Wikipedia talk:Arbitration/Requests/Case/Manipulation of BLPs/Proposed decision

Arbitrators active on this case
Active:
 * 1) Casliber
 * 2) Chase me ladies, I'm the Cavalry
 * 3) Cool Hand Luke
 * 4) Coren
 * 5) David Fuchs
 * 6) Elen of the Roads
 * 7) Jclemens
 * 8) John Vandenberg
 * 9) Kirill Lokshin
 * 10) Mailer diablo
 * 11) Newyorkbrad
 * 12) Roger Davies
 * 13) SirFozzie
 * 14) Xeno

Inactive:
 * 1) Iridescent
 * 2) Risker

Recused
 * 1) PhilKnight

SEO experiment
Regarding the SEO experiment conducted by User:Prioryman—
 * Whorlton Castle has now moved to being the no. 2 Google result for google.co.uk searches using "Whorlton Castle" as the search term (conducted from a UK location).
 * About two weeks ago, Prioryman and myself reported it being the no. 10 Google result for the same search.
 * Note that all 61 links to the page presently shown on Yahoo! Site Explorer are Wikimedia links: (60 from English Wikipedia, one from Danish Wikiquote).
 * In google.co.uk searches for Whorlton Castle (without quotation marks) the Wikipedia article is now in 11th place (one lower than two weeks ago).

Various explanations for the rise from 10th to 2nd place are possible; they include:
 * The addition of internal links over the past two weeks has affected page rank, but with a time lag.
 * There are external links to the article that do not show up yet in YSE, but are taken into account by Google.
 * The expansion of the article over the past two weeks (from 12,000 to over 16,000 bytes) has affected page rank.

A number of SEO experts have stated that Wikipedia's internal links do affect the page rank of our articles (e.g., ). At this time, it would be safest to say that we cannot be certain one way or the other. -- J N  466  04:24, 29 August 2011 (UTC)
 * I don't agree with Jayen466's evidence here. For me the wikipedia article on Whorlton Castle is currently twelfth on www.google.fr, if I search with a French IP; and twelfth on www.google.co.uk, using an IP (institutional proxy) in the English town where Jayen466 is located. Mathsci (talk) 07:09, 29 August 2011 (UTC)


 * Jayen466's 2nd placed result is anomalous - it still shows up for me at 11th place on google.co.uk and has done for some days now. I've tested this from multiple networks, logged out and with cookies turned off to prevent Google personalising the results for me. Jayen466's last three bullet points thus seem to be premature conclusions. Opinion among SEO experts appears to be divided - one that I have consulted believes that internal linking is not a factor and that the vast majority of a Wikipedia article's ranking score is due to (1) the fact that it's on Wikipedia in the first place and (2) the way an article is structured, which due at least as much to the site design and manuals of style as to the user. My experimental findings lend support to this proposition as no obvious or significant change has resulted from any editing on the article or templates.


 * My conclusion is stronger than Jayen466's – we can be certain that there is no positive experimental evidence that editors' actions have any effect on page ranking. There was no evidence of any sort before claims of search engine manipulation were made and the evidence we now have shows that no link has been positively demonstrated.


 * As such, I have to dispute principle 3.1.7 - there has never been any evidence of any kind (and none has been presented) that anyone has attempted to engage in "manipulation of search engine results", nor that any such manipulation is even possible. It is exasperating to see yet again that assumptions of wrongdoing are being made without any evidence whatsoever to support them. If this is supposed to be a fact-based decision then principle 3.1.7 doesn't belong in it. Prioryman (talk) 07:15, 29 August 2011 (UTC)
 * It is the 1st link for me. Google are privacy arseholes as they record your web surfing even when you aren't logged into a google account. You need to switch off google analytics tracking in addition to logging out of your account, and in your account you need to turn off web search tracking too, and you need to delete any google cookies as well See also this article amongst others. John lilburne (talk) 10:16, 29 August 2011 (UTC)
 * Using the text web browser lynx remotely on a different machine in the UK, it still comes up as number 11 for me. Mathsci (talk) 10:32, 29 August 2011 (UTC)
 * Is that with or without quotes, or both? It comes up 11th for me without quotes, but second with quotes. -- J N  466  13:06, 29 August 2011 (UTC)
 * With lynx the position in the listing does not change with quotes. Mathsci (talk) 13:50, 29 August 2011 (UTC)
 * FWIW, it's 11th when I do the search in the U.S. while logged in with a Google account.   Will Beback    talk    08:14, 29 August 2011 (UTC)
 * Also 11th using google.es from Spanish IP (unlogged in google account). --Enric Naval (talk) 12:18, 29 August 2011 (UTC)


 * I can't comment on the SEO tests, but I do note that 3.1.7 says "attempts". Unless a personal admits this was their intent, I don't see how we could distinguish attempts at SEO from "building the web", which Wikipedia editors are encouraged to do. There would need to be some clear guidelines on this to prevent editors from unintentionally giving the appearance of SEO.   Will Beback    talk    07:19, 29 August 2011 (UTC)


 * Yes, exactly. Let's be clear about this: in my own SEO experiment, everything I did, editing the article, structuring it, expanding it, adding links, creating a template, was entirely normal editorial practice. Although it was consciously intended to influence page ranking, it was literally indistinguishable from normal editing. It seems to me that claims of "SEO manipulation" have been used as a kind of political weapon to delegitimise editing (and one specific editor) with no evidence whatsoever presented at any stage that there was any such intention. I'm concerned that by writing this up as a principle the ArbCom is in danger of institutionalising assumptions of bad faith. There is simply no way of telling whether an otherwise proper edit is intended to manipulate search engines without making an assumption about the editor's motives. As soon as you do that, you are denying them an assumption of good faith. 3.1.7 not only invites assumptions of bad faith, it legitimises them. Prioryman (talk) 07:34, 29 August 2011 (UTC)


