Thanks, Erik, for giving me a heads-up that this discussion is happening.
Jimmy Wales jwales@wikia.com wrote:
I think we need to be very clear in a lot of different places that PR firms editing Wikipedia is something that we frown upon very very strongly. The appearance of impropriety is so great that we should make it very very strongly clear to these firms that we do not approve of what they would like to do.
And Erik Moeller eloquence@gmail.com replied:
My take on it is that if we push PR industries to far to the outside, they will just do their work clandestinely. This will damage Wikipedia's reputation far more if it becomes known, especially when an article that has 200+ revisions was started and carefully groomed by a paid propagandist.
I'm not sure how this discussion originated, but I think Erik has a point. It sounds like the proposed restriction on PR firms is a special case of Wikipedia's general rule against editing articles about oneself or one's own organization, in which PR firms are seen as agents editing on behalf of their clients. The issue of "editing articles about yourself" has always been tricky, in part because it exposes a paradox in Wikipedia's editorial policy:
(1) Editing articles about oneself is strongly discouraged, if not outright prohibited; yet
(2) Anonymous editing is allowed, which makes it difficult if not impossible to enforce the policy against self-editing. (In fact, the policy against self-editing creates an INCENTIVE to edit anonymously as a means of evading the restriction.)
I realize that Wikipedia's existing policies have a lot of history and inertia behind them, but I have long felt that the policy against self-editing is problematic. I think it should be revised to the following:
Editing an article about yourself or your client is permitted, under certain conditions: -- CREATING articles about yourself or your client is not allowed, only editing of existing articles. -- Self-editing is allowed if limited to adding or correcting noncontroversial facts. For example, if the article about me gives an incorrect date of birth, I should be allowed to correct it. -- People should not self-edit when dealing with controversial or disputed facts or interpretations. Any such disputes should be addressed on the talk page and left to others to resolve, and can be submitted to arbitration if they are not satisfactorily resolved. -- People who wish to edit an article about themselves or their client are strongly encouraged to do so transparently, by disclosing that they are editing an article about themselves on the article's talk page.
The problem with the current policy is that is is gradually expanding into an umbrella that excludes whole classes of people from participating in Wikipedia. Members of the U.S. Congress and their staffs were banned (although they could easily circumvent the ban by simply editing from their homes rather than their offices). Now we're talking about restricting PR firms. It's easy to come up with other classes of people who could also be restricted from editing on similar grounds. What about lobbyists? Employees of think tanks? Trade associations? Labor unions? Should Christian missionaries and clergy be told not to edit articles about Christianity?
A restriction on editing "articles about yourself or your client" is also a poor fit with the problem that the current policy is attempting to solve, because some PR firms do work that is not about their client but rather aimed at attacking their client's COMPETITORS. One example that we wrote about recently involved a PR firm which circulated a claim that Apple's video iPods were dangerous to children because they could be used to download and view pornography. The client in this case was Sony, which makes a rival MP3 player. But how would you enforce a policy that says PR firms working for Sony can't edit Wikipedia articles about Apple products?
Another recent example: the DCI Group, a PR firm whose clients include Exxon, recently got caught anonymously circulating a video on YouTube that mocked Al Gore's activism on global warming. Al Gore was not their client, and the video didn't mention Exxon at all. But does it make sense to have a policy that says employees of PR firms can't edit articles about politicians?
Preventing this sort of thing from happening on Wikipedia would require a policy that forbids people who work for PR firms not only from editing articles about their clients, but also from editing articles about any topics of potential interest to their clients. A policy of this nature would be so vague that it would be a nightmare to enforce -- especially since many PR firms do not disclose their client lists.
On the other hand, a policy that requires PR firms to be transparent about disclosing whenever they edit an article related to a client or a client's interests, coupled with the restrictions that I outlined above, would be reasonably enforceable. PR firms would be discouraged from anonymous or POV editing by the strong possibility that they would be identified and embarrassed for doing so.
