re banning people for suggesting Mantan's identity - We ban people for disruption and seeking to out the identity of pseudonymous WP accounts regardless of who is or is suspected of being on the other side of them. Mantan and Sammi have both denied publically and privately that they are the same person(s) or that either of them is Gary Weiss. In looking at this situation, I've considered at times that perhaps they were lying to us, that one or both accounts might be Gary, or that they're someone else but the same someone else. One would have to be an overly trusting idiot to never have the thought cross their mind, if they're paying attention to it. It's always come down to "They have different voices in talk / discussions" and "Even if one of them is Gary Weiss, are either of them doing anything which would be really bad if it is Weiss?". So, from time to time, I took a look at that. And always came back with "Well, it's a COI notification violation if one or both are him, but they seem to edit within policy otherwise, so it wouldn't be a big deal if it were". If we're going to go after people on the mere presumption that they're probably someone we have suspicions about... The list of accounts I've considered for the Duck Test for Bagley's group (basically, anyone editing his POV into [[Overstock.com]] or the related articles) and which pass a first level sniff test is immense. All of them are as credibly associated with that group as Mantan or Sammi are with Weiss. We can bring down the imperial might of bannination and be rid of both lots equally, sure. We could perma-freeze overstock.com and related articles, denying half of WRs users their levers to remain engaged with WP in the first place. Consensus has been "Leave them alone". I still disagree with the consensus that's developing that Sammi and Mantan are necessarily the same person. I've never seen an actual sock IDed that had that range of overlap and yet was consistently able to keep other edit patterns outside the overlap range separate. The timing analysis indicates that they're probably in the same time zone and edit in similar patterns during the day, but a wider analysis of NYC editors would probably turn up more, unrelated editors with similar patterns. Could it be? Yeah, sure. But we all knew that months ago. We didn't have proof then. We still don't have proof. Even if you assume the worst case, that they're the same person and that person is Weiss, the consequences in terms of sockpuppet policy violations are fairly minor. Those of you looking at that have come up with some edit patterns across months and months of edits that would be violations, including a couple of double votes I believe. That would fall into the "block the sock indef and short block and warn the main account" category. I and others would be dissapointed in Gary if it turns out that he's Mantan and/or Sammi, given that they've both assured people they aren't, but dissapointment is not actionable. We'd not trust them the same in the future, but they're not in positions of trust on the project anyways. People seem to have fallen for Bagley's "Big Lie" here, which is that if Mantan == Weiss, then this is a Big Deal, and the evil suppression of Overstock.com on Wikipedia all these years has been due to some sort of malign conspiracy. Even if Mantan is Weiss, that simply ain't so. Overstock.com's users are not welcome here because of truly gross and disgusting behavior, stalking and harrassment. Despite that we have a NPOV article on the company, the founder, etc. This was not an edit war in which sides were taken. Trying to simplify it to that is an insult to all the good Wikipedians who have been slandered and stalked and attacked over this. Bagley throws a good conspiracy theory story, but there isn't any beef. -george On Feb 13, 2008 1:34 AM, Cool Hand Luke <failure.to.communicate(a)gmail.com> wrote: -- -george william herbert george.herbert(a)gmail.com

On Feb 13, 2008 9:41 PM, Cool Hand Luke <failure.to.communicate(a)gmail.com> wrote: Note that even 1400+ edits over a year is still less than 4 edits per day on average. I don't know how broad your definition of "simultaneously" is (presumably you're using a metric like "within X minutes"). With frequencies that low you'll have trouble drawing reliable conclusions. -- Stephen Bain stephen.bain(a)gmail.com