 * It seems to me that claims of "SEO manipulation" have been used as a kind of political weapon to delegitimise editing (and one specific editor) with no evidence whatsoever presented at any stage that there was any such intention. You mean much like repeatedly calling someone's critics a "lynch-mob" out on a "witch-hunt?" I think we need to consider Prioryman's experiment with a great deal of caution. It is anecdotal evidence until it is replicated, and replicated in a manner similar to the instances where others have suggested that someone, or some group of people have been working at manipulating search engine results. It may prove correct, but it may also not prove to be correct when other variables are added into the mix (variables that exist in those other instances). What Prioryman's experiment should hopefully do is get more similar experiments moving so we can get a better idea of what is going on. I will also grant that it suggests that there is certainly no conclusive evidence that anyone has to date managed to manipulate SEO, and that no action should, to date, be taken against anyone on that premise. Cheers.Griswaldo (talk) 14:04, 29 August 2011 (UTC)

Anent this - I proposed a specific experiment at UT:Jimbo Wales in the past   - where a term would be used which has "null return" on Google, making the effect of "Wikipedia bombing" completely separable from the examples heretofore used. Collect (talk) 15:24, 29 August 2011 (UTC)


 * You're all barking up the wrong tree. Internal links are counted by Google, but Google has the ability to identify navigation and treats those links somewhat differently. It would be hard and time consuming to set up a proper experiment.  For the purpose of this case, these experiments are not useful. The relevant question to ask is whether internal link spamming is acceptable or not.  If that activity is disruptive, it should be stopped, regardless of whether the tactic is effective at manipulating Google's search results. Jehochman Talk 16:20, 29 August 2011 (UTC)
 * That still leaves the question of how to differentiate between inappropriate "internal link spamming" and appropriate linking. I guess Manual of Style (linking) is the main guideline.   Will Beback    talk    23:00, 29 August 2011 (UTC)


 * As a totally uninvolved case-watcher, I was pleased when I saw 3.1.7 proposed. I'm not sure whether it relates to this case, but the principle as written is sound and I do know there are people who have tried to manipulate Wikipedia content solely for their own personal gain via Google (either to try to knock an article off the top spot, or to plug their material in an article so it gets views). This editing behaviour is never done with the best interests of the encyclopedia in mind.  Them From  Space  03:07, 30 August 2011 (UTC)


 * I'm sure many people here are experts on Google. But for the rest, here's an interesting article on how they have changed their algorithms in the past and their plans for future changes. "Google Wants Search to Know What You Know" Technology Review.   Will Beback    talk    09:52, 30 August 2011 (UTC)

ArbCom enforcement
Based on this case decision, can we begin requesting corrective action at Arbitration/Requests/Enforcement for editors who violate the BLP policy? Cla68 (talk) 04:37, 29 August 2011 (UTC)
 * Which remedy would be enforced? If there are BLP problems with a particular article then maybe the best place to start would be either the article talk pager or BLPN.   Will Beback    talk    04:46, 29 August 2011 (UTC)
 * This proposed decision makes it clear, especially principle 2, that the BLP policy should be strictly interpreted and enforced. If the ArbCom enforcement board is an option for dealing with violators of the policy, it would help avoid having to request formal ArbCom cases for repeat offenders, and thereby reduce the Committee's future workload.  In order to keep the enforcement board from being overwhelmed with BLP enforcement requests, I suggest that a "Proposed enforcement" be added that states that the expected steps in the dispute resolution process for violators of the BLP policy follow this order:  1) discussion on the editor's talk page and or discussion at the BLP noticeboard, 2) editor conduct RfC, 3) mediation and/or report to the ArbCom enforcement forum, 4) RfAr. Cla68 (talk) 05:40, 29 August 2011 (UTC)
 * I've never seen anyone go to AE over the failure to follow a principle. All policies should be followed, but only violations of ArbCom remedies are reviewed at AE, so far as I know.
 * Is this a hypothetical case or are you thinking of any particular "violators"? Is there any active dispute that you think needs to remedying? You didn't file much evidence in this case.   Will Beback    talk    05:46, 29 August 2011 (UTC)
 * I have noticed that the ArbCom enforcement board has been fairly effective at reducing problematic behavior in troublesome topics such as Palestine/Israel. I suggest the same for BLP policy enforcement.  If editors knew that they could potentially be subject to ArbCom enforcement for violations of the BLP policy, I think they would be less bold about trying to shoehorn sketchy, negative, or pejorative information into BLP articles.  I'd like to hear what the arbitrators think of my suggestion. Cla68 (talk) 05:56, 29 August 2011 (UTC)
 * Unless they pass a sanction, there is nothing to enforce. Discretionary sanctions might be helpful.  Certain editors that keep trying to put gossip into articles tend to muck up the talk page and noticeboard discussions with endless arguments.  It would be helpful to be able to topic ban them at AE. Jehochman Talk 06:11, 29 August 2011 (UTC)
 * I agree. Arbitrators? Cla68 (talk) 06:14, 29 August 2011 (UTC)
 * There was no evidence in this case that any parties to this case added any gossip to any articles.   Will Beback    talk    06:23, 29 August 2011 (UTC)
 * Oh? Collect (talk) 12:31, 29 August 2011 (UTC)
 * Oh?   Will Beback    talk    12:43, 29 August 2011 (UTC)
 * Oh? (If one more editor joins us, we can have a quartet.) Jehochman Talk 16:21, 29 August 2011 (UTC)
 * I have no involvement in this case, and while I assume these are intended to be veiled references to specific editors, I have no idea whom they're aimed at. That said, if an editor is repeatedly trying to put inappropriate gossip into articles, please let me know and I would be happy to deal with it. I don't think admins need discretionary sanctions to handle that sort of thing, nor should they. MastCell Talk 16:36, 29 August 2011 (UTC)
 * I would suggest putting the WP:BLPN noticeboard on your watchlist, because activist editors are often featured in discussions there as they try to defend their sketchy editing of BLP articles. Cla68 (talk) 23:01, 29 August 2011 (UTC)
 * That's rather an attack, don't you think? Who are these activist editors defending their sketchy editing?   Will Beback    talk    23:05, 29 August 2011 (UTC)
 * Do you agree that adding references to published gossip, innuendo, speculation, and punditary is in itself the perpetuation of gossip, innuendo, speculation, and punditary? Unless we can all agree on that, then IMO the problems are just going to continue. John lilburne (talk) 08:31, 30 August 2011 (UTC)
 * Let me know when you've deleted this material: Sarah Palin. It's entirely speculative.   Will Beback    talk    09:39, 30 August 2011 (UTC)
 * I agree that the sentences clearly marked as "speculation" in that article are "speculation." Moreover, the rest of that material is factual and != "speculation."   Thus removal of the "speculation" would not be objected to by me as such, and I trust you will remove such similar material in all the LaRouche BLPs as soon as possible. Cheers, and thanks. Collect (talk) 15:55, 30 August 2011 (UTC).
 * What was the problem? As we are all going to be playing nice I hope you'll lend you weight to seeing that it sticks. John lilburne (talk) 16:32, 30 August 2011 (UTC)
 * Good. Let's see how long before it creeps back in. And let's not forget similar speculation in countless other political articles. As for LaRouche, I don't think anyone is speculating that he will run in 2012, though there was some about 2008 until he stated that he was too old to run but not too old to serve as an adviser to whoever won.   Will Beback    talk    21:30, 30 August 2011 (UTC)
 * Super to see that we have reached a decision that speculation doesn't belong in the articles. High five an all. When I get the time I'll come help you trim out all that 70s dross from the LaRouche article. If we put our heads together we can probably get 80% of it out. I'll cut it and you can tell me what is honestly essential. John lilburne (talk) 23:02, 30 August 2011 (UTC)
 * All you did is cut material labeled "speculation". Don't get ahead of yourself. Anyway, this isn't the place to discuss article edits. If you want to make significant changes to an article I suggest you start a thread on the relevant talk page.   Will Beback    talk    23:06, 30 August 2011 (UTC)