--Sheldon Rampton
On 8/22/06, Sheldon Rampton sheldon@prwatch.org wrote:
Preventing this sort of thing from happening on Wikipedia would require a policy that forbids people who work for PR firms not only from editing articles about their clients, but also from editing articles about any topics of potential interest to their clients. A policy of this nature would be so vague that it would be a nightmare to enforce -- especially since many PR firms do not disclose their client lists.
I'd just as soon block the PR firms entirely. Maybe once they get some semblance of a clue, but for now virtually every edit I've seen come from a PR firm has been crap.
Kelly
On 8/22/06, Kelly Martin kelly.lynn.martin@gmail.com wrote:
I'd just as soon block the PR firms entirely. Maybe once they get some semblance of a clue, but for now virtually every edit I've seen come from a PR firm has been crap.
Of course... That takes us back to the "then they well just operate in secret" argument.
I guess my counter is: We already have PR firms operating in secret. If we can't deal with secret POV pushing PR firms who can afford to learn the processes schmmoze our editors, then we've already lost. If we can deal with them acting in secret, then there is no reason for us to pretend that you are welcome.
I should bring up that when people come back in secret to edit after we have forbidden them from editing they are committing a crime. Are we honestly discussing how best to accommodate criminals here?
If we are not able to instill in the hearts and minds of the public that vandalizing Wikipedia is no more savory than spray painting the side of your local library, or using Wikipedia to advertise is no more acceptable than sticking ads in the libraries books then we will continue to find ourselves the victims of the worst aspects of human nature.
I believe our experience has shown that people today do behave respectfully with Wikipedia. Average folks are no more likely to vandalize wikipedia then they are likely to key a strangers car. But that respect starts with us respecting ourselves... few people would feel too bad about peeing on a heap of garbage.
So we must always take a public stand to protect Wikipedias integrity even when we know our measures are ineffective against a real bad-guy. Saying NO to the misuse of Wikipedia as a tool to distribute advertising and other biased information is an important part of that stand.
Like many things our rules in this regard should represent a constant effort to work towards a goal. Just as all the Wikipedias contain copyright violations although we forbid them, all will contain some advertising. Our current inability to achieve perfection is no excuse to discontinue fighting the good fight.
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
Of course... That takes us back to the "then they well just operate in secret" argument.
I guess my counter is: We already have PR firms operating in secret. If we can't deal with secret POV pushing PR firms who can afford to learn the processes schmmoze our editors, then we've already lost.
That is unhelpful rhetoric. What does it mean to "have lost"? Wikipedia is not going to disappear tomorrow just because we have some PR in some articles. We can deal with secretive activities when we detect them; that is not a logical argument for not creating a framework where these organizations can achieve some of their aims (where they are legitimate and in line with our own) while not doing harm.
You can easily compare this to the struggle against sock puppets and trolls. Sure, we can "deal with" sock puppets and trolls when we detect them. That, again, is not an argument for not building policies and community structures where harmful behavior is neutralized in ways that do not lead to such disruption.
On 8/22/06, Erik Moeller eloquence@gmail.com wrote:
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
I guess my counter is: We already have PR firms operating in secret. If we can't deal with secret POV pushing PR firms who can afford to learn the processes schmmoze our editors, then we've already lost.
That is unhelpful rhetoric. What does it mean to "have lost"? Wikipedia is not going to disappear tomorrow just because we have some PR in some articles. We can deal with secretive activities when we detect them; that is not a logical argument for not creating a framework where these organizations can achieve some of their aims (where they are legitimate and in line with our own) while not doing harm.
As you admit, we can deal with secretive activities, so obviously we haven't lost...
Our existing mechanisms are fully functional. We don't need special ones for PR firms any more than we need special cases for Congressmen, TV personalities, Gibraltarians, Friends of Gays (okay, perhaps them...) , or anyone else. My argument, which you've accused of being unhelpful, is simply that should our mechanisms for dealing with clandestine POV pushers fail, any issues with identified biased users will be trivial by comparison.
