waste of editor's valuable time argument out the window.* *Hurls waste.* Marcus Qwertyus (talk) 04:22, 21 March 2013 (UTC)
Nobody is saying that they cannot makes comments or suggestions, only that they publish them (e.g. on the company website) so that we can quote them, rather than include their material without disclosure of who wrote it. Smallbones(smalltalk) 04:37, 21 March 2013 (UTC)
I agree that if a paid editor is making suggestions, that the drafts should be sourced from the originating company's websites, but this raises potential issues when, for instance, that website is taken down, or changed, and it's not indexed by the wayback machine. I think a necessary adjunct to this is a tag, akin to the tags that get placed when a section is blanked, that can be used when material is incorporated from primary or CoI editor sources. -- [ UseTheCommandLine ~/talk ] # _ 05:07, 21 March 2013 (UTC)
But the "proper emphasis or weight of material within the article" may not be brought up? If Encyclopædia Britannica, Inc. wants to argue that their article unduly presents a pro-Wikipedia bias and gives rock-solid sources to back their view, I and many others will be happy to hear them out. If I approve that change, then I endorse that edit fully and take complete accountability for that change. Also problematic, writing material with the sole intention of getting it included on Wikipedia, though not explicitly forbidden as far as I can tell, is frowned upon. See also: WP:COS. Marcus Qwertyus (talk) 07:01, 21 March 2013 (UTC)
Again, the company can argue whatever it wants on its website or in any of the multitude of other forms of communication available to it. Except its contributions to the talk page, should be about facts, not about spin. If it has an opinion it needs to be presented in a form that we can quote, or it can't go into the article. Anything else on the talk page will simply be soapboxing, which is already prohibited. Smallbones(smalltalk) 13:27, 21 March 2013 (UTC)
  • Oppose Asking editors to refrain from directly editing articles within their COI is a reasonable (if often ignored) request. Talk pages, however, should be equally open to all, subject only to WP:TPG. Kilopi (talk) 13:23, 21 March 2013 (UTC)
  • Support the first portion (editors are invited to...). Oppose all the rest. TheOriginalSoni (talk) 19:26, 21 March 2013 (UTC)
  • Comment, I would support this if the proposer would strike #s 2, 3&4 of the not to do section. COI editors should be able tp propose changes, should be able to voice their opinion on content and why certain stuff should have more or less weight if they can do it civilly. If this proposal does anything to stop COI editors from positively and civilly make their opinions heard, then it only works against the goal of improving content on articles. Not that their views have any more weight regarding content, but they should not be stopped from voicing their concerns.--RightCowLeftCoast (talk) 20:02, 21 March 2013 (UTC)
  • No. Inappropriate. SilkTork ✔Tea time 19:08, 22 March 2013 (UTC)
  • Oppose. Very inappropriate.--Amadscientist (talk) 10:32, 23 March 2013 (UTC)
  • Oppose Here's what this says:
    1. Editors with COIs can't talk about "matters that are currently before the courts", even to the extent of saying things like "yes, I'm getting divorced [a "matter currently before the court"], but no, I'm not being charged with child molestation, no matter what my soon-to-be-ex is saying to the gossip rags"—but any uninformed drive-by editor can talk about this all day long.
    2. Editors with COIs can't talk about "the proper emphasis or weight of material within the article", even to the extent of saying "Look, it was just a plain old speeding ticket. How come that gets six screenfuls and a complete transcript of the discussion, when my Pulitzer Prize gets only two sentences?"—but any fanboy is welcome to argue about whether three sentences is sufficient to describe an actor's cameo appearance in his favorite video game, or if it really ought to be four.
    3. Editors with COIs can't talk about "the employer's opinions or interpretations of the facts", not even explaining the company's position—but any uninformed editor can speculate about this all day long.
    4. Editors with COIs can't talk about "present a proposed draft of the article or sections of the article", even to the extent of saying "You have a grammar error here, that would be easily fixed by changing the sentence to read ____" or "This paragraph is confusing, and it ought to say something like ____"—but any ignorant POV pusher can post drafts on the talk page all day long.
  • Fundamentally, the problem is that policy writing is hard, because you have to stop thinking about the situation you want to solve, and start thinking about all the other situations. Your efforts to stop one play-by-the-rules corporation from communicating with our regular editors would seriously screw up our efforts to keep WP:BLPs from being distorted. So my recommendation is that you stop trying to solve only your problem, and rethink this in terms of finding a solution that both stops major abuses by corporate publicity departments and protects living people who are trying to get garbage and libel cleaned out of their biographies. WhatamIdoing (talk) 04:12, 24 March 2013 (UTC)
  • Oppose as strongly as possible.
1 All this will do s encourage people with COI to go underground, and not declare it. The people who will be harmed are the honest editors who do declare it.
2 I am not aware of any serious misuse of this. Other editors have sometimes rejected such proposals, which is why they go on the talk pages. I certainly have rejected some, but this is not the same as the sort of misuse that would justify such a restriction.
3 The abuse of talk pages I have seen from people with COI is not by paid editors, nut by the advocates of causes. Paifd editors who want to keep being paid will learn to do things in the proper way required here. Zealots will go on regardless.
4 Blatant promotionalism is already prohibited anywhere on WP, from paid and non-paid editors alike. Perhaps this needs some stronger enforcement. I would enforce this only with respect to those who continue to add it after warnings, not against newcomers who may not realize, and I would not enforce this if the edit was a good-faith attempt to improve articles.
5. what route is left? OTRS? OTRS is already overloaded, and normally refers editors who want to make corrections or addition to put them on the talk pages. This would put the burden on the OTRTS volunteers themselves to make the edits.
6 Just as our licensing permits people to make commercial use of anything in WP, and this is the most basic of all our principles; similarly anyone can edit is an almost equally basic principle. I would not compromise it.
7 With respect to editors acting on behalf of companies and institutions, this would provide free range for those who wish to insert negative material. FThis is unfair and destructive on NPOV. I could expand on this, but Whatamidoing,above, has explained sufficiently.
8 Some people at WP have an anti-capitalist POV, or a anti-large corporation view, or an anarchist view of various sorts. I am not going to say what my own political view exactly is, though in some sense I have considerable sympathy for some but not all of of these positions. that shouldn't matter. We have no business favoring any political or economy POV or any political or economic theory. If we don't have NPOV, we're worthless as an encyclopedia.
  • 9 I see from some of the comments above that some of this is motivated by the failure of to have a person's own particular POV on an issue adopted, and the proposals by paid editors accepted instead as being NPOV, or more NPOV. This is an attempt to try to change the rules to win a few particular arguments. DGG ( talk ) 20:27, 24 March 2013 (UTC)
  • Oppose. It's one thing to limit actual article editing over COI issues, but stifling discussion is antithetical to basic principles here. Hullaballoo Wolfowitz (talk) 21:48, 24 March 2013 (UTC)
  • Oppose - there isn't a problem to solve, and articles should reflect all viewpoints, including those of subjects. We shouldn't be using articles to slander or disparage people/companies/whoever, and we should take deliberate steps to prevent that from happening, not try to silence people who don't want us slandering them. WilyD 10:39, 26 March 2013 (UTC)
  • Support in principle, but it would be so impractical – the PR people will find other ways to get their points in. If a policy is brought in, they will merely go underground (who knows how many such PR people are already clandestinely functioning on Wikipedia?). I agree with the policy, but unfortunately it will inevitably lead to even more burnout for consciencious editors. The Arturo debate has merely brought into the spotlight something we have all known goes on all the time. We should treat people like Arturo as any other editor – if they propose or add clearly POV stuff, we ignore and/or remove it as biased. BigSteve (talk) 12:18, 26 March 2013 (UTC)
  • Oppose Although it is clear that Talk pages are not intended as a forum for discussing non-article material or opinions, it is a forum for communicating what one feels should be included. It is easy for editors to judge and say that an editor with a COI should merely add the content elsewhere, but not everyone knows how to do such things, and, if the editor cannot make their voice heard on the Talk page, I don't know where else we can provide an opportunity for them to express frustrations or concerns over an article. Allowing them an opportunity to voice concerns in a centralized location is of paramount importance, and using the Talk page is a mature way to do so when a potential COI is present. Jackson Peebles (talk) 16:51, 27 March 2013 (UTC)
  • Oppose One of the fundamentals of ethics is that people who are affected by a decision should have a chance to comment. It is only fair that the subjects of an article have a chance to make suggestions or criticisms. If a COI editor goes too far, to the point of disrupting conversation, they can be asked to pull back, and if they don't they can be blocked. There is no reason to make a blanket prohibition because somebody might abuse editing privileges. What DGG said above is very good advice. Jehochman Talk 01:47, 1 April 2013 (UTC)
  • Support I agree with #1, #2, and #3. Talk pages should not be used by Editors with COIs:
  1. to argue about matters that are currently before the courts This is a no-brainer
  2. to argue on the proper emphasis or weight of material within the article of course another no-brainer--is any corp rep going to want to accent the negative and downplay the positive?
  3. to present the employer's opinions or interpretations of the facts, other than giving a link to a published source and briefly explaining its relevance well-said--a brief explanation is all that's needed
  4. to present a proposed draft of the article or sections of the article, other than pointing out material that the employer or others have published. Please do not allow this. This is the exact problem that we are having at the BP and other articles. In the first place, it gives corporation "supporters" an unfair advantage when the rep is writing "their" version, and secondly editors that have contributed nothing to the article can arrive, as happened at the BP article, and insert the reps draft into the article. Gandydancer (talk) 22:07, 1 April 2013 (UTC)
  • Support These are common-sense strictures that don't go far enough. There also needs to be disclosure within the article when article subjects are involved in the editorial process (drafting article texts, expressing opinions that form the basis of consensus decisions), apart from correcting factual errors and removing libelous stuff. Obviously, you can't and shouldn't prevent people from correcting libel about them. But that little crack in the door has been abused in a manner that has already given Wikipedia a black eye and will continue to do so. I think that there is a cultural gulf here between Wikipedia and the PR industry, which the latter is cynically (but always courteously!) exploiting. Word has spread that if you identify yourself, you can materialize on a talk page and soon have Wikipedia editors eating out of your hand. Only utter fools would engage in astroturfing when the door is wide open for abuse. Apparently articles on big corporations don't have the appeal to Wiki editors as articles about celebrities and video games, so an openly conflicted editor stands a mighty good chance of getting his way, all in the name of communal good spirit which is the hallmark of Wikipedia. In fact, editors who object find themselves accused of all kinds of wrongs. This is a bizarro attitude only found in Wikipedia, and one that makes this website a profoundly strange place when it comes to policing articles about corporations. The potential also exists for government agencies to walk through this same open door. I haven't been at Wikipedia long, so please forgive me if I don't cite "policy" and other bureaucratic tropes but simply am addressing the reality of the situation. Coretheapple (talk) 22:31, 1 April 2013 (UTC)

Extended discussion break

The main rule of the talk pages is that they are not forums and should be used only for a discussion how to improve the article. Therefore, the line should be drawn if the the contribution is about improving the article (concrete proposals, drawing attention to the incorrect information in the article etc.) or about promotion or whitewashing the company. The latest should be removed immediately per WP:TALK and WP:PROMOTION/WP:SOAP (the same applies also COI editors with a negative agenda, user:Johnadonovan being probably as the most notable that kind of COI editor]]). Therefore I support the third point (to present the employer's opinions ...) and propose to refer to WP:TALK and WP:PROMOTION/WP:SOAP. I would also support the first point if arguing in this context means just arguing and not a prohibition to propose correction of information. I fully agree that issues before the court are very sensitive (particularly but not only in respect of WP:NPOV, WP:DUE and WP:NOT#NEWS). However, I don't see how forbidding an editor to participate in the discussion would increase the quality of Wikipedia. Concerning discussion of the weight or presenting proposals about wording (every wording proposal notwithstanding its length qualifies as a draft) I strongly oppose the proposal. The problem is not that COI editor propose something but how to ensure that these proposals will get a proper review/examination and they a in line with the spirit and policies of Wikipedia, including but not limited with WP:NPOV, WP:V, WP:RS, WP:DUE, WP:PROMOTION/WP:SOAP. Therefore, more strict rules concerning disclosure of COI, disclosure of proposals made by COI editors, and review process of these proposals are needed. I don't have any concrete proposals at the moment but this area definitely needs work. I have also additional proposals concerning guidelines for COI editors:

  1. COI editors are forbidden to make any edit to the page they have COI, except minor edits concerning Typos, grammar (e.g. proper tenses, punctuation) and MOS:NUM issues (e.g. proper format of dates; delimiting etc.). Right know there is no such kind of explicit prohibition but it is clearly needed.
  2. Disclosure of COI is compulsory on the editor's user page and on COI article talk pages where COI editor edits. Failure of this requirement will result with block.
  3. COI editing means also editing of pages about organizations, persons or issues, who/which are competitors or have a negative COI. This issue was mentioned by DGG in this posting. If to try to find an example, it could be (hypothetically) Shell's representatives editing Rossport Five and Shell to Sea articles and representatives of these organizations editing Shell related articles.

Beagel (talk) 11:45, 20 March 2013 (UTC)

Explicit prohibitions have met with resistance in the past. I tried to construct a narrowly construed prohibition in cases of intractable COI even with somewhat weak wording of "should not edit", it met with considerable opposition. See if you can find my intractable COI proposal in the recent archives for some background. I agree with you though, I think we do need much clearer guidance, going beyond our current guidance of "be careful". Gigs (talk) 12:46, 20 March 2013 (UTC)
So, we even don't have explicit prohibitions (only 'strongly discouraged') to edit articles but at the same time we are discussing here prohibitions to edit the talk pages. Sounds illogical. Beagel (talk) 16:43, 20 March 2013 (UTC)
Response to Beagel - perhaps I've made the "should not" section too complicated. The intention of 2-4 is not to say that the employer may not express his or her opinions and interpretations or even drafts, only that they do it in a place where we can quote or paraphrase it and properly cite it. Placing unattributed text written by somebody closely associated with the article is unacceptable. If it crosses the fuzzy line into "undisclosed advertising" then it is illegal in the US and EU. So your suggestions on how to say "may participate, in a citable form" would be appreciated.
As far as clear enforceable rules that may include some outright prohibitions, I generally agree with you. The rules we have now are interpreted in a dozen different ways - which doesn't help the companies involved at all. Outright prohibitions can help make all this simpler, but I think many folks are afraid to get away from "the encyclopedia that anyone can edit" even the slightest bit. Taking that phrase too literally to include situations that are clearly improper, situations that lead to inherently misleading edits, or even situations where the editor is breaking the law, doesn't help anybody. Smallbones(smalltalk) 14:08, 20 March 2013 (UTC)
Could you please provide an example or two when an COI editor by editing the talk page creates "situations that are clearly improper, situations that lead to inherently misleading edits, or even situations where the editor is breaking the law," and how this is unique to COI editor? Thank you. Beagel (talk) 16:50, 20 March 2013 (UTC)
I'll see if I can role this up into one example , with variations reflecting specific points.
A member of the UK's profesional PR association places (undisclosed) advertising copy on a talk page. This is clearly improper under the association's ethics code. If the material was put into the article, it would be clearly misleading as advertising copy would be presented as unbiased information. If the PR guy hired a shill to place the material in the article, it would be clearly illegal under US and German law, and probably UK law as well. COI editors are not unique in their ability to make improper, misleading or illegal edits, but the possibility comes up with COI editors and especially PR editors fairly commonly. I think it is quite important that we protect our readers, ourselves, and, yes, even the COI editors by having clearly stated rules on this. Smallbones(smalltalk) 18:01, 20 March 2013 (UTC)
Well, if this an advertisement, it does not have a place in the Wikipedia notwithstanding who is the editor or author of it. It should be removed in any case. And why to play a game with COI editor declaring his/her COI and then find a shill if you could find a shill and send the prootional material to him/her outside of Wikipedia? Or maybe the PR company is so wicked that they have planted already an editor or a number of editors who looks like regular legitimate editors, and who may insert the text without any COI declaration and without attention which would be created by proposal of the COI editor? The scheme you proposed is, of course, possible, but it will punish most hardly the COI editors who are playing by the rules, and not these wicked souls you described who do not follow any rules and will find alternative ways. Therefore, I still not convinced by your example and still think that we need more precise rules for disclosure of COI and proposals by COI editors as also more precise rules how these proposals by COI editors should be implemented instead of forbidding their contribution at the talk page. We also should make more clear the responsibility of editors making edits based on these proposals by the COI editors at the talk page. And as I said above, it is illogical that in the situation were we don't have an explicit prohibition to edit the articles we would like to prohibit editing at the talk page, so why not start with the most logical steps? Beagel (talk) 18:42, 20 March 2013 (UTC)
It almost seems like you'd like this to be as complicated as possible. I'd like it as simple and straightforward as possible. The basic idea is "COI/PR/employer text in the article needs a citable source." Nothing at all complicated about it. Smallbones(smalltalk) 19:02, 20 March 2013 (UTC)
No, I want it to be logical, transparent and non-discriminatory. Healing a headache with a shotgun is also quite non-complicated solution (at least what concerns the headache) but I have some doubts if it is the best solution. Same applies to your proposal—I just try to explain what are the weaknesses and setbacks of it and why this is not a good solution. Also, I fully agree that "COI/PR/employer text in the article needs a citable source" but actually it applies to all edits by all editors—text which is not attributed and verified by reliable sources should be removed notwithstanding who is the editor. Beagel (talk) 19:19, 20 March 2013 (UTC)
This is creating an outright bias that disadvantages paid editors, since negative POV pushers wouldn't be bound by this rule, meaning they could just edit the article regardless. It's often this issue that paid editors get involved in the first place, as negative POV pushers have completely slanted articles many times before. This is easily and obviously true of the BP article in question. SilverserenC 15:38, 20 March 2013 (UTC)

Proposal: Robust Review

To the extent that a proposed draft from a corporate representative/paid advocate is:

a) from a large (significant) company, organization, or public figure
b) about a controversial company, organization, or public figure
c) contributing a substantial amount of text or revisions
d) contributing text about the controversies themselves...
a more robust review process is needed.