3.3.4 Future dispute resolution
Are there any "serious unresolved issues with regards to edits that are currently live in a BLP"? I don't think there was any evidence of such issues.  Will Beback   talk    07:24, 29 August 2011 (UTC)


 * As I noted before, I am neither deaf nor blind, nor do I think the committee members are.  There are most ceetainly "unresolved issued" and I think that this was made absolutely clear.  I regret that the decision is as weak as it is, as it will not prevent a host of future cases.  I wish they had seriously addressed the problems I set forth (noting that my evidence was aimed at such a decision, and was not aimed at individuals per se).   That said, I hope and trust that some additional material will be added to the proposed decision in that direction.  Ah well - I hope at least they understand that making a strong decision now would prevent future cases.  Cheers. Collect (talk) 12:30, 29 August 2011 (UTC)
 * Name these "unresolved issues".   Will Beback    talk    12:42, 29 August 2011 (UTC)


 * Not necessarily; but if there are, I would prefer them brought to the committee's attention immediately. – xeno talk  13:03, 29 August 2011 (UTC)


 * Apparently no one who reads this page can think of any.   Will Beback    talk    09:41, 30 August 2011 (UTC)

Overall comment
Overall, I think that Brad did an excellent job of writing this, particularly in not going beyond what the evidence would support. My main concerns would be about the SEO part, and the comments by Xeno and Roger Davies capture very well what my concerns would be. --Tryptofish (talk) 22:30, 29 August 2011 (UTC)
 * I agree. A well-written decision, with some caveats about the SEO material.   Will Beback    talk    22:45, 29 August 2011 (UTC)

There are some useful clarifications in the proposals from Newyorkbrad here. Off2riorob (talk) 14:26, 29 August 2011 (UTC)


 * Though to some extent I'm glad that the decision did not go even further to sanctify BLP, I also disapprove of the SEO principle. The first problem I have with it is that ArbCom hasn't really settled the actual feasibility of SEO.  Some allegations were made about templates in the santorum (neologism) case leading to some unreasonable removals of the term from navigation templates where it was appropriately listed, and I'm concerned that this principle will be used as an official imprimatur for further paranoia on the topic.  Nor do I see any reference to the tremendous double standard with which this principle is applied.  Things like Template:The Beatles are permitted to fill articles with 57k of spam intrawiki links, but they're not even being questioned.  How can we have that, but only apply the SEO principle to allow BLP extremists to do reverse SEO on articles they don't like in an effort to keep them from being listed by Google by breaking internal and incoming links? Wnt (talk) 03:30, 30 August 2011 (UTC)

I'm concerned that Arbs are rushing to sign-on to the SEO piece in a spirit of get it over with. It might be reaching too far, in that it states that SEO is a very bad thing, but leaves the door open to accusing someone of it based on undefined criteria. Would it be better to state that the community needs to attempt to develop policy to deal with SEO? --Tryptofish (talk) 19:43, 30 August 2011 (UTC)