We have an existing framework where anyone can 'achieve their aims', so long as they are aligned with the aims of the project. We call it editing, and we're very open about who we allow to participate.
If someone is acting in our interests, they are indistinguishable from any other good editor. We don't need to create a special framework for people who come to Wikipedia with little interest outside of their own benefit. We have one for them already: banning.
I'm sorry that I didn't make this more clear in my prior post.
You can easily compare this to the struggle against sock puppets and trolls. Sure, we can "deal with" sock puppets and trolls when we detect them. That, again, is not an argument for not building policies and community structures where harmful behavior is neutralized in ways that do not lead to such disruption.
I'm confused; our policy on socks and trolls is that we do not accommodate them when they make their interest known.
Wikipedia is not for trolls, just as Wikipedia is not for PR firms. We should rightly tell both to go away.
Some trolls return in secret, but we can deal with them.
I have only proposed that we treat PR firms exactly the same way.
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
That is unhelpful rhetoric. What does it mean to "have lost"? Wikipedia is not going to disappear tomorrow just because we have some PR in some articles. We can deal with secretive activities when we detect them; that is not a logical argument for not creating a framework where these organizations can achieve some of their aims (where they are legitimate and in line with our own) while not doing harm.
As you admit, we can deal with secretive activities, so obviously we haven't lost...
I said "when we detect them." When we don't, they do us harm. Logically, more secretive activity means more harm. Furthermore, there is always the risk of false positives and unfair accusations.
We have an existing framework where anyone can 'achieve their aims', so long as they are aligned with the aims of the project.
That would suggest that we should try to incorporate neutral, factual information that they provide.
I'm confused; our policy on socks and trolls is that we do not accommodate them when they make their interest known.
A sock puppet is typically a returning user who has been punished by the system and is trying to evade enforcement by creating a new identity. Our processes (ArbCom etc.) are geared towards reducing the number of cases where we have to use hard enforcement tactics that are difficult to implement in an open environment, and instead encourage the use of soft enforcement. Users are often willing to accept such soft enforcement because it preserves their existing identity and reputation. When too much hard enforcement is used in an open wiki, it often breeds an atmosphere of suspicion, as sock puppetry is difficult to detect and the rules become more difficult to implement.
We do not accommodate "sock puppets", but we accommodate users whose behavior is slightly harmful (rather than being completely harmful, such as vandals) by trying to integrate them into the community and providing a safe framework of interaction (as well as a clear record of their past activities). This reduces the risk of turning them _into_ sock puppets, and thus, the risk of more harmful activity within our community.
Analogously, the WP:COI process suggests encouraging PR organizations whose behavior we consider _slightly_ harmful to work "within the system" in a fair and responsible manner, rather than turning them towards secrecy and disruption in ways which are _more_ harmful because they, like sock puppetry, breed an atmosphere of suspicion and are generally harder to trace and detect.
On 8/22/06, Erik Moeller eloquence@gmail.com wrote:
When too much hard enforcement is used in an open wiki, it often breeds an atmosphere of suspicion, as sock puppetry is difficult to detect and the rules become more difficult to implement.
I should also add that, once a user has decided to "become a sock puppet", they will then also often feel no longer any obligation to follow _any_ rules or processes of the community, except in the interest of avoiding detection and punishment. A user who, in Gregory's words, is branded a "criminal" does not have much respect for policy.
On the other hand, if we allow someone like MyWikiBiz to submit articles through an open process, it will be hard for them to cry foul when a fluff piece is turned into a factual article incorporating criticism.
The WP:COI page is perhaps a bit poorly structured. It is really a process of article submission from interested parties. This is something, I think, we should encourage, and ideally facilitate through our own web services, community and policies, rather than externalizing it in ways which are difficult to track.