I imagine this as somewhat of a sliding scale in which a small non-profit that changes a fact about their history or operations needs just cursory review, but BP editing about their environmental record warrants serious scrutiny. Hopefully we can put some kind of language like that into WP:COI and WP:PSCOI.

The next step is to arrange for some mechanism to actually provide and ensure that review. There are some possibilities to achieve that: 1) edit requests and 2) COI noticeboard messages. 3) We could create a review council staffed with experienced editors. 4) We could have a central noticeboard where these drafts are posted and reviewed (some combination of COI/N and WP:PAIDHELP, Wikiproject Cooperation and Wikiproject Paid Advocacy Watch 5) We could solicit 'opposing drafts' from company/article critics or skeptics 5) We could appoint a neutral mediator, similar to a Mediation Committee person to review the suggestions. 6) We could seek out an expert in the field with an academic background to review or counter-prospose content. 7) We could tag talk pages with connected contributor templates and article pages with COI warning templates for a period of time after the review to alert readers and allow for continued feedback from editors. These are just brainstorming. The key is that the process be transparent and robust where appropriate. Ocaasi t | c 18:19, 20 March 2013 (UTC)

I'll repeat what I wrote on your talk page. It's impossible to organize a sufficiently robust review, because we're all volunteers with different levels of time and knowledge. But if you stop to think about it, any review that was robust enough would involve us educating ourselves about the coverage of the issue in question. Once we've done that, we can write the section ourselves.
Realistically, any review would be conducted by editors with levels of knowledge significantly below that of the PR writer (or of the company experts the writer has access to). That's where the danger lies. Where the company is a large multi-national facing criminal charges, it's too much to ask a small group of volunteers to deal with. They will necessarily be overwhelmed, and the result will be what we saw on the BP article – thousands of words of BP's being copied into the article after little or no review. One of the editors copying the words over in that case didn't even realize that the PR person was officially writing the drafts for BP. SlimVirgin (talk) 18:54, 20 March 2013 (UTC)
I have to agree with SV here. There's no way we should be simply passing BP's edits through here, no matter how well-reviewed they supposedly are. Writing our encyclopedia at the behest of a major corporation, even subject to our amateur(ish) review, is the kind of reputation-damaging act our various enemies love. Well, we could sign up Noam Chomsky or Michael Moore to do the reviewing, I suppose.... No, actually I don't think so. They can challenge factual errors for which they can present evidence. That's it. Mangoe (talk) 19:32, 20 March 2013 (UTC)
My response is mainly pragmatic. As part of my research to understand this situation, I have spoken with paid editors, freelance editors, Wikipedia consultants, and others who work with corporate representatives. Shocking or unpleasant as it may be, this is happening, right now, and I believe at increasing scale. Corporations (and everyone else) recognize Wikipedia's influence and are hiring people to change Wikipedia's content. Some companies do this out in the open, with disclosure. Others do it in secret. Some get caught, while others remain undetected--how many we simply don't know. I fear, if we do not provide guidance and a review process, it's going to simply drive more activity back underground; plus we lose the opportunity to see someone's suggestions. If we don't establish that there's a 'right way' to engage with Wikipedia, we'll lead others to think that they simply have no alternative but to make changes in secret. Thus, I think we should take steps to encourage transparent engagement, and I don't see how we do that without reviewing talk page drafts with some form of proportionate scrutiny in that process. Even in the case of paid advocates we're faced with the choice of some process or no process. Right now we don't have a robust review process and one isn't even called for in the guideline. It's not a failure to institute such a process, even an imperfect one, when the alternative is nothing at all. If we say no to direct editing, and no to talk page drafts, what options are left that will realistically be followed? Ocaasi t | c 20:13, 20 March 2013 (UTC)
Ocaasi, are you willing to spend time personally doing some of these reviews? The reason I ask is that it's easy to propose something that other people have to do, but this is what has caused this situation in the first place. We suggest that companies provide drafts for others to review, but the people who are left to do that are not the people who made the suggestion. SlimVirgin (talk) 22:25, 20 March 2013 (UTC)
I've been doing much less controversial reviews on irc-help and at AfC for years (avoiding promotional language, meeting notability guidelines, finding and citing reliable sources). I also did this along with others at Occidental Petroleum and at Monitor Group when talk page requests were made, so yes, I would do this, but I typically try to find an active editor, or several of them, more expert than me to carry on the process. (I've recently been reminded how important it is to make sure those editors are seen as unbiased, or at least balanced from opposing sides). Whether I'm qualified to do this for BP, especially given my involvement in this debate is less sure. The Paid Help board at WikiProject Cooperation was designed to do this, but it is not sufficiently aligned with community consensus to scale; we'd need to rope in the COI noticeboard and WikiProject Paid Advocacy Watch (now called Wikiproject Integrity) to really get it right. I do realize the 'workload' would fall to volunteers and we're already stretched thin. I'm not saying this process would be simple, I just prefer it to the alternative. If we decided that such a process was necessary, then we'd have to organize to make it work. I imagine some editors might like to be appointed review-clerks and given the role of researching drafts on very controversial companies, but we wouldn't know for sure until we tried to set it up. Ocaasi t | c 22:41, 20 March 2013 (UTC)
Would you be willing to do an experiment? BP's latest draft for its environmental record is at User:Arturo at BP/Environmental record. (Eight or nine of BP's drafts have been added to the article already.) I know it would take me days to review this properly, and probably longer given how contentious it is, because I'd have to start from scratch. You might be faster because you've already worked on BP.
Are you willing to post that draft on a user subpage and start doing the review, so we can see you making the edits and adding or removing sources? As you're doing it, you could use a watch to make a note of each time you start and finish (whether reading, writing, planning, or going to a library to access sources), so that you can tell us how long it took overall. I think it would make people realize how long high-quality work can take in an area like this. Without that knowledge, people are suggesting things (these robust reviews) that just don't scale. SlimVirgin (talk) 23:30, 20 March 2013 (UTC)
What concerns BP, drafts which were used were about non-controversial topics. This draft you a mentioning here is about a controversial issue and at the moment there seems to be no consensus even about the structure not talking about agreeing about mass changes of this section(s). That is what happened with draft back in December and what will (unfortunately) happen with this draft although that section needs better quality than just the current WP:Laundry list. Therefore, there is no mean to review this specific draft (at least at the moment) and if not reviewed, it will not go to the article. What concerns previous drafts, I personally went through most of them. I spent hours, but it was not so hard as you described. I have said it several times but this seems to be still ignored that these drafts are not totally new – they are based largely on the previously existed text (which also makes the proposal that Arturo should put the to the BP website look strange). Therefore, this is also a reason why your simple arithmetics counting words and making conclusion that BP rewrote 40% of the article is just incorrect (this is in addition to the other reasons I said somewhere above and which are still ignored). Overall, it took much less time than all these hours people have spent during last days here and the BP's talk page and as nobody has not said what is wrong with the text which was added to the article, I have a conclusion that this process worked and actually there is no problem with the BP article regarding these texts (there are still other problems, of course). Beagel (talk) 06:05, 21 March 2013 (UTC)
  • I don't see the issue. Arturo at BP is proposing material, and other editors are using their judgement. As long as editors are using their judgement, and fix accordingly, the system works, IRWolfie- (talk) 21:09, 20 March 2013 (UTC)
  • COI editors are a problem when (A) they subtly insert a POV into articles that slips under the radar and goes undetected for a long time (if not forever) or (B) when they edit overtly and go into major conflict with other editors. When a COI editor makes suggestions and constructively works with other editors, that should be encouraged. -- Atama 06:53, 21 March 2013 (UTC)
  • It seems to me as an objective reader and... observer... that there is an inherent issue with a multibillion dollar corporation editing a webpage about itself where users come to get information about said corporation. I think I see both sides where on is worried that the corporation is acting in it's own interests to maintain its image and slant the history record, and the other side sees no issue with a corporation - or a person who happens to work for a corporation- editing what they see to be deficiencies as far as facts or events are concerned. I would contend that the later part of my last sentence is completely viable as long as there is an independent source to verify the accuracy of what is being edited. This process can go both ways as you can get an editor who has a grudge or issue with a corporation and slants the facts against the corporation. In my own opinion I think there is something intrinsically wrong with a corporation influencing specific facts surrounding incidents in which that corporation or entity was involved in. That would be like George Bush editing his own WP page to say that he did not invade Iraq. It just doesn't make sense, nor is it ethical. I understand both sides of the argument and yes BP is and should be allowed to correct facts if they are wrong but in this case it seems like BP's motives are less than noble.BlaqkMamba (talk) 17:43, 21 March 2013 (UTC)
  • But do you seriously think other editors can't tell that a suggested edit for such a change like that George Bush one would be inaccurate? Do remember that independent editors reviewed every single one of the proposed drafts for neutrality and proper sourcing and everything. No one has pointed out anything actually wrong with the drafts thus far. By all means, if you have proof that the additions made to the article are POV, point it out, but all of these baseless accusations going around about Arturo and his drafts are pretty ridiculous. SilverserenC 18:22, 21 March 2013 (UTC)

Article on CNET right now: BP accused of rewriting environmental record on Wikipedia. The subtitle: "A British Petroleum representative allegedly rewrote 44 percent of the oil giant's Wikipedia page, including the environmental sections. Some Wikipedia editors are crying foul." Mangoe (talk) 05:35, 21 March 2013 (UTC)

You know, this is always one of my biggest fears. That something I do on occasion as a hobby ends up in a major newspaper somewhere. It's happened to me tangentially a couple of times and it's one of the things that spooks me a bit about being on Wikipedia, especially in the COI areas, which have tended to get some major media coverage now and then over the last few years. I'm just glad I'm not totally transparent about my RL identity when I edit, that's the last thing I need. -- Atama 06:51, 21 March 2013 (UTC)
This isn't real coverage. Violet Blue writes these sorts of pieces all the time and they rarely have any neutrality or truth to them (because the main purpose is to attack Wikipedia). If you read the article, she's essentially quoting everything SlimVirgin said as if it was fact and at face value at that for some reason. The only thing i'm concerned about is how she found out about this discussion, as the only way I can think of is that someone here emailed her to get her to write the piece. SilverserenC 06:58, 21 March 2013 (UTC)
It has been picked up by der Spiegel Wikipedia: BP-Mitarbeiter schreibt am BP-Eintrag mit so whatever one thinks of Violet Blue the cat is getting further out of the bag. Mangoe (talk) 15:28, 21 March 2013 (UTC)
Why do we care? It's not the first time inaccurate, lying articles about Wikipedia have been written. Heck, The Register does it all the time and The Telegraph has done it a time or two. SilverserenC 16:54, 21 March 2013 (UTC)
True, I've even seen the Wall Street Journal repeat info about Wikipedia that was questionable. I believe it was in regards to the Scientology incident, they suggested that all Scientologists were banned from Wikipedia when all that was banned were connections coming directly from the Scientology offices. -- Atama 19:05, 21 March 2013 (UTC)
It's because plenty of journalists don't do the research. They just grab what one person said and treat it like the truth, then churnalism kicks in with everyone repeating the same thing. IRWolfie- (talk) 17:59, 22 March 2013 (UTC)
  • I support Ocaasi's proposal for Robust review, but the practical details escape me. We have a similar problem, not so public or controversial, happening at Chevron Corporation (CC). I was invited back to participate at Talk - significant changes requested by CC, request refused due to bias, revised request offered, editor now flummoxed due to difficulty of assessing what's biased and what's not. Process stalled, article not modified. What to do next? Editor (me) now feels bad about appearing to stall, while CC is eager to move forward with something. I would love to throw a flag called {{Request for robust review}}. Feel free to subsection this if appropriate. --Lexein (talk) 15:39, 23 March 2013 (UTC)
  • this is focussing on the wrong end of the problem in all four aspects.
1 Articles on major corporations or organizations or public figures get much more careful watching than those on smaller, and therefore are already receiving the appropriate review. It's the minor ones where the real NPOV promotional edits are usually found.
2 controversial subjects are the ones where multi-party editing is most necessary and needs to be encouraged.
3 contributing a substantial amount of text is already highlighted in edit histories, and likely to get attention. Further, an large addition is normally an attempt to add relatively factual material.It's the small insidious edits that need the watching. (What does need very careful watching are large deletions--for which we already have an edit filter to call them to attention, and substitutions of one body of text for another.
4 Text about the controversies themselves is where we need to encourage all parties to edit, so we can form a consensus and benefit from the cooperative editing. What does need watching is attempts to add a controversy section, because though some are valid, some are attempts at adding POV negative material. DGG ( talk ) 20:41, 24 March 2013 (UTC)
  • With the benefit of some intervening days, I think you've overgeneralized in two places, DGG:
1: Articles about large corporations, in my experience, do get some oversight by the community, but still not enough IMHO. Example: Vonage. Example 2: At Talk:Chevron Corporation, it was just me and one other editor, and he had to drag me back in by my Talk page to jump back in about the requested edit (still pending). I was thinking (with love), "Where are all the lefty bleeding heart tree huggers when you need them?" I didn't know where to go to drag more eyeballs into that page, sad to say.
3: "normally an attempt to add factual material" - that's not supported in my experience. Big 'ol chunks o' hoo haw, is what I've seen. Example: (again) Vonage. It was so egregious that it was called out on Reddit, but it blew by me and every other editor watchlisting it (see Talk:Vonage). At Talk:Chevron Corporation, the first requested edit removed previously used independent RS (a few were advocacy sources, admittedly), and added mostly primary RS. If that had gone in, it would have stayed in, unchallenged for a long time, IMHO.
But I have no quarrel with your (I think) intention. --Lexein (talk) 01:32, 29 March 2013 (UTC)