 * "Attempts to manipulate Google rankings via internal link spamming are wrong" would be a good statement. Of course, people will argue endlessly about what is link spamming, and what is proper linking for usability.  Jehochman Talk 20:43, 30 August 2011 (UTC)
 * The problem is indeed distinguishing between normal linking, adding to templates, etc, and attempts to manipulate SEO. Creating a principle that people will argue over endlessly isn't a good idea.    Will Beback    talk    21:32, 30 August 2011 (UTC)
 * How do you distinguish between "normal linking" and "attempts to manipulate SEO"? My linking on Whorlton Castle, the subject of my SEO experiment, was deliberately intended to manipulate the ranking of that article, yet to any outside editor it wouldn't have - and didn't - look any different to normal linking. Prioryman (talk) 21:44, 30 August 2011 (UTC)


 * Actually, I did not think Google is the world's only search engine.  I would have stated it as Attempts to increase visibility of any topic or claim regarding a topic or living person covered in a Wikipedia article, or of any outside website, by increasing links within Wikipedia or outside of Wikipedia are improper, and may be dealt with by any administrator as improper behaviour.  Collect (talk) 20:53, 30 August 2011 (UTC)
 * That pretty much covers all internal linking, doesn't it? – xeno talk  20:55, 30 August 2011 (UTC)
 * Any excessive linking - how often is there a valid reason to iterate links over multiple articles for the precise same information? Yet there are cases where the precise same information is conveyed in dozens of articles, all carefully then cross-linked as well.  Have you not seen any such cases? Collect (talk) 21:23, 30 August 2011 (UTC)
 * Evidence?   Will Beback    talk    21:25, 30 August 2011 (UTC)
 * Try Anders Breivik -- now or formerly linked at over fifty separate pages - including under the kinfd of gun reportedly used. Need Milk? Try for Matt Breivik  this example of COATRACK:  Breivik is not related to mass murderer Anders Behring Breivik.   Care to explain why someone not having any relationship needs to have that in their BLP?   It is important to state that Breivik finds a tv show "hilarious"?  How many examples are needed, Will?   Cheers. Collect (talk) 21:35, 30 August 2011 (UTC)
 * But how do we know that someone is trying to alter the SEO results? This seems like normal, if exuberant, editing. Looking at the list of inbound links,, I'm sure there are some that could be trimmed but most look legitimate. Between the murder, the manifesto, the coverage, and the reactions, is inevitable that the article will have links from many others. It's just a fraction of the links to Charles Manson, who killed no one himself, and whose followers killed far fewer people than Breivik.   Will Beback    talk    22:54, 30 August 2011 (UTC)
 * Indeed; the point of writing an article is to have it be seen and read. Where to draw the line? – xeno talk  22:56, 30 August 2011 (UTC)
 * It seems to me that the line is being / has been drawn purely on the basis of speculation about editors' motives. At the risk of sounding like a broken record, nobody has ever presented a shred of evidence that any editor intended to, or actually did, engage in any SEO activity in relation to disputed articles. That is why 3.1.7 is so bizarre - it's as if the ArbCom was presenting a principle stating "It is an extremely serious abuse to enter Wikipedia's server room and turn off the servers" on the basis of unevidenced allegations that a recent outage was due to someone doing that. I mean, it's nice to know that the ArbCom is keeping on top of hypothetical misdeeds, but really, what is the point? Prioryman (talk) 23:08, 30 August 2011 (UTC)
 * I've seen a lot of things that look like gaming of the system pertaining to the Breivik article. If ArbCom wants an eye-opener, they should look at the image deletion closes for that article - people deleting images from Commons because they "glorify" the shooter, playing copyright games where they claim that practically all of the world's media have violated copyrights of a purely hypothetical third party photographer, deleting an image from Wikipedia because it's on Commons then using that deletion as precedent for deleting similar images from Commons, and once the least unusual images were deleted, arguing that the remaining image of Breivik with a gun is unacceptable because it'll encourage serial killings.  Likewise, using unfounded and excessive copyright claims, they don't want anyone to link to the manifesto or YouTube video, the definitive primary sources by Breivik about himself!  I haven't heard of this idea to criminalize relevant Wikilinks before, but it sounds like more deletionist gaming to me.  And despite any number of allegations I've seen here that it's immoral to cover Breivik's in detail, I don't believe this is a matter of ethics for one minute - I think that some people just don't want readers to link the conservatively dressed nice European boy look and Tea Party politics of Breivik with the idea of terrorism; I suspect some of them want to keep it so that when people think of terrorists they think of Arabs in turbans.  I ask people to consider that it is not the inclusionists who are editing with an agenda here. Wnt (talk) 00:38, 31 August 2011 (UTC)
 * Thank you very much for showing why I feel the committee should have made stronger findings. It is not up to editors to show how evil Breivik and the Tea Party are. and to link them all with "terrorism." It is up to us to follow only what reliable sources state as fact - and injecting what we know to be the truth is not what the project is, or ought to be, about.  Again thanks for making this so crystal clear. Cheers. Collect (talk) 00:42, 31 August 2011 (UTC)
 * There's no here evidence that policies have been broken in editing the Breivik article. If there have been problems then a separate case should be started on that dispute.   Will Beback    talk    00:54, 31 August 2011 (UTC)
 * I did not propose going beyond the existing sources, which are more than sufficient. Wnt (talk) 01:07, 31 August 2011 (UTC)
 * Wnt, are you stating that existing sources are sufficient that Matt Breivik needs a claim that he is not related Anders Breivik in his article? And that there is sufficient RS sourcing to call Anders Breivik related to the Tea Party and that both are "terrorists"?  I fear that others do not agree.   And Will -- if you see no here evidence of any violation, then you are quite in the minotity and, in so doing, giving clear evidence that the decision is far from strong enough. Thanks! Collect (talk) 08:03, 31 August 2011 (UTC)
 * I very much doubt such a disambiguation is necessary, and absent evidence to the contrary I have no trouble with taking it out. What I reject is the suggestion that someone added that link to that article in some POV crusade to increase the Google rank of the Wikipedia article about Breivik!  What possible point is there for anybody to do that?  Besides, adding one link from one article is about 1/40000 as potent (if it has any effect as all) as keeping a template like "The Beatles" which to be best of my recollection adds something like 200 links to 200 articles.  So suggesting SEO in this context is absolutely absurd, and contributes to the very palpable feeling that a certain group of deletionists is hunting down and eliminating whatever it can about the Breivik article out of their unsubstantiated belief that it is "unethical" to provide information about this person because it "glorifies" him.  What I'm seeing, right here, without the ink even being dry on this decision, is that SEO is merely a tool, like WP:BLP or WP:NOTNEWS or WP:Notability or any other policy that comes to hand, to be used without regard to what it actually says in an effort to eliminate unwanted information.  Now yes, that's harsh, but that's how I feel, and I really can't think of a way in which someone could really believe that that disambiguation link was genuinely intended to affect search engine rankings. Wnt (talk) 02:21, 1 September 2011 (UTC)