Wikipedia has a tradition of soft responses to slightly harmful behavior, which I believe to be the only appropriate strategy in a technically open environment. If we required a real world ID and background check on everyone ever editing Wikipedia, that might be different.
On 8/22/06, Erik Moeller eloquence@gmail.com wrote:
I said "when we detect them." When we don't, they do us harm. Logically, more secretive activity means more harm. Furthermore, there is always the risk of false positives and unfair accusations.
Lets look at the possibilities:
-In a world we where tell PR Firms not to edit: * PR firm follows the rules, doesn't edit. **Pool of editing people decreased **Harm of biased insertions avoided or * PR firm ignores the rules, secretly inserts biased material. **We detect it: Harm is mitigated. **We fail to detect it: harm remains.
-In a world where we invite PR firms to edit after jumping through hoops: *PR firm follows the rules, edits (with boiler plates and notices) **Editing community size increased by the addition of editors driven by monetary interests in certain POVs. **Rightfully distrustful users overscrutnize every word, things we would normally accept are rejected **Worst of the bias removed, their ability to use Wikipedia for PR mostly suppressed. **Still some bias and POV leak through, but now through a blessed channel **Patient and well funded PR firms are able to wikilawyer to game the system, because we have no blanket prohibition. * PR firm ignores the rules, secretly inserts biased material. **Feel justified because we've permitted them to edit, but not allowed them to actually accomplish their goal. **We detect it: Harm is mitigated. **We fail to detect it: harm remains.
To me it seems like a simple question: Is it in the project's interest to invite a group of editors who for the purpose of enabling their outside agenda? If our agendas coincide, they can just edit as individuals... it's only in the case that they differ that we must consider here.
To me it seems obvious that any successful system of inviting edits by PR people would ultimately permit them to only make the same edits they could make as normal editors (and likely a bit less due to community paranoia). The PR firms that wish to insert bias will quickly learn to do it in secret, so we've gained nothing... and we lose the high-ground position of a blanket rejection policy.
We have an existing framework where anyone can 'achieve their aims', so long as they are aligned with the aims of the project.
That would suggest that we should try to incorporate neutral, factual information that they provide.
We accept neutral and factual edits from any person.
[snip]
We do not accommodate "sock puppets", but we accommodate users whose behavior is slightly harmful (rather than being completely harmful, such as vandals) by trying to integrate them into the community and providing a safe framework of interaction (as well as a clear record of their past activities). This reduces the risk of turning them _into_ sock puppets, and thus, the risk of more harmful activity within our community.
Do we have any data to back up the position that use of hard enforcement typically turns a slightly harmful user into willy on wheels? As far as I'm aware, this position is just speculation. I won't argue that being soft may allow a slightly harmful person to go on contributing while remaining slightly harmful, but I'm not aware of any solid evidence that being hard makes them into something worse, and I'm aware of fairly little evidence of slightly harmful folks becoming unharmful.
Analogously, the WP:COI process suggests encouraging PR organizations whose behavior we consider _slightly_ harmful to work "within the system" in a fair and responsible manner, rather than turning them towards secrecy and disruption in ways which are _more_ harmful because they, like sock puppetry, breed an atmosphere of suspicion and are generally harder to trace and detect.
[merge]
I should also add that, once a user has decided to "become a sock puppet", they will then also often feel no longer any obligation to follow _any_ rules or processes of the community, except in the interest of avoiding detection and punishment. A user who, in Gregory's words, is branded a "criminal" does not have much respect for policy.
Should we really expect the same behavior from bored 13 year old boys in London, and judgement proof nationalist trying to do their part in their nations wars via Wikipedia, as highly funded US corporations with substantial liability and a reputation to protect?
PR firms can't effectively completely change their identity twice a week... doing so would substantially obstruct their ability to obtain and maintain customer relationships.
On the other hand, if we allow someone like MyWikiBiz to submit articles through an open process, it will be hard for them to cry foul when a fluff piece is turned into a factual article incorporating criticism.