Arturo may be compelled to edit request edits by BP

I've got a feeling that Arturo may be compelled to edit the article as part of the editor's duties at BP. If so, Arturo is faultless and BP is to blame. Is there a way we can notify BP that this is a violation of Wikipedia policy? — DragonLord (talk/contribs) 16:08, 21 March 2013 (UTC)

Compelled how? And this isn't a violation of any Wikipedia policy. SilverserenC 17:32, 21 March 2013 (UTC)
How is this not a violation of policy if there is a conflict of interest? My point is that the user may be required by his/her boss at BP to perform the edits. — DragonLord (talk/contribs) 17:48, 21 March 2013 (UTC)
Have you ever read WP:COI? Editors with a conflict of interest are even allowed to edit the articles directly however they want. They are discouraged from doing so because of the difficulty of writing neutrally when one has a COI, but there is absolutely no restriction on doing so. And, instead of editing directly, Arturo has even been going a step above and has been putting up suggestions on the talk page for review and allowing other people to implement the material if they feel it is appropriate. If you feel that he is being "compelled", why don't you go ask him? SilverserenC 18:11, 21 March 2013 (UTC)
(edit conflict)First of all, COI isn't a policy. Secondly, there's no policy or rule that prohibits anyone from being paid to edit, see WP:PAID for a proposed guideline (there have been a number of proposed guidelines and policies, but none gained consensus). You'd have to find Arturo violating one of our actual policies, like WP:NPOV, WP:BLP, etc. -- Atama 18:11, 21 March 2013 (UTC)
Arturo may be compelled to edit the page? Arturo hasn't edited any article, let alone the BP page [1]. IRWolfie- (talk) 18:26, 21 March 2013 (UTC)
Compelled to request edits, I think. I've revised the section heading above. --Lexein (talk) 15:45, 23 March 2013 (UTC)
What's wrong with that? The more suggestions, the better eventual article. And much much better by a known person than someone hiding from scrutiny. DGG ( talk ) 20:47, 24 March 2013 (UTC)
I just wanted to clarify his actual actions, and to clarify the title of this section. The discussion seemed like it was focusing on his "edits" as if they were to articles. In reality they were only in Talk, and that's the WP:COI suggested practice. I'm not arguing a case here, DGG. --Lexein (talk) 09:00, 26 March 2013 (UTC)

Please provide evidence of POV-pushing by Arturo

Can anyone on either side of this debate please provide some evidence of the alleged POV-pushing or white-washing being done in Arturo at BP's drafts? A lot of people seem to be "crying foul" over the mere fact that a corporation is paying a person to edit, rather than evaluating the content of those edits. I'm not prepared to take sides yet because I legitimately can't tell if there is anything shady going on here, based on the limited amount that I've seen. Axem Titanium (talk) 21:35, 21 March 2013 (UTC)

I think we need to start routing comments better. This page is more for discussion about potential changes (or not) to the current CoI policy. The Talk:BP page is a better place for discussion of specifics of this incident. -- [ UseTheCommandLine ~/talk ] # _ 21:56, 21 March 2013 (UTC)
I've asked the same thing multiple times on the BP talk page and have yet to get a response. SilverserenC 22:03, 21 March 2013 (UTC)
The sentence that alerted me to this situation was Arturo's most recent draft, which Silver seren was about to insert. It began: "In the 1990s and 2000s, BP has had a mixed environmental record, according to government regulators, journalists, activist groups and environmental monitors." The source used [2] doesn't really support that sentence, and it's not the best source to use. Lots of high-quality sources, including the New York Times and PBS, indicate that BP's poor record goes back certainly to the 2000s, at least. Other sentences are red flags too, e.g. that even after the Prudhoe Bay spill in 2006, "the company continued to receive praise in the media for its investment in alternative energy and its focus on greenhouse gas emissions." Did it?
The whole thing is problematic: (a) have the right sources been used; (b) do they say exactly what BP has used them for; (c) is anything missing; (d) is anything carefully worded; (e) has the correct weight been given to all the issues? And so on. Checking this thoroughly – making it policy-compliant – would involve a lot of research. It would be faster, easier and more ethical to write it from scratch. SlimVirgin (talk) 22:25, 21 March 2013 (UTC)
"which Silver seren was about to insert"
That's an outright lie. As you very well know, I was actually in the middle of pointing out other sources (covering negative information, mind you) that I felt should be included in the draft.
As for BP's history, the sources are quite clear on that. While they have been criticized for the oil spills and other incidents, BP has still be ranked year after year in a number of lists just within the past five years, as the top oil company in respect to its attention to environmentalism and renewable energy and climate change. SilverserenC 22:40, 21 March 2013 (UTC)
"BP has had a mixed environmental record" is cited to an article titled "BP Shows It's the Best and Worst at Once". Sounds pretty mixed to me.
"It would be faster, easier and more ethical to write it from scratch". If this were the case, why hasn't someone else just done that? The fact of the matter is that someone is being paid to spend effort to create a "first draft" of new content which no one else has spent the effort to do. I don't see a reason to discard that effort, regardless of the source, and even if that first draft requires other editors to scrutinize it more. The way I see it, allowing people like Arturo to do whatever it is he's doing isn't going to prevent any other non-COI regular editor from doing a "faster, easier write from scratch" at any time since that regular editor was going to spend that effort anyway. Until that munificent editor strolls along, let the regular editors of BP spend their effort improving/NPOVizing/de-promo-ing Arturo's prose, if that is indeed a problem. Axem Titanium (talk) 01:50, 22 March 2013 (UTC)
Axem Titanium, what you've explained is one of the main reasons why we've never had a policy that forbids COI contributions. Sometimes the COI editor is more able or more willing to do the work to create or improve an article than other editors. Preventing them from contributing based on a nebulous principle that the article is somehow "tainted" by those contributions is not in Wikipedia's best interests. (That's not an attempt at a straw man argument, I'm not suggesting anyone here has actually suggested that, but that's often the underlying objection to COI contributions.) As long as there is oversight, as long as everything is done transparently, and especially as long as there is no conflict between Arturo and other editors at the article, there shouldn't be any reason why Arturo's contributions should be discounted (especially since they're being presented to the talk page for review before other editors). -- Atama 02:37, 23 March 2013 (UTC)

Wikipedia editors still have editorial control of articles

I have watched this become so overblown I can't help but wonder how it got this far...oh wait a minute, now I remember. It was because someone decided to create a figure that I have yet seen qualified that the media has now attached themselves to as gospel and now the community has attached itself to that. This is by far the worse case of circular referencing I have ever seen on Wikipedia. Someone should be ashamed of themselves. Editors are forgetting that a contributor is not a COI editor if they are editing within policy. The page here makes that clear but I still see people calling anyone that has a close association of any kind a COI editor. It ain't that simple. Perhaps the page needs to clarify this in much stronger terms.--Amadscientist (talk) 10:42, 23 March 2013 (UTC)

Proposal for additional wording to the lede

I propose text be added to "When advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest" to be followed by "When editing within Wikipedia standards, editors with close associations are not considered in conflict of interest and are advised to edit with extreme caution to avoid a perception of COI editing"--Amadscientist (talk) 11:11, 23 March 2013 (UTC)

  • Support as proposer.--Amadscientist (talk) 11:11, 23 March 2013 (UTC)
  • Oppose only due to its rather extreme flip of emphasis. We (I) really want to clearly declare that COI is defined by association, and that a (presumed) COI editor must work hard to maintain trust by avoiding all activities which work against Wikipedia's goals mainly by sticking with the WP:COI content and behavioral guideline. --Lexein (talk) 15:55, 23 March 2013 (UTC)
    • I kinda like that wording you used and see your point. Since we don't actually have a strict policy against close associations editing articles that are associated with (just sternly worded suggestions and advice) we just need to make it clearer about that. Let me try some of your suggestions above.--Amadscientist (talk) 23:28, 23 March 2013 (UTC)
  • Oppose COI exists regardless of your behavior; it's a property of contributors. Neutrality on the other hand exists precisely dependent on your behavior; it's a property of conduct. Ocaasi t | c 17:45, 23 March 2013 (UTC)
    • No, that is incorrect. Editors are not in conflict of interest if they are following policy and guidelines properly. You do indeed have to "behave' in a manner that crosses a line to be in conflict. Just being related to someone does not make a conflict of interest.--Amadscientist (talk) 23:29, 23 March 2013 (UTC)
  • Cross purposes: two separate kinds of conflict exist. The association itself establishes the existence of conflict of interest, and is unambiguous, and fixed. A corporation's interest simply is not and cannot ever be, the interest of Wikipedia, for too many reasons to list here. There is a small intersection of interests and actions, and a way to act within that sliver, by our policies and guidelines: our table, our etiquette. Actions taken (edits and edit requests) can then conflict or not conflict with WP:COI and other WP guidelines and policies. It's unfortunate that we use one word to refer to several things. A COI editor at an "interested" article is never not a COI editor, but the edits done can be WP:COI-compliant or not, IMHO. So I think it's important to clarify our grammar wherever possible. --Lexein (talk) 20:45, 31 March 2013 (UTC)
  • Oppose. Whether you have a conflict of interest depends on your roles and relationships. It has nothing to do with behaviour or opinions. If you are a judge and your husband is the defendant, you have a conflict of interest, no matter what you think of the case, no matter how you would have handled it, so you recuse. The argument that if you follow the relevant rules you don't have a conflict is to misunderstand what is meant by COI. SlimVirgin (talk) 23:40, 23 March 2013 (UTC)
    • That's COI as it concerns the off Wiki definition. Yes, there is a misunderstanding of what a COI is...on Wikipedia. We have a different standard as to what constitutes a COI editor and behavior IS part of that...or did you miss the disclaimer on the article here . This is a "behavioral guideline". But I think that we could add a footnote instead of additional wording to the lede itself.--Amadscientist (talk) 23:43, 23 March 2013 (UTC)
  • We use the same broad definition as anyone else, namely whether any external relationship/role could reasonably be said to undermine your primary role as a Wikipedian, which requires putting Wikipedia's interests first while you're editing. Ocaasi put it well above: COI is a property of contributors, while neutrality is a property of conduct. SlimVirgin (talk) 23:55, 23 March 2013 (UTC)
Actually we don't use the same broad definition. This is a behavioral issue. One is not in conflict just for having a close association. One does have to cross a line to be "in conflict". I trust Ocaasi's good faith even if the wording seems rather confusing to me.--Amadscientist (talk) 23:59, 23 March 2013 (UTC)
Amadscientist is correct. We don't use the normal definition of COI. Our COI is based on an actual conflict: you want X, and Wikipedia wants not-X. This failure to use the standard definition—the one in which a husband removing libel or vandalism from his wife's BLP has a "conflict of interest" rather than what we actually have, which is a synergistically aligned interest that Wikipedia has every reason to encourage—is a source of perpetual confusion. I'm increasingly sympathetic to suggestions that we rename this guideline to something like WP:Damaging Wikipedia due to a conflict of interest. WhatamIdoing (talk) 03:36, 25 March 2013 (UTC)
  • Oppose I agree that there are three distinct concepts 1. COI editing 2. Having a COI 3. Being seen to have a COI. Unless we separate these we will go nowhere. Rich Farmbrough, 04:09, 24 March 2013 (UTC).
The text spells out that one must cross a line to be in conflict so whether other editors agree or not the text makes it clear what are guideline is here. I don't see any ambiguity but many editors here seem to.--Amadscientist (talk) 20:52, 24 March 2013 (UTC)

Instead of adding additional wording to the lede

We could actually add a footnote as was done at WP:VERIFY.--Amadscientist (talk) 23:38, 23 March 2013 (UTC)

COI - does it matter if they are paid?

Currently editors who have a COI who are paid effectively can not edit the article in question. Why was this not extended to all advocates? IRWolfie- (talk) 02:35, 24 March 2013 (UTC)

Simply put, that is not accurate. They are very strongly discouraged. So are editors that indeed do have an actual conflict of interest, who have been found to be editing in a manner that could compromise the project.--Amadscientist (talk) 03:11, 24 March 2013 (UTC)
They aren't as discouraged as much, as the text of COI makes clear, IRWolfie- (talk) 18:47, 24 March 2013 (UTC)
Obviously the text in COI makes nothing "clear" if people don't even agree with the basic principle that: "When advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest". To me that spells it out very clearly that just having an outside interest is not COI, but when advancing them is more important than our goals, then they are. Yet, even when editors are complying with this they are accused of COI editing. This is unlikely to change until editors can agree on the simple definition of what a COI editor even is.--Amadscientist (talk) 20:57, 24 March 2013 (UTC)

Our way or the real world's way?

We've just acquired a bunch of text re-defining COI according to the real-world standards. As many editors have noted elsewhere on this page, Wikipedia does not use, and never has used, the corporate-standard definitions.

Real-world corporate definition
You have a position of fiduciary trust at Company A. You also have a position of fiduciary trust at the unrelated Organization B. This represents a potential conflict of interest. Company A wants to cut a deal with Organization B (or perhaps Organization B is a newspaper, and the editor just asked you to write an article about your other employer). You now have an actual conflict of interest: you have a duty to work in the best interests of two organizations on opposite sides of a deal. (Solution: recuse yourself.)
Wikipedia definition
Wikipedia has a goal (good, neutral, encyclopedic articles). You have a goal (to get unverifiable garbage out of an article about your business). There is no conflict of interest here. Wikipedia's interest and your interest are exactly aligned. Wikipedia's interest and your interest are not in conflict at all. (Solution: make the edit.)