(od) Kindly point out where I made a claim about SEO. Really. I dislike dealing with straw arguments as to my position. Cheers. Collect (talk) 02:56, 1 September 2011 (UTC)
 * You're saying you raised this example in this section even though it had nothing to do with SEO? Wnt (talk) 18:10, 1 September 2011 (UTC)
 * Um -- what are you saying? I said I did not mention "SEO" in this section.  The post to me rather implied that I dad mentioned it, and setting the record straight seemed like the logical thing to do.  Do you feel the Breivik ref is suitable where the actual fact is that Anders Breivik has absolutely nothing to do with the person the article is about makes any sense at all? Cheers. Collect (talk) 18:50, 2 September 2011 (UTC)
 * Somehow we're not connecting here at all. You asked me that a few paragraphs above and I said I very much doubt that we need to have that link.  When there's no specific evidence about such confusion (like media reports or people asking repeatedly on the talk page) the choice whether or not to add a disambiguation link for clarity is truly at the whim of the editors involved.  I'd vote against it myself, and I'd expect it to be voted down.  I'll also say that the best practice, what I'd like to see on Wikipedia, is for Breivik to be Wikilinked to an article all about the name, when and where it arose, how many people bear it and where they're distributed in the world - alas, we're not close to that level of coverage of genealogy.  I'd further say I think WMF should have a proper Wikigenealogy project carrying all the old public records that are now treated like the private property of sites like ancestry.com, so that you could click on Matt Breivik and Anders Breivik, pull up the known records, and get back any linkage between them dating back however many generations.  Back in 2006-2007, before the rise of the deletionists, it really seemed like all kinds of things like that were just around the corner.
 * When I read what you wrote above, first you propose a SEO principle (I would say, since you're suggesting an alternative wording of one labelled as "SEO" by three editors at the beginning of the thread, ending in Prioryman's question How do you distinguish between "normal linking" and "attempts to manipulate SEO"?), namely "Attempts to increase visibility of any topic ... by increasing links within ... Wikipedia are improper". Then you're asked what it covers and you say "excessive linking".  Then you're asked for evidence and you name the Breivik disambiguation as an example of excessive linking.  How on Earth can anyone interpret that as anything other than a statement about SEO?  I can present this only as an exhibit of a remarkable lack of communication. Wnt (talk) 20:21, 2 September 2011 (UTC)
 * I suggested: Attempts to increase visibility of any topic or claim regarding a topic or living person covered in a Wikipedia article, or of any outside website, by increasing links within Wikipedia or outside of Wikipedia are improper, and may be dealt with by any administrator as improper behaviour.  which does not say "SEO" nor apply to "SEO" or have anything to do with "SEO" other than your apparent assertion that it must do so.  Now if a person does not say "SEO" and his posts are not related to "SEO" and he says he did not post about "SEO" I would humbly suggest you accept that he did not mention "SEO" nor post about "SEO" not mention "SEO."      Is this sufficiently clear? Cheers. Collect (talk) 21:58, 2 September 2011 (UTC)
 * So you're not talking about search engine optimization, but attempts to increase visibility of any topic by increasing links outside of Wikipedia. I think we're done here. Wnt (talk) 22:28, 2 September 2011 (UTC)
 * OTOH, you could say you misapprehended what I wrote.  You might also note that I asked on UT:JW about having a fully controlled experiment thereon, by the way, as I remain unconvinced one way or the other as to whether Google, Yahoo etc. might be misled by editors on Wikiopedia. Cheers. Collect (talk) 22:31, 2 September 2011 (UTC)

Questions about search engine optimization
A proposed principle supported by seven arbitrators states "It is an extremely serious abuse of Wikipedia to utilize editorial and structural features of the site—such as internal links, external links, and templates—in an attempt to inequitably or artificially manipulate search engine results. This is particularly so where the purpose is to disparage a living person." I think I should be clarify my reservations with a few basic questions.


 * Does this decision imply that there should be a maximum size (bytes or number of linked articles) for navigational templates like Template:The Beatles? If so, what is a fair limit?


 * If Template:The Beatles is not a problem to be fixed under this principle, what other template would be? For example the disputed templates from the santorum debate included a much smaller number of target articles, and therefore linked to the santorum article from a much smaller number of pages.  I would wonder if the intention of ArbCom then was to rule that SEO for commercial purposes, such as bringing eyeballs to articles about Beatles albums, is not a significant concern.