It's hard for them to cry foul when their business is outright prohibited as well.
Wikipedia has a tradition of soft responses to slightly harmful behavior, which I believe to be the only appropriate strategy in a technically open environment.
No one has proposed a depature from soft responses to editors.
In fact, I'm arguing that our handling of editors works well enough that we don't have to create a special "permission to promote" compromise in order to reduce the amount of secret editing that will happen.
Our wiki's were an interesting target for spammers long before they were an interesting target for press agencies. We did not respond to this activity by creating permission to spam and "This page has been edited by an identified spammer". We've addressed it with diligent editors, a few technical measures, cooperation with others (like google), and by showing respect for our project by having little tolerance. ... and I believe (and can substantiate with data) that our efforts with basic SEO and spammers have been remarkably effective.
I have seen no evidence why PR firms would be less responsive to a similar handling. If anything they should be more willing to accept that no means no.
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
PR firms can't effectively completely change their identity twice a week... doing so would substantially obstruct their ability to obtain and maintain customer relationships.
This is an interesting point. If we can get to a point where PR firms edit under a disclosed name (by making transparent editing beneficial to them), then they have an incentive in not doing anything stupid here: protecting their reputation. No PR company would want a story "PR company blatantly inserts bias at Wikipedia" written about them. That's just bad PR after all.
I'm still keen to hear what this "appearance of impropiety" thing is that we're so keen to avoid.
Steve
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
-In a world we where tell PR Firms not to edit:
- PR firm follows the rules, doesn't edit.
**Pool of editing people decreased **Harm of biased insertions avoided or
- PR firm ignores the rules, secretly inserts biased material.
**We detect it: Harm is mitigated. **We fail to detect it: harm remains.
-In a world where we invite PR firms to edit after jumping through hoops: *PR firm follows the rules, edits (with boiler plates and notices) **Editing community size increased by the addition of editors driven by monetary interests in certain POVs. **Rightfully distrustful users overscrutnize every word, things we would normally accept are rejected **Worst of the bias removed, their ability to use Wikipedia for PR mostly suppressed. **Still some bias and POV leak through, but now through a blessed channel **Patient and well funded PR firms are able to wikilawyer to game the system, because we have no blanket prohibition.
- PR firm ignores the rules, secretly inserts biased material.
**Feel justified because we've permitted them to edit, but not allowed them to actually accomplish their goal. **We detect it: Harm is mitigated. **We fail to detect it: harm remains.
What I got out of this analysis, which I don't see any basis to contest, is that by allowing PR firms to edit through some sort of approval process, we get all the costs of a bureacracy to manage approved PR firms, without in any way eliminating the cost of monitoring for PR firms (and others!) who attempt to push biased content clandestinely.
From a cost-benefit standpoint, I see insufficient value in
encouraging PR firms to edit through an approval process; it increases our costs and gains us rather little. We'll still have to hunt down and deal with the defecting PR firms and other bias pushers, and frankly I doubt that the gain from the cooperating PR firms will be significant.
Kelly
On 8/22/06, Sheldon Rampton sheldon@prwatch.org wrote:
I realize that Wikipedia's existing policies have a lot of history and inertia behind them, but I have long felt that the policy against self-editing is problematic. I think it should be revised to the following:
Editing an article about yourself or your client is permitted, under certain conditions: -- CREATING articles about yourself or your client is not allowed, only editing of existing articles. -- Self-editing is allowed if limited to adding or correcting noncontroversial facts. For example, if the article about me gives an incorrect date of birth, I should be allowed to correct it. -- People should not self-edit when dealing with controversial or disputed facts or interpretations. Any such disputes should be addressed on the talk page and left to others to resolve, and can be submitted to arbitration if they are not satisfactorily resolved. -- People who wish to edit an article about themselves or their client are strongly encouraged to do so transparently, by disclosing that they are editing an article about themselves on the article's talk page.