Now, we could change our definition to match the real-world definition, but I don't think that there's significant support for it. I think that the more sensible approach would be to remove all of this information about real-world concepts from the guideline as being irrelevant to Wikipedia's approach. What do the rest of you think? WhatamIdoing (talk) 19:44, 26 March 2013 (UTC)

I think WP's definition hews more to the definition of conflict-of-interest in the medical context, as outlined here. I think your assertion of the corporate definition as "the real world's way" comes off as being a bit dismissive. -- [ UseTheCommandLine ~/talk ] # _ 20:09, 26 March 2013 (UTC)
It's not practical to just tell PRs not to get involved, when (a) we have an obligation to our employer/client to do our jobs and (b) our reputations are often treated unfairly on Wikipedia. Besides that there's (c) we often make substantial improvements and correct errors. However, if we're going by the real-world, we don't call it COI at all when a PR person works with the editors of an independent site. That's just called PR. And it would be very odd for a PR person to tell a journalist their coverage is bias and they want to re-write the controversies themselves. But then Wikipedia is a very unusual site.
I would actually advocate for (in full knowledge it's a losing battle) that we abolish PR as "COI" and create a guideline on "Doing PR on Wikipedia" just as many news organizations provide guidelines on how they want to receive news, bylines and other pitches, that outlines how Wikipedians want PRs to behave, what type of participation is valuable, what is acceptable and not, etc. CorporateM (Talk) 21:09, 26 March 2013 (UTC) (PR guy and frequent COI contributor)
The existing fudge is actually pretty good. Don't edit the article directly, say who you are, propose changes on Talk, leave others to make the actual edits. It even scales, for most subjects. It's understandable to PR because it's analogous to how newspapers and the trade press work - you send press releases, they review them and they review it. Mind you, given the prevalence of churnalism these days maybe that accounts for the confusion too. Guy (Help!) 00:05, 29 March 2013 (UTC)

I think that the initial issue has been incorrectly stated. The real-world definition of conflict of interest is much more stringent. Not only is conflict of interest (defined broadly) prohibited but so is the appearance of conflict of interest. Wikipedia seems to be an outlier that takes an almost comically lax view of COI. This has become evident for all the world to see in this BP incident, where BP is drafting entire chunks of text and many Wiki people are defending that, including Jimbo. What all are ignoring, including Jimbo, is how this hurts Wikipedia's credibility. It doesn't matter whether BP is adding good or bad text. The only thing that matters, from a real-world COI perspective, is that BP is participating actively in an article about BP. The debate over this on various Wikipedia pages has actually hurt Wikipedia's credibility even more than the BP edits, as it has exposed Wikipedia's tolerant attitude toward large corporations actively participating in their articles. This will encourage other companies to do so. Coretheapple (talk) 14:01, 27 March 2013 (UTC)

An appearance of a COI is not illegal in the real world. Most large corporations ban it, but most small businesses are happy to live with it.
I personally doubt that Wikipedia's reputation is affected much by anything other than individual readers' personal experiences. 99% of the world has no idea that BP (and other companies) is trying their best to follow the rules that we put forth. WhatamIdoing (talk) 00:25, 29 March 2013 (UTC)
  • Our traditional definition of COI is closely aligned to the academic definition: any interest which may be in conflict with neutrality, must be declared. Guy (Help!) 00:02, 29 March 2013 (UTC)
    Actually, we don't require disclosure. We only encourage it, and hope that good-faith helpful edits won't be reverted by people who get hung up on the editor's self-disclosed identity. WhatamIdoing (talk) 00:25, 29 March 2013 (UTC)
Fair point, but an undisclosed COI tends not to end well. Guy (Help!) 16:52, 30 March 2013 (UTC)

Here's my basic question: SV added about 5000 characters' worth of stuff about real-world definitions of COI. Do we want to keep this here? WhatamIdoing (talk) 00:25, 29 March 2013 (UTC)

It's the standard academic view. The problem we've seen on this page is some editors not knowing what a COI is in the real world (which, as Guy says, is how Wikipedia uses the term too). It's therefore important to be anchored by mainstream academic sources, so we know we're not making up our own definitions. I can't see how it could be a problem. SlimVirgin (talk) 00:35, 29 March 2013 (UTC)
Yes. We need to be clear on what it is, and what it means for people dealing with content edits. It's pretty clear that understanding is muddled for at least some people. There is a small group who believes paid editing is absolutely OK (though that's often external as the most prominent advocates are banned); there is a small group who thinks that nobody with a conflict of interest should touch Wikipedia at all; and there's the pragmatic middle ground, which is where we should be - and generally are give or take the occasional human error. Guy (Help!) 16:51, 30 March 2013 (UTC)


I think we need to be practical:

  • If you are a corporate PR person and find vandalism, slander, mis-spellings, grammar, broken links, poorly formatted references, or other non-controversial housekeeping items needing to be done on your employer's page, just go ahead and do them. If somebody complains, assume the edit is controversial and don't repeat it.
  • If you want to add, remove or change content, then ask via the talk page. Edits that could be controversial should not be done by somebody who is working for the subject.
  • There are other situations not enumerated where common sense should apply. For instance, the subject of an article should be able to remove {{prod}} but probably should not get involved in {{AfD}}. We should not try to list all the fine details because the nuances are complex and could change depending on each situation. Jehochman Talk 19:23, 30 March 2013 (UTC)
I disagree with the suggestion that "Wikipedia has a goal (good, neutral, encyclopedic articles). You have a goal (to get unverifiable garbage out of an article about your business). There is no conflict of interest here." I see this as being as clear a COI as organisation A sells product B, organisation C is considering buying product B. As someone involved in both organisations I declare an interest, recuse and leave others to negotiate a deal. The key point here is that even where you think the two organisations align and don't conflict you need to recuse - COI is not just for scenarios where you are involved in two organisations that are bidding against each other to buy the same unique product. ϢereSpielChequers 19:54, 30 March 2013 (UTC)
So if I'm the paid professional publicist for a BLP, and I see unsourced, unverifiable libel in the article about my client, do you really think that what I want (to remove the libel) and what Wikipedia wants (to have the libel removed) would be better characterized as being "the same" or "different"? Where is the "conflict" in our interests? WhatamIdoing (talk) 15:02, 3 April 2013 (UTC)
I think the academic background Slim has provided is very interesting and it's worth asking whether our guideline reflects the academic and professional consensus on COI. However, I think that discussion belongs first on our talk page (here), so that we can look at incorporating such language into our actual guideline. I don't see the value in citing academic consensus for mere background, just like I don't support citing laws that are not binding on Wikipedia itself. Instead we should streamline the guideline to reflect the parts of scholarship and law that we want to take on as our own. Otherwise it's tangential and has no impact. I think editors should propose specific changes to definitions and advice rather than merely reference them in what amount to background and context sections--these seem out of place for me in a guideline unless they have some direct bearing on what the guideline suggests or requires. Ocaasi t | c 20:00, 30 March 2013 (UTC)
Our guideline has never reflected the academic, legal, or professional consensus on COI. WhatamIdoing (talk) 15:02, 3 April 2013 (UTC)
Wikipedia is part of the real world. We can't make words mean whatever we want them to mean, and there's no point in people talking about COI without knowing anything about it. I agree with Jehochman that this boils down to common sense, and recognizing that there are different levels of COI and different levels of involvement, but common sense hasn't been the most noticeable attribute of these discussions. At least if we know we're using words the way the rest of the world uses them, that will be progress. SlimVirgin (talk) 21:54, 30 March 2013 (UTC)
We can too make words mean whatever we want them to mean, and we do it all the time. We have redefined "Notable" and "Neutral" to mean things that totally surprise new users, and we do it here, too. WhatamIdoing (talk) 15:02, 3 April 2013 (UTC)

Proposal: new article tags for COI editor input

On my sandbox [3] are several possible alternative proposed article tags, which can be utilized in situations in which the subject of the an article is involved in the editorial process on the talk page, beyond pointing out factual errors and bias, and/or has drafted text that is either copy-pasted or largely adopted by other editors. Such practices are allowed by our policies, but no suitable article tag exists to address such situations. Input from editors interested in this situation is definitely welcome there. Frankly I don't see why any of the tags being proposed so far can't be utilized where appropriate.

The assumption behind these tags is that the current ones that we have don't cover situations that are problematic and a potential embarrassment for Wikipedia, and areas in which disclosure to the reader is warranted but not currently the practice. I think that readers should find out from Wikipedia, not the media, when such things take place. Coretheapple (talk) 16:42, 31 March 2013 (UTC)

I do like the last one, though maybe something a bit smaller for a section heading? I also continue to think that these edits should be tagged somehow in the edit history, with the source and original text if necessary. If we are going to, either implicitly or explicitly, make it convenient and easy for outside actors with monetary interest in what is said on their WP page tohave some sort of control over what is said there, then we also need to similarly enable not just the current review of that material (such as via [[[User:Occasi|Occasi]]'s proposal for robust review) but also the future review of that material. -- [ UseTheCommandLine ~/talk ] # _ 20:10, 31 March 2013 (UTC)
Yeah, the last one, which i think was drafted by Ocasi, is definitely a keeper. Coretheapple (talk) 20:25, 31 March 2013 (UTC)
How do you envision these being used? You put them at the top of an article, and then what? Leave them there forever as a badge of shame or a warning to the reader? Take them down as soon as one person looked it over for puffery? WhatamIdoing (talk) 15:04, 3 April 2013 (UTC)
I'd like to see them come off when the article and its talk page are both free of COI influence. i.e., the PR people and corporate or government reps or personal press agents or subjects of articles are no longer proposing text, no longer participating in article decision-making, except to correct BLP issues and errors. These tags are predicated on the expectation that readers have that Wiki articles were conceived, written and executed by disinterested editors. When these expectations are not met, when the PR reps and the article subjects themselves are involved in the article decision-making, they need to know that their expectations are not being met. It doesn't have to be a permanent tag. The amount of time the tag is there would be totally up to the subject of the article. Once he or she or it withdraws, the tag comes off. That's how I saw this. (By the way, if I don't respond further it's not disrespect, but I am off-Wiki for a few days and not checking in here too much if at all.) Coretheapple (talk) 03:59, 5 April 2013 (UTC)
Better: "Some of the material on this page may have been provided by the article-subject" (at the bottom as a disclaimer)." We already have bias tags. But I think what is more appropriate, is that if editors are uncomfortable with the behavior of a PR person (undue influence, arguing and nit-picking in controversial areas, advocacy, etc.), they should tell that person and their effectiveness in contributing here should rely - in part - in behaving appropriately.
I do find that Request Edits that involve pointing out an error or something missing tend to be more helpful than "can we re-write it this way?" A lot of Request Edits are just asking to re-write material in a way that's more promotional or aligned with corporate messaging with no reasonable rationale for it. I think the guideline would be more helpful if instead of "COI editing is discouraged" - article-space editing was practically outlawed, content contributions were discouraged, corrections, images, sources and other help were encouraged, based on the ratio of how often each are helpful. CorporateM (Talk) 13:31, 15 April 2013 (UTC) (frequent COI contributor)
That's good language in certain specific situations, but I was thinking of something that would more broadly deal with article subjects (and their reps, employees, etc.) shaping the article content. Coretheapple (talk) 13:36, 15 April 2013 (UTC)
Actually I added "may have" - being that we never know for sure if an editor is a PR representative even if it is declared. CorporateM (Talk) 16:42, 15 April 2013 (UTC)

whether eventually to keep or delete a Connected Contributor tag on an article's talk page after the article's COI tag is deleted

Should the {{Connected contributor}} template be deleted from an article's talk page or kept there even after the {{COI}} template was deleted from the article after a review for COI effects or NPOV? In this case, the conflict of interest was undeclared and the COI editor was not blocked or banned, but it would be useful to know regardless of particulars of a COI or an editor. Although I'm asking about the use of the template where another editor has the COI, I also have a COI for that same article and have not edited the article except to add the COI template into a blank line, but I have posted to the talk page. Regarding the talk page template, I have no opinion. I previously asked at Template talk:Connected contributor#whether eventually to delete this template from an article.27s talk page or keep it there, but no one replied. So, I am asking here. Thank you. Nick Levinson (talk) 16:32, 3 May 2013 (UTC)

FTC rule

An update of the FTC's interpretation of online advertising rules is at [[4]]. Of interest is that it covers any advertising, marketing, or promotion online (well presumably any that reaches US consumers), and that the disclosure requirements, e.g. for paid endorsers, are very strict. I think a reasonable interpretation of this would be that any promotional activity for a product or service on Wikipedia involving a paid endorser would have to be disclosed very near the promotion on the article page. Since Wikipedia does not allow the disclosure of the editor on the article page, this will cause quite a problem. Any undisclosed promotion on the article page, would be illegal, and thus against Wikipedia's terms of service. In short, it looks like WP:COI needs to be tightened up. Smallbones(smalltalk) 19:50, 22 May 2013 (UTC)

I drafted some possible article templates a while back.[5] Coretheapple (talk) 20:26, 22 May 2013 (UTC)
Frankly those don't look strong enough to me. They couldn't be just placed at the top of the article, but would (according to the FTC) have to be near the paid contribution and likely identify the paid contributor. The disclosure has to be very direct. I don't think we'd allow this type of disclosure, e.g. "the following paragraph was written by Mr. XYZ who works for Company ZZZ." What would then happen if somebody else edited that paragraph, or moved half of it to another location? Just wouldn't work IMHO. Smallbones(smalltalk) 21:24, 22 May 2013 (UTC)
Example 21 suggests that a disclosure at the bottom of an article would not be considered a "clear and conspicuous" disclosure. However, I don't see there ever being consensus to add it to the top.
How the FTC's disclosure laws apply to Wikipedia is probably not so simple to unravel and it would be interesting to get an opinion from the FTC itself (or a lawyer at least). The guide applies to all mediums, but is focused mostly on advertising or at least on channels that have authorship. Wikipedia is unique in the regard that readers never know the author of any article, which cannot be said for tweets, blogs, ads, television shows, radio, news media, etc. Additionally, the authors of the article are not the source of the communication, rather the references are. (or maybe, this is confusing to think about philosophically)
However, I do think interpreting the law is a much more constructive way to establish best practices. At the very least everyone should agree that it is unlawful for a PR person to create blatant advertisements through anonymous editing. CorporateM (Talk) 23:14, 22 May 2013 (UTC)

"Non-controversial"

My personal experience as a PR person has been that in almost every case, this clause is misunderstood even by good-faith participants with a COI. I believe the confusion lies in this:

  • People with a COI believe it's referring to "controversial content" like controversies or lawsuits
  • It is actually referring to "controversial edits" such as anything that could be reasonably contested

I would kindly suggest something less prone to confusion, such as "clerical edits." CorporateM (Talk) 17:05, 2 June 2013 (UTC)

Like the phrase "controversial edit," "clerical edit" may mean different things to different people. To some, it may mean fixing typos. To others, it may mean routine work done under the direction of or on behalf of another, such as that of an office clerk. Someone out there will probably think it has to do with religious edits *joke*.
I've made and reverted this edit which makes it clear what a "controversial edit" is. 1) Are you okay with this as a solution to the problem, and 2) are there any objections to anyone else? If you are okay with it and there are no objections within a few days, I'll probably make it "go live."
Note - my proposed clarification makes a significant change to the text: It changes

If another editor objects for any reason....

to

If another editor acting in good faith objects for any reason....

This is to explicitly state what is already implied in this and most other policies/guidelines. davidwr/(talk)/(contribs)/(e-mail) 21:19, 2 June 2013 (UTC)
A few thoughts:
  • I think the criteria that no reasonable editor could possibly object to the edit is a good bar to set, rather than whether an editor actually did object, as most of our company articles are not closely watched.
  • Currently grammar, citation fixes, etc. seem to be on-target with how non-controversial edits are defined
  • We must consider not just the criteria itself, but how it will be interpreted by COIs who are inexperienced, have poor judgement and are likely to interpret it in a way favorable to their edits, even if unintentionally
  • I feel uncomfortable having someone ask me if an edit is acceptable even here, where I have no COI (unless you believe I am conspiring to make my PR edits more widely accepted), but do have the "appearance of COI" (or something), which is a concept I think would be useful to introduce.
Consider the following circumstance, which I find very common. The article on XYZ contains a list of vendors. In general, a volunteer would delete the list on-site, but a COI sees the list and believes this is an indication that it is acceptable practice. They would think that it was a non-controversial edit to add themselves to the list, when it is not.
Therefore, in addition to the criteria that no reasonable editor would object, I would suggest that we ask PRs to defer to a disinterested editor anyway as best practice, or do so if they are not 100% confident it is indeed a non-controversial edit.
I've seen a lot of Request Edits, where I could see the editor thinking it was a non-controversial edit, but boy was I glad they used Request Edit instead. And there's no harm in asking. The exception being grammar and such, where a Request Edit is just annoying and unnecessary. CorporateM (Talk) 15:03, 5 June 2013 (UTC)

About the "law of unintended consequences"

On a few occasions throughout my WP "career" I have used the following sentence to explain to concerned people why negative information about them or their connected person/org will not be removed: "Subjects of Wikipedia articles do not have the right to not be embarrassed."