 * Can a simple internal link to a Wikipedia article disparage a living person, when the article itself is not in violation of policy?


 * Does the prohibition of SEO manipulation extend to efforts to decrease search rankings for a Wikipedia page, e.g. by preventing robots from crawling a controversial article, or (as was proposed at one point for the santorum article) moving an article and deleting the redirect so that links from news organizations and other third party sites into Wikipedia would be broken?


 * Does ArbCom suggest that adding relevant wikilinks to articles can be "an extremely serious abuse"? How would you distinguish between the normal effort any editor might make to link to an article from various related topics from an improper SEO?


 * What is the theory by which adding external links to an article would affect its search engine ranking? What distinguishes such improper links from normal references and external links?


 * Does starting noticeboard discussions about disputed articles and/or and repeatedly linking to them from RFCs and other policy discussions count as SEO manipulation? Wnt (talk) 07:04, 31 August 2011 (UTC)


 * Noticeboards are not indexed and used by Google AFAICT, so the straw issues seem to be made of ... straw. Cheers. Collect (talk) 08:04, 31 August 2011 (UTC)
 * In this search, the second entry is a noticeboard. There seems to be substance to that point, though whether SEO is affected remains the question.    Will Beback    talk    08:10, 31 August 2011 (UTC)

Some broader questions:


 * What is the basis in policy for this principle? Is ArbCom declaring it as a new policy?  Is ArbCom accepting WP:SEO, an essay started by a few people on one side in this case, as policy?


 * Does this principle cover efforts to change the name and/or first sentence of an article so as to make the Google result look a certain way? If so, does it command that people make these edits ingenuously, without regard to the search engine, or does it command that these parts of the article be altered so that when taken out of context in a search result they don't appear to disparage a living person?  If the latter, does this (what?) override existing naming and editing style guidelines?  (Note that these types of changes were made to the santorum article) Wnt (talk) 15:47, 31 August 2011 (UTC)


 * I'm flattered, but my essay WP:SEO, written mainly in 2007, isn't policy (yet). You may be thinking of WP:SEOBOMB. Jehochman Talk 18:40, 31 August 2011 (UTC)


 * Oops! That's the one. Wnt (talk) 02:07, 1 September 2011 (UTC)

The principle as written deals with situations in which an editor screws around with links or similar feature with the intention of artificially and inequitably manipulating search results. Good-faith editing will rarely, if ever, bring the principle into play, and it is obvious that Template:The Beatles is utterly remote from the concern the decision addresses.

More generally (as I have seen this type of argument from Wnt in multiple contexts), I do not agree that because there are borderline cases in applying any rule or principle, no rules or principles can ever be adopted. By that logic (and admittedly exaggerating slightly for effect), there are borderline-vandalistic edits so we shouldn't have a policy against vandalism, and there are debatable copyright infringements so we shouldn't have a policy against copyright infringement, and there are disputed claims of harassment so we shouldn't have a policy against harassment. This type of argument is not reasonable. (It might have greater validity in some situations as applied to governmental action that raises First Amendment issues, but this is a private volunteer website that has acknowledged our obligations both to ourselves, our readers, and our article subjects to discuss people with an appropriate degree of decorum and responsibility.)

I do agree with Wnt that it there should be a community discussion of policy in this area, particularly if what I believe has so far been an isolated problem of misuse becomes more extensive, which I hope it will not. Newyorkbrad (talk) 01:06, 2 September 2011 (UTC)


 * Has there ever been a known case of someone intentionally trying to enhance the SEO of a Wikipedia article? there was no evidence of any presented in this case. (I know there are many cases of people using Wikipedia articles to try to improve the SEO of their private websites, which is not explicitly prohibited but is indirectly discouraged through various guidelines). This has the appearance of ArbCom trying to create a policy out of thin air.   Will Beback    talk    01:29, 2 September 2011 (UTC)
 * There were credible allegations that this was intended during the Santorum (neologism) situation a couple of months ago. We haven't made a finding on this issue, but if there's a possibility this was intended, I have no problem trying to nip it in the bud. If you're right that the situation has never happened, then saying it shouldn't happen shouldn't be create much of a problem. And to me the principle is frankly an obvious one; after all, the converse would be "it is permissible to utilize editorial and structural features of the site—such as internal links, external links, and templates—in an attempt to inequitably or artificially manipulate search engine results, including to disparage a living person," which I hope no one would support.
 * All that being said, I think the SEO principle is getting disproportionate attention on this page, to the detriment of the rest of the decision. If this means that the rest of the decision doesn't raise issues worth discussing, fine, but I always hate (in any community discussion) for a peripheral issue to swamp the main one. Newyorkbrad (talk) 01:34, 2 September 2011 (UTC)
 * As explained on this page, there is a danger in creating principles like this because they can become the basis for alleging misbehavior by editors. I would be surprised if, in the coming year, this principle is not used against an editor without any actual proof of intent. If that does happen, I think the community should be in a position where it can revoke this unsupported principle that's based on guessing editor intent.   Will Beback    talk    01:42, 2 September 2011 (UTC)
 * If the allegation fits, it will probably be made, whether there's a relevant policy or not, or a principle in an arbitration case. Frankly, it's not as if people all over the wiki are studying the principles in arbitration decisions and either conforming their behavior to them, or keeping track of them to invoke in disputes later on. Outside a few specific topic-areas whose editors spend too much of their lives on the arbitration-enforcement page, the reading public for the Committee's decisions, as we arbitrators have to remind ourselves sometimes, is smaller than we think.
 * An irony is that I was just going through the arbitrator comments on the Cirt-Jayen466 proposed decision page, and one of my colleagues opposed a principle that included the words "... can reasonably be perceived as ...", on the ground that it is too subjective. (In fact, reasonableness is usually meant to connote an objective standard, but I digress.) I had included those words precisely to avoid creating an issue of motive and having to read editors' minds. I wouldn't mind a similar rewording here to remove the element of mind-reading, but I haven't been able to think of a rewording that would resolve this issue without creating new ones. If and when the community turns to developing a policy in this area&mdash;essential only if the problem becomes more serious, more likely&mdash;with more time and more participants it may do better in framing the point to be made than I have. Newyorkbrad (talk) 02:10, 2 September 2011 (UTC)
 * I'm surprised and disappointed to see you take this line about the Santorum article, especially given your professional background. I would have thought you of all people would have appreciated the need for evidence. I'm sure that if someone approached you in real life and asked you for help in suing a third party, one of the first things you would ask would be what factual basis they had for pursuing a claim. What makes allegations "credible" - the fact that some established editors have made them, or their basis in hard evidence? Let's recap:


 * No evidence has ever been presented that SEO was the motive for adding a template to the Santorum article, and the editor concerned has strongly denied this.
 * No evidence exists that adding a template had any effect on the article's search ranking.
 * No evidence exists that it's even possible to affect search ranking by adding templates.
 * No evidence exists that adding a template to the article was an abnormal action in any way. As others have pointed out, we're meant to make articles easy to find ("building the web").


 * The allegations were made by people campaigning to have the article deleted. As such, they had a vested interest in promoting claims of wrongdoing in order to discredit the article and the editor they were attacking. Don't you find it remarkable that given all the other mud that has been hurled in the course of these cases, nobody has stepped up to present any evidence whatsoever on the SEO issue? That suggests to me that the allegations were never based on evidence but were essentially a political weapon, used to gain leverage on the Santorum article and the template, and discarded as soon as they were no longer of use. Put bluntly, the claim of "SEO via templates" is the Wikipedia equivalent of "death panels". It's concerning that you seem to be indirectly lending your support to this claims. The fact that you are raising it as a principle at all gives it undeserved credibility and legitimises similar unsupported claims in the future. Will is right; this is going to cause more problems than it solves. Prioryman (talk) 08:06, 2 September 2011 (UTC)
 * Precedent would be against you. John lilburne (talk) 08:28, 2 September 2011 (UTC)
 * I don't think it should be any surprise that the SEO principle attracts attention - the other proposed principles are largely an affirmation of existing policy, without any obvious impact on one side or the other, but that one is a new mystery. I can't really think of anything on Wikipedia that we could do in a deliberate attempt to SEO that would much surpass the effect (if any) of Template:The Beatles (which at 57k of html source is already about as large as such templates can reasonably be made without slowing page loads to a crawl), nor can I think of a better motive for someone to do SEO here than to promote an intellectual property likely worth $100 million or more by ensuring that collectors are presented with a full list of songs that they can own.  It seems to me that if this template is far from being a concern, it should follow that nothing, and I mean absolutely nothing, should be regarded as violating this principle. Wnt (talk) 08:38, 2 September 2011 (UTC)
 * I've never been convinced about the alleged Google aspects of Cirt's activities. My experience is that Google tends to favour a site rather than individual pages, then new and modified pages over old unchanging pages. That a site's ranking is based on external links, and links to other sites. Internal links to a page simple make it more likely that a user or bot will find the page when looking at something else. IOW internal site links promote a page within the site, and that would be my assessment of template and DYK activity following the addition of excessive polish or tarnish to an article page. John lilburne (talk) 12:02, 2 September 2011 (UTC)
 * Has there ever been a known case of someone intentionally trying to enhance the SEO of a Wikipedia article? I've found many cases of this happening. If I would have known this case would have focused on the matter, I would have provided detailed evidence confirming this. Whether any of these attempts were successful, I don't know.  Them From  Space  14:38, 2 September 2011 (UTC)
 * You might as well present this evidence now or as you find convenient - it's all too clear that this issue hasn't been resolved here. I should mention that I suggested early on in discussions, and still support, a neutral bright line putting a limit on templates.  I think that limiting navboxes to linking 50 articles or less is more than fair, and limiting them to 25 is reasonable.  I think that any navbox too big to show in its entirety by default is too big to use in an article.  I think we can do a lot better with list articles to make them pretty like the templates, and I think with some dev work the fuddy-duddy old category system could be made just as pretty.  But what I don't accept is that we set a policy where adding "santorum" to a template is SEO because some people want the article censored, and adding a few Wikilinks pointing to the Breivik article is SEO because people hate him, and meanwhile there are no restrictions whatsoever about other articles that are liked or which are protected by an outside commercial interest. Wnt (talk) 20:30, 2 September 2011 (UTC)
 * Hmmm. Why not also pass a principle that "It is an extremely serious abuse of Wikipedia to utilize witchcraft in an attempt to inequitably or artificially manipulate search engine results. This is particularly so where the purpose is to disparage a living person."?  True, no credible evidence has been presented to show either that anyone has done such a thing, or that, if attempted, it would be effective.  However, those objections are irrelevant for some reason or other.  And surely, if such a thing were attempted, and worked, it would be pretty serious.  Moreover, if this proposed principle is passed, we can piously hope that good faith editing will only rarely trigger accusations of witchcraft.  Cardamon (talk) 07:22, 4 September 2011 (UTC)

When search engine optimization became a featured article it moved from fourth to first on Google Was this a bad thing? I still jokingly tell people that my article ranks first for SEO. Th punchline is "Yep, it's a Wikipeia article." SEOs generally hate Wikipedia becuse it ranks first for everything (exaggeration). When Google ranks pages, they are saying "these are the best". What's wrong with trying to make an article be the best? There is no problem with SEO. The problem is internal link spamming. Jehochman Talk 13:21, 4 September 2011 (UTC)

Going forward
In drafting this decision, it was my hope to emphasize, once again, the importance of applying the BLP policy appropriately and evenhandedly. In conjunction with the concurrent Cirt and Jayen466 decision, I believe we have emphasized the points we needed to make and we can close the case soon. I'd like to add a few concluding thoughts.