On Feb 13, 2008 2:41 AM, Cool Hand Luke <failure.to.communicate(a)gmail.com> wrote: We're going to have to agree to disagree on the quality of the evidence. There's always been suggestive evidence that they're connected - but also suggestive evidence that they aren't. Your interpretation of the analysis you've done is suggestive, but I still object to insufficiently rigorous math and statistics and dataset sizes for comparison. Your evidence isn't qualitatively changing the situation, just quantitatively. A large mound of low-grade goo isn't better structural material than a small mound of it. Again - Even assuming the worst case, that these two accounts are the same person, and that person is Gary Weiss - the COI issues that would raise have never been persuasive to me. They just don't seem to edit in a problematic manner in general, only on controversial topics. Encouraging an investigation of everyone we suspect of something on Wikipedia, to avoid any COI issues, is tantamount to blowing the pseudonymity and user real ID protections we have and most people hold rather dear out of the water. We *do* block people for that. And we probably should. If you intend to challenge the generally pseudonymous operating mode, feel free to start with yourself, by letting us all know who you are and where you work. Then, we have a few million other users to convince to similarly identify themselves, some core policies on user privacy to junk and rewrite, and eventually around 2035 we can get back to working on the encyclopedia. I personally am perfectly fine with people using real names on Wikipedia, and I think it will improve the project. As you may note, I have used my real name online for ... oh dear lord, it's 20 years now. Other than the infrequent "edited without remembering to log in" from IP addresses that are publically registered to a company I admit freely to owning, I don't do any IP or sock edits, and am happy to have anyone who doubts that take a look at me. I have nothing to hide other than the fact that I am sadly overly heterosexual for a San Francisco Bay Area resident. Well, had nothing to hide. However... Forcibly outing people who fall into disfavor with our critics, however, seems like a short road to destroying the project. A large number of editors and admins I otherwise respect have happily run to do Bagley's bidding on this one, and that's highly disturbing. If we're going to enforce all the policies equally, then a number of people have met the policy definition of "acting as proxy of banned user" in this, in addition to attempts to reveal the real-life identity of a Wikipedia user in public. -- -george william herbert george.herbert(a)gmail.com

Jeez. I'm not talking about forcibly outing anyone or demanding verification. I modestly posit that fewer sock puppet abuses would be possible if user names were not more anonymous than "anonymous" IP addresses. Admins should be able to look them up, and block open proxies as appropriate. I'm already outed, thank you very much: http://72.14.205.104/search?q=cache:e6cvRAh0rzIJ:www.slweekly.com/article.c… As for the evidence, 27 and counting disagree with you, and the others who are dragging their feet even though they have long suspected a COI. http://en.wikipedia.org/wiki/Wikipedia:Requests_for_comment/Mantanmoreland#… Oh, and I already re-ran it as you suggested. Thanks for noticing. Of all 21 accounts compared, including 8 New Yorkers (as you suggested), I compared with the accounts most similar, having the same number of edits--one of them was one of the three best fits in the whole sample. They had crossed editing dozens of times on over 10 different days. Name me any account, and I'll compare them too, but nothing found so far matches as well as they do. And to re-iterate, the editing times are not themselves unique, but considering all of the common writing traits they share (which few other editors share), their common POV, article interests, and their history of past sockpuppetry, it's easy to see why 27 and counting disagree with you. On Feb 13, 2008 5:05 AM, George Herbert <george.herbert(a)gmail.com> wrote:

On 2/13/08, George Herbert <george.herbert(a)gmail.com> wrote: It is one thing for three people to each say "I'm certain that neither of the other two is me". But for each of them to go a step further, and say (referring to the other two names) "...and I'm certain that *he* is not *him* either" -- this would suggest that all three of them are more than casually acquainted. Otherwise, an observer would expect at least one of them, by this point, be skeptical as to the uniqueness of the other two individuals. This might seem a little complicated, so you might draw a little triangle, label it with three random names, and study it for a while. Put yourself, momentarily, in each person's shoes and ask yourself how many things you would actually be sure about. That's what I did anyway, though I have no way of determining whether one, two, or three people are now laughing at me for admitting this. (Or possibly none at all... *shrug*) —C.W.

On Feb 13, 2008 5:40 PM, David Gerard <dgerard(a)gmail.com> wrote: David, did you read the evidence? Nothing's ever conclusive in these issues but I, who have never edited these things or run into these people before, think that it would be difficult to file a more comprehensive indictment of a sock, and definitely this is the most efficient I remember. (There's always the probability that we're wrong, but the methods in this case have reduced it to considerably less than it has been before, and that probability never seemed to have bothered us much in this area, did it?) GWH, I'm all for ensuring that similarities are checked with the appropriate control set, but the bit on "interleaving" in the timestamp statistics comparison is pretty damn convincing, and independent of time- or area- specific control sets. RR

On Feb 13, 2008 11:37 PM, Nathan <nawrich(a)gmail.com> wrote: I don't really care about the product, I care about the fact that this has pissed off so many people that the little battles they've fought over it have spilled all over the project and made everyone else's time miserable. Somebody needs to ensure that doesn't happen again. Err, you'll have to expand on that. I have some fairly advanced statistical training, and I see several problems with the collinearity here, but its a pretty good rough reckoner, and - as I said - light years ahead of anything I'd seen before. Definitely, 10 data points is too small, but in the interleaving comparison in particular, the difference is stark enough that I don't think the degrees of freedom matter. And what do you mean by "serious approach"? RR