A few editors have commented how well the sentence sums up the situation - and one editor remarked that my "double negative" construction was particularly good. So, I present to you this sentence, to use as part of a standard explanation template text if you think it is suitable. Roger (Dodger67) (talk) 20:27, 7 June 2013 (UTC)

If it were phrased more directly, it would sound something like "We have the right to embarrass subjects of Wikipedia articles", which is the same content but does not make us sound like kind people. WhatamIdoing (talk) 22:08, 7 June 2013 (UTC)
That's a bit too direct and not really accurate, it implies we intentionally and maliciously set out to embarrass subjects. A better explanation: "Subjects do not have the right to remove properly verified content just because it is embarrassing." Roger (Dodger67) (talk) 07:01, 10 June 2013 (UTC)

Foontroo Incorporation

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


Foontroo is a privately held company in Indian market that is primarily an internet domain registrar and web hosting company.We are providing all Information Technology and Software services.Including Email Web services Etc. — Preceding unsigned comment added by Foontroo (talkcontribs) 14:56, 14 June 2013 (UTC)

Archiving per this edit which removed this section entirely. Keeping in case another account is connected to this one at a later date, the COI declaration will carry forward. davidwr/(talk)/(contribs)/(e-mail) 19:50, 15 June 2013 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

propose to edit for what to do for article talk and separating discussions

Notice provided on Template talk:COI#Proposed new parameter: concern.3D

I propose to add to the guideline that how to handle a conflict of interest should include posting to an article's talk page what editorial issue/s result from the conflict. This is because subsequent deletion of the {{COI}} template depends partly on resolution of issues raised on that talk page, but this guideline doesn't mention posting to the article's talk page as part of what to do. Either the template documentation should be changed or the guideline should be, and changing the guideline seems the likelier path. In that case, I also plan to combine this with the existing first step, discussing with the editor, which sounds like posting to the editor's talk page about the conflict. That's because the editor should be notified but any editor watching the article should be encouraged to help with neutralization of content, at least after the possible-COI editor has had a chance to do so in non-BLP contentiousness-repair cases. I'll wait a week for any comment. Nick Levinson (talk) 18:47, 15 June 2013 (UTC)

I've drafted a change to {{COI}} to make it easy to add a concern directly in the template. I have not updated the template's /doc though. See Template:COI/testcases for an example of how this might work. davidwr/(talk)/(contribs)/(e-mail) 20:16, 15 June 2013 (UTC)
Interesting idea; maybe it's a better alternative. Thanks. Nick Levinson (talk) 20:58, 15 June 2013 (UTC)
No one acted on the above post by editing the template, so I edited the guideline to encourage adding the concern to the article's talk page. If the template is edited per the suggestion, the new guideline provision already covers that. Nick Levinson (talk) 19:29, 23 June 2013 (UTC)

Suggested additional guideline for existing users

I suggest that an additional line be added to this line pertaining to our existing users as follows -

If a user has disclosed a COI or disclosed their paid editing interests here, other users are suggested to treat their suggestions and discussions as if they had been suggested by another non-COI editor.

I believe this additional line is important because we currently have a negative incentive for the COI editors to declare their interests. If they do not declare it, they might even continue to be acting functionally, but if they do, their actions are faced with a great deal of backlash. This suggestion is in line to remove that negative incentive so that editors working under COI are encouraged to be open with their editing, rather than hide behind anonymous accounts, which ends up hurting the encyclopedia even more. TheOriginalSoni (talk) 18:27, 7 June 2013 (UTC)

P.S. Feel free to make any minor tweaks with the wording, if you think so. Please also add a note in the discussion stating your change.

  • For those who might be concerned with this addition, I must also make another point which should be considered about COI editors - If their additions are not being constructive, they would be breaking atleast some other policies too. Conversely, if they are not breaking any other policy, i see no reason why their contributions must not be taken for what they are, and not who makes them. TheOriginalSoni (talk) 18:44, 7 June 2013 (UTC)

Discussion

    • I must not understand. Are you proposing that once an editor discloses their COI on a central place, they must thereafter be treated as if they had none? But the entire point of their disclosing their COI is so that their edits can be watched to see if the COI influences it. DGG ( talk ) 02:49, 11 June 2013 (UTC)
  • Whether or not the edits are neutral can be judged based simply on the edits themselves, and the fact that there is a COI would not change that. I do not think anyone's reaction to the edits would/should change depending on whether the other party have COI or not. TheOriginalSoni (talk) 03:26, 11 June 2013 (UTC)
  • I believe that some COI edits may be ok (I have committed one myself today — go look in my edit history) but that we should continue to view paid edits, when discovered, with extreme prejudice. I think having a guideline telling us we shouldn't do that is a mistake. —David Eppstein (talk) 06:10, 14 June 2013 (UTC)
  • I agree with User:David Eppstein. This proposed change would be a giant step backwards. The COI rules are weak enough. This change would make them even worse than they are, and make Wikipedia even more of an oddball and an outlier on conflicts of interest than it currently is. Coretheapple (talk) 13:16, 14 June 2013 (UTC)
  • Comment from a COI contributor: This is related to the issue of stereotyping COI editors. If you look at other stereotypes, they are based on a demographic that is statistically more likely to have a certain attribute (criminal behavior, poor driving, etc.) than other demographics. We frown on these stereotypes, because it is still unlikely that a specific member of that demographic will exhibit the stereotype, even if they are more likely to do so than someone else. The difference with COI, is that the editor is not just "more likely" but "most likely" to make COI edits. And so it's impossible to fight against the stereotype when our experiences will be that they are true in almost all cases. I think DGG worded it best why we shouldn't pretend the COI doesn't exist, but some other wording may be more palatable. For example, by saying we shouldn't assume their suggestions lack validity. CorporateM (Talk) 13:42, 14 June 2013 (UTC)
That really misses the point. This has nothing to do with "stereotyping." The language proposed would undermine the very purpose of having a COI guideline. It could impact on the talk page consensus so as to allow conflicted editors to shape the discussion and editorial slant of an article even more than they already do in some pages. This problem would be particularly acute when there are multiples of COI editors. If a COI editor, such as the subject of an article, wishes to point out that an error exists in an article, then certainly the talk page can be used to point that out. But we don't need any more incentives than currently exist for COI editors to influence the editorial direction of articles about them. Readers come to Wikipedia with the expectation that the articles are written and edited by disinterested parties. While that may be a naive expectation, it is one that I think Wikipedia needs to encourage rather than to discourage. Coretheapple (talk) 13:51, 14 June 2013 (UTC)
Treating PR pros as "fellow editors" is a problem that the proposal touches on. I have the advantage of knowing the circumstance of this proposal, which is that a legal antagonist made what appeared to be an outrageous and obviously COI suggestion, which turned out to be valid upon reviewing the available source material. CorporateM (Talk) 16:00, 14 June 2013 (UTC)
  • We do have to consider the case when paid editors are breaking the law by covert advertising. How can we possibly ignore this? If you want to ignore it, please don't let us have a rule that says others have to ignore it also. Smallbones(smalltalk) 05:42, 15 June 2013 (UTC)
  • Well let's take a look at an example of how this suggestion would have affected my response at the Chevron Corporation article. See Section 6.4- Environmental Damage in Ecuador Edit Request where I say: Considering that Chevron stated the court finding was, "the product of fraud (and) contrary to the legitimate scientific evidence" it would be expected that the company rep here on WP would propose an edit that represented their desired way of seeing the entire disaster and the resulting lawsuits. And in my next edit I say: I find the rewrite a brazen attempt to bias our readers to the Chevron viewpoint rather than an unbiased telling of this unfolding incident. [6]. Should I really have been expected to pretend that this suggested edit was written by an editor with a neutral point of view? Gandydancer (talk) 11:53, 15 June 2013 (UTC)
Just a brief point here - You are not supposed to pretend that it was written by an editor with a neutral point of view. All that this proposal implies is that You take that edit request purely by its own merits, ie "You take the edit request as if you would have taken it if (say) I had suggested it". If the edit is not neutral, it is not. There does not need to be any other reason to reject it. That was indeed the reason I added this proposal. I wonder if there could be a change in wording which would make it clearer. TheOriginalSoni (talk) 13:12, 15 June 2013 (UTC)
I know very well what you meant. Gandydancer (talk) 14:03, 15 June 2013 (UTC)
The difficulty is, it's impossible to assess whether the editor is trying to be neutral. If Gandydancer was correct in his assessment, then he is speaking his mind, pointing out the obvious and defending the Wiki from corporate influence. If he was incorrect, then he is badgering a PR person that attempted to do the right thing, even if perhaps unsuccessfully.
I think the guide could be improved with a lot more basic, down-to-earth advice. For example, I would like us to explicitly tell PR people that it is not helpful to propose re-writes of material that is already neutral and properly sourced. Too many Request Edits are "can we write it this way?" (usually in a manner that is more aligned with corporate messaging) and it's disrespectful, though perhaps unintentionally. CorporateM (Talk) 04:21, 16 June 2013 (UTC)
@TheOriginalSoni: I think that CorporateM is making a valid point about it not being possible for editors to always judge whether an edit (whether directly in the article or a talk page suggestion) is neutral. A COI editor may be cherry-picking facts to support his point of view, which a non-expert editor would have difficulty ascertaining without spending a large amount of time vetting the COI editor's suggestions or edits. This is not a hypothetical or theoretical problem, as that kind of behavior has been a serious problem in one article I monitor, causing a number of problems: conflict among editors, large amounts of time having to be consumed vetting edits, and a generally unhappy editing environment. That problem would only be exacerbated by treating all edits by all editors as neutral and in good faith once the COI has been pointed out. On the contrary, once the COI has been pointed out, it is incumbent for editors not to treat those edits as neutral, but to give them the highest degree of scrutiny, because by definition the COI "single purpose" editor/paid pr person has the interests of his or her client paramount, not the advancement of Wikipedia. Coretheapple (talk) 15:52, 17 June 2013 (UTC)
I think what you mean is that it is difficult to assess whether something is "an improvement," because we do not require that any editor be neutral, only that they be neutral enough to make the article "better." My approach is as follows: If the article has significant problems, I will fix them regardless of the quality of the PR engagement. If the article is basically fine how it is and the value from the PR contributor is marginal or debateable, I will do what any journalist would do, and basically blow them off. CorporateM (Talk) 20:53, 17 June 2013 (UTC)
Oppose. Why have a WP:COI policy at all in this case? It's hard enough to do anything about editors who blare their POVs all over talk pages and then consistently get away with violating policies to push their POVs (perhaps because no one cares any more among users and admins?) and thus are allowed to get away with it. Sigh... CarolMooreDC - talkie talkie🗽 16:46, 17 June 2013 (UTC)
I don't think I can support this, although I'm sympathetic to the cause of wanting to encourage disclosure. I do think we can encourage disclosure in other ways, similar to how we approach this already when we say "Some editors declare an interest in a particular topic area. The benefits of this are that most editors will appreciate your honesty and may try to help you...." Disclosure evidences transparency; transparency builds trust; trust helps you get (and *should* help you get) a fairer hearing. It's a simple as that. I would not object to some text reminding editors, when assessing a situation in which a conflict of interest is in play, to see disclosure in this light, while not turning that reminder into an easily-gamed formalism. --j⚛e deckertalk 22:52, 17 June 2013 (UTC)
Good editors will disclose and be careful. Even with this policy bad editors won't disclose and will do what they want. So at least we can encourage the 70 percent (or whatever number) who are basically honorable to disclose such paid activities. CarolMooreDC - talkie talkie🗽 23:19, 17 June 2013 (UTC)
Or editors will disclose and not be careful and not abide by COI rules. Unfortunately, the rules are so toothless that they are as good as nonexistent as it is. Coretheapple (talk) 15:09, 19 June 2013 (UTC)
If only Wikifoundation could hire professional admins for various ANI roles and pay them well enough to put up with the harassment from the jerks of all stripes who need frequent blocking to keep them in line. It's the biggest thing that makes people quit. (If I didn't have such a thick skin I would have quit years ago.) I'm all for lots of 24 hour blocks til people get the point. It would have helped me get the point earlier when I was "young and wreckless" on wikipedia :-) CarolMooreDC - talkie talkie🗽 18:37, 19 June 2013 (UTC)
Personally I think the most egregious situations are the ones in which the COI editors are not jerks, and are playing the wiki-game expertly. Coretheapple (talk) 21:46, 19 June 2013 (UTC)
But in general discussion is only place you can call someone out on it - very generally... :-) CarolMooreDC - talkie talkie🗽 22:50, 19 June 2013 (UTC)

COI submissions page similar to AfC

I was up until about 3 a.m. whipping up an AfC-like submission page for COIs to request corrections, contest unsourced material and (after reading some disclosures) offer content for consideration. It comes to mind that I've seen posts several years old where editors have pondered why this doesn't already exist and it seems like a no-brainer.

It needs some coding work before the forms would actually "work" but I would be interested in (a) anyone who can help code the forms and (b) any thoughts generally. CorporateM (Talk) 16:31, 1 July 2013 (UTC)

We do not need an elaborate mechanism to encourage COI contributions. If an article contains an inaccuracy, the COI editor can come to the talk page and say so. If there is a genuine problem, it will be fixed. If the editors do not agree that there is a problem, it will be contested or, more likely, ignored. Coretheapple (talk) 19:39, 1 July 2013 (UTC)
T13 might be able to help setting up the form and other such technical issues. TheOriginalSoni (talk) 13:31, 5 July 2013 (UTC)
This might be useful but it currently duplicates WP:PAIDHELP. I'd agree that a more neutral wikiproject location would be ideal. In any case, should you go through with a new noticeboard, PAIDHELP should be merged. You might notify them and ask for feedback. Cheers, Ocaasi t | c 15:58, 5 July 2013 (UTC)
I was thinking PAIDHELP is already a duplicate of COIN, but my conversations with other editors have shown that there appears to be relatively broad support for a noticeboard similar to BLPN. Request Edit is slightly different, in that it involves issues that only require a single editor's attention, while a board is usually to advertise something that requires additional help from multiple editors. Anyway, it should probably be moved to an actual board (and maybe renamed?) at the least. It could be implemented as a tab of the COI submission page to keep everything in one place. Though the COI submission page is not specifically for "paid editors". CorporateM (Talk) 16:35, 5 July 2013 (UTC)

Hi! As a follow-up to Wikipedia:ANI#User:Usmanwardag, I'm wondering if we should form a stance in regard to paid !votes. The AfD just raised at ANI had two, and possibly at least four, accounts !vote to keep the article, another currently open AfD has had two people !vote to keep, both of whom appear to have been hired (one certainly was, a second looks likely), and going back a bit, I can recall one case this year where an editor was paid to have an article taken to AfD, and three additional editors were then hired to !vote delete. The discussion at ANI seemed to come down more to disruption than paid !voting, so it may be that the current stance should still be 'strongly discouraged, but not specifically banned", but given that there have been two cases in the last two weeks, this might be worth visiting.