This is already largely the case. The autobiography guidelines don't forbid all editing related to one's self or related persons/activities and in fact explicitly mention the kind of uncontroversial editing you bring up as acceptable. http://en.wikipedia.org/wiki/Wikipedia:Autobiography
The problem with the current policy is that is is gradually expanding into an umbrella that excludes whole classes of people from participating in Wikipedia. Members of the U.S. Congress and their staffs were banned (although they could easily circumvent the ban by simply editing from their homes rather than their offices). Now we're talking about restricting PR firms. It's easy to come up with other classes of people who could also be restricted from editing on similar grounds. What about lobbyists? Employees of think tanks? Trade associations? Labor unions? Should Christian missionaries and clergy be told not to edit articles about Christianity?
I'd say it depends. Any editor who can write neutrally -- so that no one can tell which point of view they hold, so that they don't seem to be pushing any particular POV, so that they are receptive to criticism and feedback -- I don't have a problem with. When they become obvious or disruptive in their attempts to impose their views, and through that it becomes clear they have a conflict of interest, that is a problem; perhaps we should be quicker to restrict those with obvious potential conflicts of any sort from editing where their editing has been problematic.
A restriction on editing "articles about yourself or your client" is also a poor fit with the problem that the current policy is attempting to solve, because some PR firms do work that is not about their client but rather aimed at attacking their client's COMPETITORS. One example that we wrote about recently involved a PR firm which circulated a claim that Apple's video iPods were dangerous to children because they could be used to download and view pornography. The client in this case was Sony, which makes a rival MP3 player. But how would you enforce a policy that says PR firms working for Sony can't edit Wikipedia articles about Apple products?
The same way we enforce policy saying that PR firms working for Sony can't edit articles about Sony; the same way we enforce the general policy stating that if your sole purpose is to push a certain point of view that you will be disinvited from the project. (Which is to say -- unevenly, spottily, sometimes through long and painful processes, on a case-by-case basis, but once it is spotted it is usually looked upon disfavorably and the editors involved restricted or banned.)
Another recent example: the DCI Group, a PR firm whose clients include Exxon, recently got caught anonymously circulating a video on YouTube that mocked Al Gore's activism on global warming. Al Gore was not their client, and the video didn't mention Exxon at all. But does it make sense to have a policy that says employees of PR firms can't edit articles about politicians?
Preventing this sort of thing from happening on Wikipedia would require a policy that forbids people who work for PR firms not only from editing articles about their clients, but also from editing articles about any topics of potential interest to their clients. A policy of this nature would be so vague that it would be a nightmare to enforce -- especially since many PR firms do not disclose their client lists.
On the other hand, a policy that requires PR firms to be transparent about disclosing whenever they edit an article related to a client or a client's interests, coupled with the restrictions that I outlined above, would be reasonably enforceable. PR firms would be discouraged from anonymous or POV editing by the strong possibility that they would be identified and embarrassed for doing so.
I would prefer that PR firms do not edit in mainspace at all, but confine themselves to talk pages, where indeed they should identify themselves if they are at all ethical, or risk embarrassment. The goal of a PR firm is pretty basically not compatible with NPOV; while the basic information they have to spread may be valuable, any contributions made by them would need careful checking to minimize the effect of outside entities trying to compromise our neutrality. In talk space at least they are simply suggestions to be acted on as time permits rather than something being presented to the reader as encyclopedic material. No, we cannot identify all of them, but we don't want to give the impression that this is something we want or will allow if only someone will jump through enough hoops to do it.
We are already an incredibly attractive target for those who have a view to push; they know about us and how to game us and aren't going to disclose their biases if they think they can get away with not doing so, policy or not. (Indeed, I've seen enough who try to deny their association when called on it.) I worry about what will be allowed to slip in if we do not explicitly take a stand against it. That we have to rely on the judgment of our editors to determine when our policies are being violated instead of having a nice clean line to draw makes it difficult, but it doesn't mean we shouldn't do it.
-Kat