We has emphasized in the decision the strong suggestion that the feuding parties here should probably do their best to avoid each other for awhile. We have also emphasized that additional dispute resolution procedures, culminating in a new request for arbitration if necessary, are available if BLP-related disputes persist after the closing of the case.

I see that one of the parties to the case is already preparing a user-conduct RfC against another party. I have to admit that this is not what I had in mind as the immediate aftermath of the case. I would like to see dispute-resolution steps initiated promptly if there are future disputes involving an alleged pattern of BLP violations against one of the parties to this case. Such dispute resolution should, ideally, be initiated and resolved by editors who have not been involved in the argumentation to this point.

I do not think that an RfC commenced by one of the parties to this case against another with respect to another is the best path forward, either for resolving BLP problems or (a secondary but still important goal) for preserving a reasonably harmonious editing environment. That is not to say that such a course is forbidden by the decision, but it is not my own view as to how matters should best proceed. If there is a perceived pattern of BLP violations or issues involving a user going forward, they should be addressed promptly and civilly in the appropriate forumsm which of course may include an RfC. On the other hand, if there are no future problems or perceived future problems, I don't see much value to an RfC aimed at alleged past ones.

To anticipate a future question: if future problems (or alleged problems) do come about, then I'm not saying past conduct before this decision is off limits for discussion. (Although, if conduct has already been evaluated and found to be acceptable&mdash;I haven't gone back to check to what extent that's happened&mdash;that would be relevant as well.)

Of course, to the extent that there are BLP concerns with existing articles, they need to be addressed through the appropriate mechanisms as well&mdash;a separate issue from user conduct, though of course at times an related one.

Finally, and significantly, it may be best for some editors to consider taking a break from certain topics from time to time. If I were spending a significant amount of my editing time adding content about a particular group of people for whom I had little respect, I could see myself losing perspective after a while. And, the more negatively I viewed the group of people, the more likely I would be to lose that perspective; and some of the groups and organizations and individuals we cover on Wikipedia, because we aim to cover the gamut of notable subjects, are quite despicable. But note that this too is a voluntary suggestion for editors to think about, and a general one; I am not aiming this at any one editor or group of editors&mdash;and it is certainly not written for the purpose of having wiki-enemies throw it in one another's faces and I don't want to see any of that, please. Newyorkbrad (talk) 01:25, 2 September 2011 (UTC)
 * I have explained my rationale for starting the RfC in a post at WR. Cla68 (talk) 23:30, 2 September 2011 (UTC)
 * If it's relevant to on-Wikipedia discussions please post it on Wikipedia.   Will Beback    talk    00:03, 3 September 2011 (UTC)
 * Not what you had in mind? Brad, (insert chosen deity) knows you're not a stupid man, but sometimes I really have to wonder.  It's pretty simple: one side or the other is in the right and one is in the wrong.  To take the above example, either Cla86 is a pestering harassing machine, or Will Beback is a game-playing POV pusher.  You need to make a decision about which one.  Once you've done that, we can get on with business.  Until you do, you're going to get, depending on your point of view, a continuation of harassing RFCs or snide, tendentious, passive aggressive bullying. The reason?  If you refuse to make your decision, BOTH sides are going to conclude they are right.101.118.30.250 (talk) 02:50, 3 September 2011 (UTC)
 * This would have been an urgent problem ten years ago if Cla68 and Will Beback were the only two editors in this topic area. Because this isn't the case, it is possible to completely ignore what may have happened in the past, focus only on future problems, and let the dispute resolution process be handled by as of yet uninvolved editors. Count Iblis (talk) 02:59, 4 September 2011 (UTC)
 * That sounds good in theory, but in practice, new, as-of-yet-uninvolved editors show up and attempt to resolve these issues, only to be met with stubborn opposition on every point from Will Beback. Here are examples from recent article RfCs: Waalkes (talk) 18:40, 4 September 2011 (UTC)
 * Yet Cla68 is there too, despite his apparent lack of interest in other American political figures or movements. In one of those RfC you, Waalkes, agreed with me. Is it a violation of any Wikipedia policy, guideline or essay to take a different view in an RfC from the majority? It's incorrect to say that I stubbornly oppose every newcomer. For example, user:Scott MacDonald dropped by and expressed concerns which I addressed immediately. But I suggest we look forward rather than back.   Will Beback    talk    21:22, 4 September 2011 (UTC)
 * Unfortunately, looking forward may mean looking forward to more of the same. -- — Keithbob • Talk  • 16:48, 5 September 2011 (UTC)
 * Wikipedia isn't owned by Will Beback. If people don't like his (proposed) actions, he can be ignored/reverted. Count Iblis (talk) 20:52, 5 September 2011 (UTC)
 * Just because Will Beback disagreed a whole crew of parties to this case on that new thread doesn't mean that he did wrong in doing so. It only shows that the battle lines are still drawn, especially on the other side of that debate. Wnt (talk) 22:04, 5 September 2011 (UTC)