Just to make things a tad harder, people aren't generally hired just to !vote, but are instead hired to fix the article and save it. This can result in article edits, but sometimes just comes down to COI !votes at the AfD discussion. It seems that being paid to fix the article may be a different concern to being paid to express an opinion in the AfD. There's also probably a distinction to be drawn between people paid to create an article voting in an AfD, and people paid specifically to vote. - Bilby (talk) 03:34, 17 July 2013 (UTC)

It's unfortunate that we have to spell this out, but I was surprised to find comments at ANI of the type "We don't actually have a rule against this." We could, maybe even should, allow *comments* by paid editors at AfD's (and at similar discussions and noticeboards) as long as the paid editor identifies himself as paid, but !voting is totally out. It is essentially meat puppeting. Similar discussions would obviously include nominations for good articles, featured articles, really any discussion where there are !votes or similar. The simple way to express this is something like
"Other than comments where the paid editor self-identifies as being paid, paid editors are prohibited from !voting, promoting, or discussing the relevant article on pages other than the article talk page or the paid editor's own user pages."
Trying to make fine distinctions between AfD's and other types of pages, !voting vs. nominating or other types of promotion would simply invite gaming. If a paid editor wants to avoid breaking the proposed rule, all they have to do is start their comments with "Comment (I have been paid to edit this article)". Simple. Smallbones(smalltalk) 13:22, 17 July 2013 (UTC)
^ Smallbones' comments is basically what I recommend to PR people. If you have something to say, disclose you're a PR rep and post a comment, but don't vote. I would think a stronger word than "discouraged" would be appropriate.
This is an ethical grey area in PR. It's well known that it's unethical to pay for votes in an election. OTOH, when it comes to things like awards that are won by a popular vote, many companies will encourage their employees, customers and partners to vote.
That's pretty shady if the company wins the award because a bunch of its employees came out to vote, many of whom are not even users of the product. The problem is that this is the most effective way to win the award and there are no repercussions for doing it, so PR reps are therefor obligated to serve their employer's best interest. The best way to deter behavior you don't want here is to show PR reps that it is not in their employer's best interest, so they can advise their stakeholders accordingly. CorporateM (Talk) 14:13, 17 July 2013 (UTC)
BTW the wording I suggested above is just a suggestion. If folks have other suggestions or tweaks, please include them here for discussion. The only things I ask is that the word "prohibited" be included and that the new rule be simple (so as to be less open to gaming). Smallbones(smalltalk) 14:36, 17 July 2013 (UTC)
A couple quick comments from a PR rep's perspective:
  • "are prohibited from...promoting... the relevant article..." <-- the "promoting" of the article itself (as oppose to promotion in the article) seems acceptable to me in at least some circumstances and open to sweeping interpretations.
  • sigh, Wikipedians are stuck on this term "paid editors" but when I've used the term "paid editor" in front of a PR person, they ask "what's that? like if we pay someone?" Because they are not paid to edit Wikipedia - they are just general employees. It's like working in a Verizon store and 1% of the time you do repairs on a cell phone, then people start calling you a "paid repairman" You wouldn't know what they were talking about. CorporateM (Talk) 15:09, 17 July 2013 (UTC)

Request for advice

I support a non-profit research organization, RTI International, that regularly publishes academic studies in peer-reviewed journals that could be very useful for Wikipedia. As such, I have a self-citation COI. (See example here). My question is, should I still use Request Edit, disclose, etc.? I'm not sure I really have a COI, (or at least a very significant one) because the fit with Wikipedia's objectives feels much more natural in this case. CorporateM (Talk) 18:06, 22 July 2013 (UTC)

Since editors with much more knowledge about Wikipedia COI policy than I have not responded, I'll give a common sense answer. If you are not receiving a salary from the organization I see no reason that you should have to report a COI. Gandydancer (talk) 19:37, 29 July 2013 (UTC)

Circular wording in Template:Uw-coi

The very first paragraph says "if you are affiliated with some of the people, places or things you have written about on Wikipedia, you may have a conflict of interest or close connection to the subject", which seems like an empty, circular explanation ("if you are affiliated with FooCorp, then you may have a close connection to FooCorp"). I mentioned it last January and a couple of people have made edits to fix the wording since, but the template seems to get dialled back to an ancient earlier version every few months. Does anyone have any objection to just dropping the words "or close connection to the subject" here? --McGeddon (talk) 16:42, 29 July 2013 (UTC)

Well, I'm not so sure. I'll admit it's mealy-mouthed. However, I suppose that one's affiliation could be so remote that it would not be a conflict of interest. I agree that it can be improved. Coretheapple (talk) 17:31, 29 July 2013 (UTC)

Science Genius

Emdin is a partner with Rapper, GZA from the Wu Tang Clan and website Rap Genius of the Science Genius B.A.T.T.L.E.S. initiative to utilize hip-hop music and other aspects of rap to engage youth in New York Public Schools In Science.http://www.npr.org/blogs/codeswitch/2013/08/08/207348197/science-rap-b-a-t-t-l-e-s-bring-hip-hop-into-the-classroom — Preceding unsigned comment added by 75.150.65.29 (talk) 18:52, 16 August 2013 (UTC)

declaration section

I don't like the advisory writing tailored towards corporate communications writer and the tone that sympathize with them. Also, suggesting following the trade organizations for PR editors is sanctioning the values held by its industry organization onto wiki editors. Cantaloupe2 (talk) 16:39, 28 August 2013 (UTC)

Code of conduct

Regarding the part on professional ethics codes:

"and professional public relations firms"
To the best of my knowledge, individuals join the PRSA, not organizations like PR firms and those individuals do not necessarily work for a firm, rather than a corporation. The word "professional" could also be trimmed.

"may be required"
"Required" is a strong word. Members "pledge", but it's mostly an educational document. IMO, I am not convinced the code has more than a marginal impact on behavior.

The PRSA (the most well-known ethics code) and GA (the most global) is one way to go. It could also be written in a way to just make it general. CorporateM (Talk) 05:55, 29 August 2013 (UTC) (PR Guy)

In the scope of Wikipedia editors, Conflict of Interest refers to dissenting goals between that of the encyclopedia and the interest of the editors or the party editors is representing. Inducing systemic shift to any article or a group of articles to advance interest is a foul play here in Wikipedia. Practicing by an industry ethical standard does not mean interest in common with Wikipedia. There are some who add biased view inadvertently as a fan, or unequal experience on all sides.
Here's a hypothetical example. An electric industry/alternative energy industry sponsored Editathon that drive editors to elaborate on and add credible references to favorable contents about electric appliances and vehicles and constantly adding comparison to technologies the industry wants to differentiate against only in favorable aspects but remaining deliberately silent, or drown it out in other contents. Such manipulation can be highly calculated and subtle, such as including suggested list of good sources that may be WP:RS but designed to instill systemic bias they're seeking all in the name of raising "awareness". Some are more obvious by the ways of correlation in commonalities such as sources used, user name, editing articles that are fairly closely related to the interest of cited sources, etc.
I think linking to PRSA guidelines on this policy page is an endorsement of the viewpoint of industry advocate which is inconsistent with the non adversarial position we are supposed to follow. Not every conflicting interest is about public relations or anything that has to do with it. I find instilling adoption of one industry group's suggestions unacceptable and I feel that removal is in order from this page. Cantaloupe2 (talk) 06:42, 29 August 2013 (UTC)

A slight change

Hi all,
I undid this edit, because I felt that:

  • "Consideration" is less clear to readers than "Money";
  • Removing "... is a subset of paid editing" might remove valuable context, and certainly breaks the punctuation in what remains of the sentence;
  • "an instructor asking you to write up its warts-and-all history" is problematic, because instructors are usually human and consequently get gendered pronouns, or "their", instead of "its".

Nonetheless, it's probably possible to improve the wording around there... bobrayner (talk) 17:50, 28 August 2013 (UTC)

There's a reason I changed it money to "for consideration". When you use the term money, it conveys currency is the only form of payment. Consideration is a broader term which describes something of value, which can be status or in-kind items.
As for the its/theirs nit picky grammar thing, feel free to fix something like that.
Is a subset of paid editing.. as opposed to just paid editing? Cantaloupe2 (talk) 01:15, 29 August 2013 (UTC)
Paid advocacy is a subset of paid editing and is treated differently. There seems to be consensus that some forms of paid editing are okay, but that paid advocacy isn't. SlimVirgin (talk) 22:23, 29 August 2013 (UTC)

Fitness of Further Readings material

HERE, I removed these materials. The contents are again, tailored to business of "professional ethics" on the matter of Public Relations producers. These are not the principles followed at Wikipedia and I find these links inappropriate. Cantaloupe2 (talk) 08:20, 29 August 2013 (UTC)

I would not remove the whole section. If there's a specific link or links you find objectionable, raise that here. Gigs (talk) 17:14, 29 August 2013 (UTC)

Neutrality applies here too

As I read through here, I feel that there's some partisan prose here. There are some rhetorical writing that are tailored to the interest of public relations editors; as well emotive expressions that reflects personal. I will be going over and copy editing over time towards neutrality. Cantaloupe2 (talk) 01:57, 29 August 2013 (UTC)

I've reverted C2's last 2 edits, even though I recognize the motivation as being very good. Wording has always been very tricky on this and it probably shows in some compromises made that may not exactly look right on the surface. I too think that there are some subtle things that seem tailor made for PR folks to split hairs on. In short, I'll call out the usual folks I trust on this and we can go over the tricky wording in detail. Smallbones(smalltalk) 02:54, 29 August 2013 (UTC)
What do you find disagreeable about my edits? The break down of the edits thus far is given in the talk discussion right above this one. You didn't really give me specific examples or rational for why you did what you did. Cantaloupe2 (talk) 04:38, 29 August 2013 (UTC)

POV explanatioon given for avoiding COI edits

Currently, it reads "COI editing is strongly discouraged. It risks causing public embarrassment to the individuals and groups being promoted (see Wikipedia is in the real world), and if it causes disruption to the encyclopedia, accounts may be blocked." . This explanation emphasizes the importance of looking out for the SUBJECTS, which obviously is of importance to Corporate Communication/Public Relations editors. The reasoning should be stated in a non-partisan way and presented in a way that reflects the purpose of Wikipedia and not tailor to a specific focus group. Let's refrain from editorial reflection. There are credible, notable news stories about PR editing. Again, we should just report and summarize what happened and not interpret the consequences in a way that instill partisan expression of concerns for specific issues. Cantaloupe2 (talk) 04:54, 29 August 2013 (UTC)

Look, while not a core policy, this is a significant and substantial one. Even if you're of the opinion that one should be able to just make bold edits to such policies, you should not be surprised when they are reverted when they so not have a significant consensus behind them. supreme executive power derives from a mandate from the masses, not some farcical WP:BRD ceremony.
in all seriousness, while i recognize that your intention might be good, it needs a whole lot more discussion and consensus behind it before you change a core freaking policy again. -- UseTheCommandLine ~/talk ]# ▄ 04:59, 29 August 2013 (UTC)
From the version of Wikipedia page current as of the time of this writing "Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy." Under majority rules process, it goes like "75% of people at the time of poll chose God created man, therefore it is generally accepted that we were made by God". It doesn't consider premise presented by opposing parties. A PR firm with vested interest can easily send a crew to outnumber a perspective they oppose to gravitate the perspective in their direction. In consensus building, the challenging parties have to present their premises and provide valid counter argument on points of contentions beyond ad populum argument that only reason it as there are more people on their side. You argument is that this is substantial and significant. Now, where's the supporting argument? Cantaloupe2 (talk) 05:32, 29 August 2013 (UTC)
Expanding on this, here is a direct example of the very concern I have voiced above.

Hi Cantaloupe, your edits to the guideline have been reverted by six editors, which means your changes don't have consensus. The way forward is to start to argue your case on the talk page, rather than continuing to revert. Please see WP:BRD, and more importantly WP:3RR. Many thanks, SlimVirgin (talk) 01:39, 30 August 2013 (UTC)

The assertion is "your changes don't have consensus" or that my edit "waters it down too much" for the premise "because six editors reverted your edits". WP:CONLIMITED reads quality of argument in reaching consensus. The reverting editors made assertions of personal opinion with no substantial supporting premises. No explanation of what and how it watered down to support that claim. Reiterating my concern, "A PR firm with vested interest can easily send a crew to outnumber a perspective they oppose to gravitate the perspective in their direction." If this was to be a consensus process, a group of industry advocates would have a way to induce systemic shift in POV using "I don't like it" "its not better" and such without producing any argument and say "xx number vs you and lack of opponent participation in discussion therefore no consensus" to resist page change unfavorable to their industry.
I explained in no uncertain way why POV of PRSA shouldn't be endorsed. No real argument in favor was put forth why it should remain, and the only counter claim given was that "unexplained edits by 6 usernames vs you, therefore consensus not reached" which depends on ad populum tactic as a means of evading explaining their position.
Cantaloupe2 (talk) 02:42, 30 August 2013 (UTC)
My revert was again reverted. CommandLine provided "this is really not the way to convince the community)" in the summary. What does this user mean by "convince"? Where are the opposing arugments? More usernames said "nay" is no consensus. Cantaloupe2 (talk) 05:36, 29 August 2013 (UTC)
Can I ask you to please make your argument here, and then we will draft some changes in language? I understand your point about consideration rather than payment, i think. i also think it would be good if we had examples or references for this same type of language being used in institutional CoI policies. I am most familiar with medical institutions' CoI policies, and I can't remember seeing such language, but i am eager to see some. -- UseTheCommandLine ~/talk ]# ▄ 05:44, 29 August 2013 (UTC)


In the talk discussion, I provided a link to the word consideration and I find the explanation in prose was sufficient to be clear. More obviously clear like "money" doesn't mean its accurate. Money given in exchange is a consideration, but not all consideration is monetary. As an example, a high school teacher, who has a family member who owns a business gives extra credit assignment to his class to edit the articles of relevance in a favorable direction for the family member's business. The students are graded at teacher's arbitrary discretion. The "consideration" doesn't have to be of value. see this. The problem is the consideration element, which disrupts neutrality. A more open ended task with a different consideration such as a chemistry teacher asking a student to improve an article on an element of their choice from the periodic table for grade is not an issue, because it doesn't the outcome doesn't disrupt the encyclopedia in a partisan manner. The point of reference in our scope of discussion is the interest of our project(improving encyclopedia) and contents presented in impartial manner. Interest group has vested interest in shifting contents portrayed in a certain way, which often is exaggerating achievements and downplay unflattering contents. The two interests are conflicting and priority as an editor is to protect the neutrality. Cantaloupe2 (talk) 06:11, 29 August 2013 (UTC)
(that "see this" is a dead link, for me anyway). I think i understand your point. I do wonder about undermining the existing definitions though. Defining things in such a way, (though I agree that that's really how things should be defined) would suggest to some people that this definition was meaningless. I can go and look someone up (so long as i'm not outing them) and see that they are employed by so and so. in the case of a doctor, i can go to propublica's dollars-for-docs website, or one of the websites being put up because of the new sunshine laws in the us, and that will detail in concrete terms how much money a physician is getting from a particular source. these are quantifiable numbers (which are more important to most people, i've found) and tie things directly and concretely to a particular influence.
switching to a consideration standard, though, is much more ad hoc, and would seem to be more prone to rumor and innuendo. this could conceivably undermine the weight of the CoI criticism, by diluting it.
All kinds of government officials go and take a management job with a regulatory agency for a few years, and then leave for private industry where they lobby for the industry they were just regulating. I think that's the sort of place your suggestions are coming from, please correct me if i'm wrong. but that's also very very hard to prove before or even after it happens. and this standard of evidence is really my primary concern with these changes. -- UseTheCommandLine ~/talk ]# ▄ 06:24, 29 August 2013 (UTC)


Among the issues that were raised by editors, such as "its" used in place of "theirs", I don't believe that anyone would object it if one just went and fixed it. The work we do here is constantly evolving and it does not have to be up to publication quality every single time. The link did not work, because of a spacing error. I added the source as a way of elaborating on "consensus". I fixed the link, so it should work now. The change in wording from "money" to "consideration" was merely for accuracy. The rewording was to address the clashing between the meaning of "payment" vs "money" as the money is not the only form of payment. Phrasing it as "consideration" removes the technical loop hole of declaring as a "not a paid editor" when compensation by money is used to define "paid".

In reviewing the general idea of Wikipedia, the greater concern is not that someone is paid but the type of editing that gets done. If the contents written has a consequence to the editor, it affects neutrality. If the portfolio of work done is evaluated by how well it reflects the client interest, remain without major reversion and not raise public outcry, its going down a different path than wikipedia mission. I would personally describe it as the game of how much BS one can get away with and smooth talk to gain their rapport. Cantaloupe2 (talk) 07:18, 29 August 2013 (UTC)

Most COI and ethical policies are not drafted in terms of consideration, but in terms of relationships. Consideration is a better word than money, but ultimately that still leaves a loophole where someone has a conflict of interest due to a relationship where consideration (i.e. something of tangible value) might not change hands. Ultimately a conflict of interest arises from the existence of two relationships with conflicting duties and obligations (fiduciary or otherwise).
For example, a real estate agent representing both the buyer and the seller of a property is a common conflict of interest, even though neither party is directly paying the agent ahead of time. The agent has two relationships, with conflicting duties (to get the best price for the buyer, and to get the best price for the seller as well)
Regarding BRD on guidelines and policies, it definitely still applies, and edits should not be reverted solely due to "lack of consensus for the change", but they still can be reverted by anyone who believes that the edits are contrary to wider consensus, or problematic in a way that requires discussion. Gigs (talk) 17:23, 29 August 2013 (UTC)
"farcical" was just my attempt to have a little humor by shoehorning the policy into a monty python quote. i think highly of it and think i've adhered to it; for the moment at least i still don't think enough discussion has been had to support the bold changes Cantaloupe2 has suggested. -- UseTheCommandLine ~/talk ]# ▄ 18:03, 29 August 2013 (UTC)
Gigs, the relationship is an important point. I believe that requiring solid argument would protect the integrity. Without such a requirement, multiple accounts representing partisan interest can gain control in numbers game. WP:CONSENSUS is clear on "just don't like it" is not a valid argument. Cantaloupe2 (talk) 07:40, 2 September 2013 (UTC)

Please comment on this edit

My argument is presented in the last few sections of the talk headers. The existing prose looks like its written to resonate with public relations writers and advantages/disadvantages are not written from the perspectives of impact on articles on Wikipedia, but rather comparing personal impact. Member policies for PR industry groups are for their members and not Wikipedia. These advisory comments do not appear appropriate for Wikipedia's COI guideline page. One editor contested the edit and commented that it "waters it down" too much. Please compare and comment if I omitted anything that was omitted in the scope of relevance for Wikipedia editing.

Existing contents:

Some editors declare an interest in a particular topic area. The benefits of this are that most editors will appreciate your honesty and may try to help you; you lay the basis for requesting help from others to post material for you, or to review material you wish to post yourself, and professional public relations firms may be required to abide by code of ethics, such as the GA code of ethics or PRSA code of ethics. The disadvantage of declaring your interest is that people outside Wikipedia, such as reporters, may identify you and generate negative publicity for you, your group or your company. Some COI declarations have the effect of announcing your real name (see WP:REALNAME). Do not publicly declare an interest if this could put you at harm in the real world, e.g., from stalkers.

Rejected contents

Some editors declare an interest in a particular topic area or that they edit on behalf of another party. Disclosures creates more transparency about your editing. Conflict of interest editors are encouraged, but not required to disclose it. In some situations, declaring your interest can deduce enough clue to deduce your real life identity, so discretion should be exercised.

Cantaloupe2 (talk) 08:06, 29 August 2013 (UTC)

The new version looks fine to me. You raise valid points about the ethical code mostly being a personal pledge, and also being biased since it's kind of softball self-regulation, that we should not implicitly endorse. You need to fix your wording a little, "deduce enough clue to deduce" is awkward, as well "required to disclose it" should probably say something like "required to disclose their conflicting interest". Gigs (talk) 17:13, 29 August 2013 (UTC)
Thanks for your comment. I don't really focus my effort on punctuation/grammar as I try to focus on contents. Right now I am trying to get consensus over contents. Grammar/technical error was one of the reasons cited for reversions to some of my work, but it appeared to me that some are just finding a reason to revert something that don't agree with their interest to avoid having to engage in a real debate. As I was looking around, i came across this ESSAY and this really concerns me. I know civility is an important aspect in HOW we edit here, but impartial, well balanced presentation are the core values on WHAT is presented. Cantaloupe2 (talk) 00:52, 30 August 2013 (UTC)
Hi Cantaloupe, I reverted your edit because it's not clearly an improvement. What is your objection to the current version? SlimVirgin (talk) 01:08, 30 August 2013 (UTC)
I provided a thorough explanation in the paragraph at the top of this section. You provide an assertion that "not clearly an improvement" but it ended there. Please provide supporting evidence. So far, this is not any more convincing than "green is a better color", just because. Cantaloupe2 (talk) 01:46, 30 August 2013 (UTC)
Cantaloupe, grammar and punctuation matter a lot, when you are editing a guideline. I think we could make a more limited edit to remove the mention of advertiser ethical codes just by striking the word "and" and everything following it in that sentence. Failing that, it's really not very correct to say that "firms" are bound by an ethical guideline, we should at least change that to "pr professionals". Gigs (talk) 16:51, 30 August 2013 (UTC)


True that they do, but grammar/punctuation is something that can be uncontroversially corrected and misspelling/punctuation/grammar are not not justifiable basis to "I don't like, therefore I revert". If grammar/punctuation was the only premise to support their assertion, then it would appear addressed once it is fixed.

1.) Wikipedia is an non-adversarial place and I believe we're in an agreement that language is partisan.

2.) "PR Professionals" may not be definitive on how "professionals" is determined. Even if the article was focused to practioners of the art of public relations, coverage a private trade group value that is a minority interest is questionable. Percentages of PR practioners who belong to any trade group like PRSA was reported to be 10%in 1995 as quoted by Fitzpatrck, a professor in the field.

3.) It appears that there's enough controversy even to within PR community that endorsing PRSA, but trade organizations are described as having ethics, transparency and disclosure in common. Neutrality is not part of this. So, this is where endorsing instill interest clash.

  • Verčič, Dejan, K Sriramesh (2003). The Global Public Relations Handbook: Theory, Research, and Practice. Taylor & Francis. ISBN 9780805839227.{{cite book}}: CS1 maint: multiple names: authors list (link)
  • Gudykunst, William B. Communication Yearbook 26. Psychology press. p. 227. ISBN 9781410606662.
  • Philip M. Seib, Kathy Fitzpatrick. Public relations ethics. p. 24. ISBN 9780155019430.

Cantaloupe2 (talk) 18:38, 30 August 2013 (UTC)

Hi Cantaloupe, I'm still not sure what your concerns are exactly, but I agree that the current text is a bit wordy, so how about we cut it down to:

Some editors declare an interest in a particular topic area. Public-relations firms, for example, may be required to abide by a professional code, such as the GA code of ethics or PRSA code of ethics. The benefit of declaring your COI is that most editors will appreciate your honesty and may try to help. The disadvantage is that, if the declaration identifies you, it may generate negative publicity for you, your group or your company.

SlimVirgin (talk) 20:13, 30 August 2013 (UTC)
I explained my rationale why. The new suggestion does everything but address my concern. I find it the fundamental issue of tailored writing and undue credibility towards private interest group Cantaloupe2 (talk) 23:33, 30 August 2013 (UTC)
I changed it from "firms" to "professionals" since it is a valid point that the members of these organizations (who are bound to uphold such codes) are not firms, but individuals. Gigs (talk) 20:03, 5 September 2013 (UTC)
Instead of "required", something along the lines of "may violate" is probably a more accurate representation. Members "pledge" but there is no enforcement really for "required" to be realistic. Also, most PR reps are familiar with the ethics code, but few are actually members that have made the pledge, so the relevance is beyond those that have obligated themselves to follow it. CorporateM (Talk) 21:04, 5 September 2013 (UTC)

method for IP COI and risk of outing

I have just come across editing on an organization by an editor who works for that organization, but the editing was from an IP address that has a history of editing that article. I found out about who it was by following an external link, so following up an apparent COI would seem to risk outing. It's possible the editor has a Wikipedia username that they just didn't use to edit the article in question. This raises three questions, relevant both to the case at hand and to possibly clarifying the COI guideline:

  • Procedurally, do we proceed with an IP COI just as we would with a non-IP COI?
  • If so, and if we should present evidence, how do we prevent outing (other than by the editor themself)?
  • If the editor does have an English Wikipedia username but never used it to edit this article (someone with the same username has edited one other article on en-WP), how does that affect COI procedures, since perhaps the editor is trying to avoid being outed?

Nick Levinson (talk) 15:37, 10 September 2013 (UTC)

If there is a risk of outing, see Wikipedia:Blocking policy#Confidential evidence for help. davidwr/(talk)/(contribs) 17:36, 10 September 2013 (UTC)
I would not be overly concerned regarding our outing guidelines when it comes to single purpose accounts editing with a likely COI. Those guidelines were written to prevent harassment of established and good faith editors on and off-wiki, and it's an overly conservative interpretation of them, in my opinion, to assume that they bind us from publicly investigating spammers and single-purpose COI editors. A common example of this come with every COI investigation that involves an autobiography, most of which we do publicly investigate. Obviously such things should be treated with extreme care and accusations and posting of names should not be made recklessly.
All that said, how could you have conclusively tracked an IP to a real name? It would risk libel for you to publish such a claim if you weren't absolutely sure. It would be best to just publish the links you have discovered between the IP and the COI subject matter, and leave personal names out of it, if possible. Gigs (talk) 18:06, 10 September 2013 (UTC)


I have a different take on this. Nick, COI editing is not against policy. But IMHO what you are doing is getting close to crossing the line on a very important policy. North8000 (talk) 18:08, 10 September 2013 (UTC)

It sounds like you may know more details of the situation than I do, and that very well may be the case. Gigs (talk) 21:34, 10 September 2013 (UTC)
No I don't Here's what I meant:
  • The policy rightly defines a COI as "When advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest." So. unless their investigation has opened up the editor's skull to see if this "more important" exists, they don't have sufficient basis for accusing of a COI.
  • IMO "Investigating" an individual is a case of the the cure being worse than the disease.
  • Being just a member of an organization is not even remotely close to establishing a COI. Shall we investigate whether there are any persons of that party editing the Democrat or Republican Party articles?
  • Getting dangerously close to wp:outing or partial outing.
Sincerely, North8000 (talk) 21:45, 10 September 2013 (UTC)
It's hard to comment on this case without knowing the sorts of edits involved. Ultimately it's the nature of the edits that justifies the investigation, not the potential COI in and of itself. As you point out, we don't go on witchhunts when there isn't problematic editing. Gigs (talk) 17:40, 13 September 2013 (UTC)
Sorry that I missed these posts earlier.
I don't have a personal name for the person in question but I do have a username tied to the IP, but that does not eliminate that more than one person may have used the IP, which has appeared multiple times in editing the article, nor does it prove that no spoofing occurred spoiling the username linkage, thus those questions can be presented for investigation, but there is a prima facie case for a COI. If I say publicly how I linked the username to the IP, it would be very easy for anyone to out the person with what I would have said about how I know. I don't object to submitting the evidence in a way that avoids outing. There has been problematic editing. Whether a COI has infected the article's content does not require examining an editor's personal motives but, given our policies and guidelines, can be based on what is known about their motives plus what is in the article, what has been deleted from it by that IP, and what is available in an off-Wiki search for sourcing, including with respect to NPOV. The IP is not a single-purpose account and there may not be a COI for that IP at any other article; but, for this article, the user wrote that they "work for" the organization the particular article is about, thus, apparently, is not "just a member", and the organization is a small one (my guess is no more than a few hundred people), not nearly as large as the Republican party (I think the Democratic party is even larger). I'd post a notice on the IP's talk page, the article's talk page, or the article about the likely COI but for the outing risk.
The responses by User:North8000 seem to object to any COI for an IP unless the editor is self-exposing. I don't read the COI guideline that restrictively, and it would almost certainly expose the Wikimedia Foundation to legal liability if an article is a disguised advertisement and uses someone's likeness so as to violate the law. So I assume the Foundation wants IP COIs to be prevented or repaired much as with non-IP COIs, quite apart from editorial quality issues (including NPOV).
I therefore mainly need to know the procedural differences between IP and non-IP cases. For example, instead of posting a notice to the IP talk page, should I go first to the Checkuser procedure (which allows confidentiality) and state how, with high probability, I linked a username to the IP and then, if Checkuser produces a positive result, post about the apparent COI to the IP talk page?
Nick Levinson (talk) 17:45, 14 September 2013 (UTC) (Deleted problematic nonnegative case: 17:51, 14 September 2013 (UTC))
Nick, that's not what I meant. I meant that what you are implying (that merely being in the org flatly constitutes a COI) is in conflict the the definition of a COI. Sincerley, North8000 (talk) 21:18, 14 September 2013 (UTC)
North8000, you're technically right that it doesn't but it virtually does, under the guideline. Someone who works for the organization, if paid by it and that's commonly what is meant by working for an organization at least in the U.S., is "very strongly discouraged" from much editing of Wikipedia, and that would likely include some recent editing from the IP in question on the article in question. Nick Levinson (talk) 17:12, 15 September 2013 (UTC)
Nick the link that you provided discusses getting paid for editing Wikipedia not just working for/getting paid by the organization. HUGE difference with respect to risk of COI. Sincerely, North8000 (talk) 12:58, 16 September 2013 (UTC)
Clause 2 says "you expect to derive monetary or other benefits or considerations from editing Wikipedia (for example, by being an owner, officer, or other stakeholder of an organization; or by having some other form of close financial relationship with a topic you wish to write about)" and the next paragraph says "if you have a financial connection to a topic – including, but not limited to, as an employee, owner or other stakeholder – you are advised to refrain from editing articles directly". The notion that someone paid as a staff publicist responsible for influencing media but not specifically to edit Wikipedia who then edits Wikipedia on their employer's time should be considered as possibly violating their employer's rules on use of resources but is not in a COI relationship would be virtually unworkable regarding COI enforcement, would expose WMF to more legal risks, and, I think, runs counter to what seems to be the normal judgment among Wikipedians about who has a COI and what COI editors may do. I don't know what the editor in question means by working for the organization in question, but I think there's enough to raise the question about a specific COI. Nick Levinson (talk) 17:23, 16 September 2013 (UTC)

Making sure this is not a COI violation/issue

I have a personal friend who in his professional capacity has a financial interest in specific commercial publications (from a third party) but he is otherwise not an editor. He is well aware of COI issues on WP and knows that he should not edit on topics related to these publications. I have a personal interest in these publications only in that I want to buy and enjoy them and hope they do well, but otherwise have no profession or economic ties to them.

If my friend gives me a suggestion that one of these publications does not have an article, and then I proceed to research the topic on my own, finding the appropriate third party/independent sources that already exist to satisfy WP:V/WP:N for a standalone article without any other input from my friend, would this be considered a COI problem?

I really don't see this being an issue - as long as those sources exist and I'm not being fed them via my friend, it would be the same as if anyone else, unprodded, wrote the article. It's just happening sooner than later. But I do want to double check to make sure that this isn't stepping on any toes with regards to COI. --MASEM (t) 16:52, 11 October 2013 (UTC)

The issue is this: Would your creating the article or not creating the article affect your friendship in anything but the most trivial way? In short, would he thank you if you did it and/or wonder why you didn't if you didn't? The ethically safest course of action is to prepare a userspace draft then have a neutral editor review it or submit it as an articles for creation. Be sure to declare your COI to whoever reviews it, and don't make significant edits without discussion after it becomes an article. davidwr/(talk)/(contribs) 22:26, 11 October 2013 (UTC)

Discussion at WT:NOT

There's a discussion of a proposal I made at WP:NOT to incorporate the Bright Line Rule over there. Some folks say that is just a repeat of what's said here. Any comments appreciated. Smallbones(smalltalk) 02:19, 13 October 2013 (UTC)

WP:COI guideline seems contrary to the interests of WP. Make it an essay? Delete? Or can you defend it?

After seeing it referenced at WP:AN, I just read WP:COI for the first time, and I don't get it at all.

All edits are made by humans biased by something or another. COI is a potential source of bias, sure, but how is it different from the bias of a biologist editing a controversial topic in biology, a lawyer editing a controversial legal topic, or a runner editing about a controversial running issue, or any other bias? In all these situations, what matters is that the resulting content is compliant with NPOV, notability, etc., not WHO is making the edit, or WHY they are making the edit, or what biases may or may not be influencing them. What's so special about a COI bias that it deserves this kind of attention and scrutiny?

I'm considering proposing demoting this "guideline" to an essay, or maybe nominating it for deletion, but would like to first give others a chance to defend it. --B2C 21:51, 12 October 2013 (UTC)

The difference between COI and simple bias is that the COI bias is caused by an external relationship, not by something internal to the editor.
So if you're editing in the interests of strawberry muffins because you believe they're nutritious and delicious, it's possible that you'll be persuaded to change your mind, or soften your views, or take other views into account. But if you're here because you're being paid by a manufacturer of strawberry muffins, you're not allowed to change your mind if you want to keep that job. Or, more accurately, if you do change your mind, you're not allowed to reflect that in your edits.
It means that someone external to Wikipedia – an invisible hand – is in control of your edits, and therefore of Wikipedia's content. That's what's objectionable about COI editing, just as COI is objectionable in every other area of life. SlimVirgin (talk) 22:07, 12 October 2013 (UTC)
I think it should remain a guideline. In the examples you gave, if the biologist's career or reputation would be affected by the issue he is writing about, that's a conflict of interest under this guideline. The same goes with a lawyer writing about a topic in which he has a current case or which he reasonably expects to have cases on in the future or when his writing about it would otherwise impact his professional life. If the runner were writing about something that affected his running career, such as an article about a fellow runner he knew personally or about an official race sanctioning body in which he was a member or in whose races he ran, that would also be a conflict of interest. On the other hand, a biologist, lawyer, or runner contributing their expertise in areas that aren't closely tied to their profession, such as a runner creating articles about races or runners from before he was born, is likely not a conflict of interest under these guidelines even though, as you suggest, bias may be present. davidwr/(talk)/(contribs) 22:08, 12 October 2013 (UTC)
  • If COI conflicts with another guideline I would favour altering the other guideline. Who makes the edits matters, that's why we block editors. COI editors have an inherent difficult in editing neutrally, even when they think they are editing neutrally. The distinction between paid and unpaid COI is largely irrelevant. IRWolfie- (talk) 22:18, 12 October 2013 (UTC)
    • We block editors who demonstrate repeatedly problematic behavior when other attempts to rectify don't work. Yes, in that case who makes the edit matters, but that's an exception for very good demonstrable reason. Prohibiting editors with a COI from editing is not that. --B2C 23:26, 12 October 2013 (UTC)
  • (edit conflict) At first blush I thought this was a shocking suggestion. But I am not sure. If it were eliminated entirely, it would remove any doubt that about Wikipedia's utter absence of integrity and independence. A disclaimer could be put at the top of every article, warning readers that Wikipedia makes no claims about the completeness, accuracy or objectivity of the article, and that the subject of the article, or its/his/her paid representatives, may have created, instigated or shaped the development of the article. That's an honest approach, and it most closely approximates the reality of Wikipedia. Coretheapple (talk) 22:20, 12 October 2013 (UTC)
    Every page has Disclaimers at the bottom. davidwr/(talk)/(contribs) 22:29, 12 October 2013 (UTC)
    Most readers, myself included, have never noticed that. I'm talking about a visible disclaimer, one that every reader can see, prominently displayed. Coretheapple (talk) 22:31, 12 October 2013 (UTC)
  • WP:COI is a legal protecton against the Wikimedia Foundation getting successfully sued if an article turns out to be a disguised advertisement posted by some editor and someone's likeness was commercially used without the latter person's consent. Such a lawsuit has happened regarding another publisher. I don't think COI applies merely to a member of a profession; e.g., a lawyer working as one can write about law, but not about the firm that employs that lawyer. Nick Levinson (talk) 22:40, 12 October 2013 (UTC)
  • Our policies are not fig leaves for WMF. If they need legal protection, they can say so transparently and set up whatever rules they need. This is not our job. Jehochman Talk 00:39, 13 October 2013 (UTC)
    • Yes, it is our job to include legal protection. Everyone needs legal protection; the Foundation doesn't have to say so. Policies and guidelines are part of that protection, even while they fulfill other functions. Nick Levinson (talk) 18:41, 13 October 2013 (UTC)
  • After having worked on this guideline for a number of year, spending time at WP:COIN long ago, and then thinking about it for half a decade since, I find this guideline to be problematic. I think it causes more problems than it solves. We should simply enforce our content policies without reference to the user's real life identity. Our content policies, especially WP:NPOV, are relevant to cases of problematic COI editing. As for non-problematic COI editing (e.g. a company corrects trivial errors on their Wikipedia page, such as spelling and grammar because that might make them look bad), if editing is no problematic, why should we care? Jehochman Talk 00:39, 13 October 2013 (UTC)
  • Noncontroversial editing is sometimes allowed now by WP:COI where the COI is nonpaid. In paid cases, good arguments lie both ways, one because institutions often need to lean over backwards to avoid coming so close to the line between good and bad that people cross over. Nick Levinson (talk) 18:41, 13 October 2013 (UTC) (Indented & releveled as list item: 18:53, 13 October 2013 (UTC))
  • You would presumably agree that a judge should recuse if her husband is the defendant, or that a journalist should avoid praising a business if he owns part of it. The same principle applies here. We can't judge whether articles are neutral a lot of the time, because it requires background knowledge. We therefore ask that editors have no COI so that they're approaching the issues honestly, rather than under pressure from an external source.

    That editors have internal biases is unavoidable, but we can at least ask that they have no external relationships that control or undermine their relationship with Wikipedia. SlimVirgin (talk) 00:48, 13 October 2013 (UTC)

  • The problem is that Wikipedia editors are anonymous. There's no way to know who anybody is, unless that person declares their identity. We don't allow anonymous people to be judges or journalists. I think it is a lot simpler to evalate the edits without regard to the editor's identity. If the edits are good, we accept them. If they are bad, we don't. The COI guideline is good when it warns COI editors what problems they face and how to avoid them. The COI guideline is bad when other editors wield it as a club. Jehochman Talk 00:55, 13 October 2013 (UTC)
  • For all but the simplest of articles there's no way to judge whether an article accurately reflects the majority and significant-minority views reported in the secondary literature, unless you're willing to spend weeks educating yourself. Most of the time we have to trust, despite the anonymity, which is why the idea of AGF is so important on Wikipedia. But it's impossible to extend trust when an editor is employed by the company he's writing WP articles on behalf of. It's a very simple bright line. If you could reasonably be said to have a COI, edit something else instead, please, or stick to the talk page. SlimVirgin (talk) 01:12, 13 October 2013 (UTC)
  • Do you support the animal rights movement? Do you recuse from editing any articles on topics that are important to you? I think editors who are employed as PR professionals need to disclose their involvement. Once involvement is disclosed, editors can scrutinize the contributions to make sure they are good. The question to ask is whether PR professionals can make our articles better. In many cases, the answer is yes. Jehochman Talk 01:18, 13 October 2013 (UTC)
  • Having opinions about something doesn't in itself give rise to a conflict of interest. A COI is created by external relationships: being employed by someone, having an involved family member, volunteering for a campaign, etc. The point you make about scrutiny ignores the point I made above, namely that a lot of the time it's impossible to judge whether the edits are neutral and accurate, because the issue is so complex. We had that problem exactly at BP. There was a BP person suggesting text and sources, and you had to be really knowledgeable about the issues to spot when things were being minimized or omitted. It can take months or even years for the average Wikipedian to get up to speed, so the "scrutiny" argument just doesn't scale. SlimVirgin (talk) 01:27, 13 October 2013 (UTC)
  • If you think it's that simple, by all means go to Chevron, BP, or any of the big oil or pharmaceutical companies (and their products), then come back here – after you've read all the sources and the sources people have omitted – and let us know whether the articles are accurate and neutral. I'll see you in a couple of years' time. SlimVirgin (talk) 01:39, 13 October 2013 (UTC)
  • I'm with SlimVirgin on this issue. Born2cycle appears to have no experience working with contentious corporate articles in which the corporation has a strong on-wiki presence. It is a nightmare for NPOV, I can tell you. Many times I have brought verifiable and reliable sources to the BP article and had the material removed. I say keep the COI page as strong as possible. Binksternet (talk) 02:04, 13 October 2013 (UTC)
  • Somebody who works for BP can't touch the article, even if fully disclosed, to correct an error such as a wrong title associated with a corporate officer, or a wrong name associated with a corporate office. However, an activist who hates BP can do whatever they like to the article. That seems...asymmetrical. Jehochman Talk 02:07, 13 October 2013 (UTC)
  • The asymmetry is that a corporation can hire a team of editors who can argue forever. Violations of NPOV take lots of time to confirm, behavioral guidelines are loosely enforced, sock-puppetry is difficult to uncover. It all takes time, and a volunteer editor often cannot stand up to the hordes of paid editors employed. That's asymmetry. Smallbones(smalltalk) 02:17, 13 October 2013 (UTC)
  • That would be a stupid thing for them to do because when outed, they would get written up in mainstream media and take a huge PR hit. I constantly advise clients not to do that, and think that all smart PR professionals would dispense similar advice. Whenever I'm hired as an expert witness one piece of the assignment is usually to look at whether the opposing party has been doing anything dirty. Big companies are constantly involved in lawsuits, and constantly have adversaries looking to find any dirty activities. Hiring shills to turn their Wikipedia pages into advertising or fluff would be extremely stupid. This is something done primarily by small companies or the unsophisticated. Jehochman Talk 02:21, 13 October 2013 (UTC)
  • Yelp Inc., a highly controversial company forever being dragged into all manner of litigation, has a paid editor working actively on the article. I am not aware of any articles being written about it. BP and Chevron's paid editors were the subject of publicity and those editors continued their good work for the company, undeterred. Coretheapple (talk) 02:49, 13 October 2013 (UTC)
  • I didn't see a declaration for a Yelp, Inc., COI on the article or the talk page, unless it's suggested/mentioned in a talk topic/section, which isn't prominent enough, assuming there should be a declaration. Could someone please rectify the absence or counter the claim? Nick Levinson (talk) 18:41, 13 October 2013 (UTC) (Reworded: 18:53, 13 October 2013 (UTC))

All of the above discussion is depressing; bordering on disturbing. What I'm hearing is a lot of experienced and respected editors essentially saying WP policies on notability and verifiability, etc. are a failure. What I have always loved about WP is that on any controversial issue the coverage would be reasonably fair because editors from each side would be involved to keep the other side honest. It seemed to me that as long as you had some people from each side involved, and it didn't even have to be close to balanced, all of the biased POV material would be kept out, and the result would be solid/balanced coverage. It's a brilliant system. I still see that. I still believe that. If some on one side are paid does not change this one iota. To believe it does, to argue it does, is to argue that the underlying system on WP is a failure. I don't buy it. I don't see it. This campaign against COI editing undermines the pillars upon which WP sits. It sanctions going after WHO and WHY instead of WHAT. It's a horrible idea which goes against WP interests. It needs to go. --B2C 05:16, 13 October 2013 (UTC)

The real issue is that we are allowing paid advocates and that the COI policy does not have enough teeth to effectively deal with them. We do not want the pharmaceutical industry writing about themselves and their products. They already write much of the literature (through ghost writers with MDs names applied after the fact), they hide evidence that is not to their commercial advantage, they are the top advertiser on TV, they fund the committees that draft up guidelines, there is more than one lobbyist per member in Washington acting on their behalf, NO they do not get to write Wikipedia too. Wikipedia is respected because we are different. Because we are independent. This suggestion would irreparablely harm the project. Doc James (talk · contribs · email) (if I write on your page reply on mine) 05:27, 13 October 2013 (UTC)
I agree, but I think that you'd have to admit that this incredibly depressing conversation indicates that the so-called "community" is utterly clueless on the subject of COI, hasn't the foggiest idea why it's important, and has its collective head buried so far up its collective rear that the situation is going to only get worse. That's why I'm coming around to the view that perhaps all vestiges of COI restraints should go, Wikipedia should surrender to the p.r. people, as long as a clear disclaimer is written at the top of each article indicating that the content may be compromised by COI. Coretheapple (talk) 05:41, 13 October 2013 (UTC)
No! Our policies and guidelines about content keep THE CONTENT from being compromised by all kinds of bias, including COI bias. If the policies don't protect the content from COI bias, they don't protect from other biases either. That's simply not the case. --B2C 06:06, 13 October 2013 (UTC)
The problem I've found is that having one person on one side being a paid editor does unbalance things, so that established policies don't always have a chance to work. In particular, I feel that paid editors (n general) are far more willing to act against other policies to get the requested change through. I've seen false representation, vote stacking, false references, (including reviews paid for in order to provide reliable sources), socks, meat puppetry, and at one stage, I saw what looked like someone being hired to take on the role of a mediator in dispute resolution. Money changes the equation, perhaps because people are more willing to compromise on other aspects when there is a decent reward sitting behind them. I agree with DocJames that we need teeth - personally, I'd be happy just to see a requirement that editors with a financial COI must declare their COI before they edit an article, even if we don't stop them from editing completely (which won't happen anyway), and have sanctions that we can use if problems arise. I feel that paid editing is going the be a given on WP - what I'd like is to make editing within policy more effective than editing outside of it, and for that we need clear rules and an ability to act against those who go against them. - Bilby (talk) 06:39, 13 October 2013 (UTC)
I hear you. But it just seems to me that the magnitude of the COI problem is greatly exaggerated, and measures taken to combat it are largely ineffective and cost more harm to WP than the COI problem itself. It's like the war on drugs. Some wars are just not worth fighting. I have faith in our CONTENT polices and guidelines that they will keep problems caused by COI editors in check. Will there be problems? Of course. Will they be dealt without special COI rules? Yes. --B2C 14:01, 13 October 2013 (UTC)
I think we have a tendency to underrate the size of the problem, because we base it on what we see, while we overrate the ability of existing policies to cope. NPOV and other policies are great when faced with articles watched by enough (or just the right) people, but fall down where the editors have different levels of experience and resources, or where the articles fall under the radar. Just to clarify, though - I don't feel that we should ban editors with a COI from making changes. I do feel we should follow the sorts of established practices in other areas, such as journalism, by requiring editors to disclose any financial conflict of interest if they wish to work on an article where that applies. - Bilby (talk) 22:11, 13 October 2013 (UTC)
WP:COI is a behavioral guideline about editing. Its purpose along with multiple other WP policies and guidelines is to create and maintain a neutral encyclopedia. If someone is editing non-neutrally it should be apparent in their edits. If it is not, then they are functioning as a productive contributor to the project and should be supported. Everyone has bias and editors are naturally drawn to subjects and topics that they are interested in, that they have opinions about based on their education, lifestyle, training, occupation, life experiences, religion, moral values etc. Some have very strong opinions and clear agendas like this skeptic group which recruits and trains like minded people to edit WP.[7] Creating a police state on WP where we investigate, or worse deduce, someone's occupation or income source or belief level, based on anything other then the same editing standards as others, only serves to calcify a specific, limited POV and create an insiders only project. WP works through collaboration and diversity (the wisdom of crowds). Any attempts to set up more rules with bigger "teeth" will only serve to bite WP in its arse and limit its diversity and progress. --KeithbobTalk 16:02, 13 October 2013 (UTC)
Some journalism publishers have much more stringent restrictions on journalists than Wikipedia has, and the ones I regularly see do good journalism. Nick Levinson (talk) 18:41, 13 October 2013 (UTC)
Keithbob's "police state" concerns are unwarranted as all of the most problematic COI situations that I'm aware of involve editors who have declared themselves to be paid editors, so the COI is not in doubt. Nick's points are correct, and I would add that Wikipedia is unique in its lack of integrity, for not disclosing the COIs of article contributors, even when open and declared, to readers. Coretheapple (talk) 18:46, 13 October 2013 (UTC)
Unfortunately, you'll find that there is a degree of selection bias there. - Bilby (talk) 22:11, 13 October 2013 (UTC)
It also appears to be a "slippery slope argument" which makes no sense because financial conflict of interest is a well known, widely observed concept, limiting the issue. Alanscottwalker (talk) 19:25, 13 October 2013 (UTC)

As I argued here we should make changes to some of our core policies. If medical science isn't as reliable as the other hard sciences, then our core polcies shouldn't assume that it is. Doing so mkaes Wikipedia vulnerable to COI editing, but enforcing the COI policy is itself problematic. Therefore it's better to deal with the source of the problem. Count Iblis (talk) 19:48, 13 October 2013 (UTC)