As far as I know, no one has ever thought to do what I've done here. The SevenOfDiamonds case was happy just to list similar traits and plot a similar graph of editing patterns between the users. That wasn't enough for me: I wanted to know how meaningful similar editing patterns and traits are. So I did surveys with data as random as I could, and no one else is even remotely similar. You would have that count against the data. GWH protested that I didn't include enough similarly situated (ie, presumably New Yorker) editors. So I grabbed some of them too, and still nothing is as close as these two. It seems that some editors would not be satisfied until every last Wikipedian is compared. That simply isn't necessary. Their editing patterns match well, and are rare (probably less than 1-in-21 based on what I've looked at so far, and that's while I'm trying to find users who will match). Many of their editing traits are shared by almost no other accounts (1-in-10 or less for at least a half dozen or so traits). If these variables are even somewhat independent, they multiply together into very long odds. And the interleaving is really compelling based on the editing style. GWH protested that my sample users don't resemble their editing patterns well enough (as indeed no accounts do), but by comparing an editor who makes a significant proportion of their edits while Mantanmoreland doesn't edit, I would have expected to find LESS intersections, not more. These users never edited at the same time as each other, and this is at least eyebrow-raising. All of these things together are damning--and that's without even keeping our eye on the ball--that these users shared POV, that Samiharris started editing Wiess within one day of Mantanmoreland quit, and that Mantanmoreland knew his edits would be monitored, so had a motive to spin a sock for possible COI abuse. Quack. Cool Hand Luke

I'm really, Really, REALLY getting sick of all the drama, so wouldn't it solve matters one and for all if Mantamoreland and Samiharris simply didn't edit those articles? The easiest way to get rid of a COI issue is for them to recuse themselves; then we can forgo this never-ending inquisition.

On Feb 13, 2008 11:43 AM, Cool Hand Luke <failure.to.communicate(a)gmail.com> wrote: Some points - 1. A lot of the statistical stuff still needs a lot of work. The section 13 analysis really does need re-running with editors who are whole-day better matches to the edit pattern, for example. You and others brushed this off - that's not a reasonable response. At this point section 13 is faulty. 2. *All* of the statistical stuff, particularly the user phrase analysis, should be flipped around and run the other way to generate anything like a solid profile. If user X uses phrases A, B, and C a lot, and user Y uses phrases A, B, and C a lot, then they're similar in that sense. But to understand the level of confidence in their similarity, one then should flip it around and analyze the database contents, and find users who use phrases A, B, and C a lot, and then see if the number and distribution of users implies that the similarity in that aspect is truly uncommon or not uncommon. It's been baldly asserted by a number of people that "Of course" similarities imply linkage - the reverse analysis shows how likely the linkage is to be causal as opposed to merely statistical. It may well be that 5% of total users fit profile of "uses A, B, and C", and that we would statistically *expect* that out of (these numbers representative but made up) about 1 million US users, we would find about say 3,000 in New York City, of whom about 150 would also use phrases A, B, and C, and about say 25 of whom would share a similar editing time profile. Given a whole US population, the odds would then be high that somewhere, in some metro area, are two factually unrelated people who use a given set of similar phrase patterns, the same edit time of day profile, and some overlapping article interests. Those are made up numbers - however, they might well be pretty close to reality (or not - I haven't run them, and statistical guesses are a really bad analytical tool 8-). 3. This whole incident has several aspects which are very unlike previous sockpuppet analysis I am aware of. In a sense, this is good - we're starting to develop some tools to determine analytically stuff that previously has been gut feelings. It's also showing that we have some disagreement as to the validity of the statistical methods, and experience with statistics and analysis. In my opinion, having done statistics in college, in workplaces, and in scientific analysis from time to time, the investigative methods used so far are applicable but have not been applied with sufficient statistical depth and rigor... Yet. I don't see any reason why people can't perform the rest of the research to answer these other questions, and I think that we can eventually reach the point that my doubts are answered from a truly statistical likelyhood sense of "proving the case". I believe that IF you intend to use statistical methods to try and prove the case, the burden of proof should be on those "prosecuting" the case to convince skeptics that the statistical analysis has gotten good enough. I am not yet convinced, from a statistical sense. There are too many unknowns, and the rigor of the work isn't there yet. We don't yet have agreement on what level of rigor and testing is required. But I think we can find that agreement. 4. All of that said, through the haze, there's a sense that a duck has been sighted. How we handle a combination of duck test feeling and statistical findings to reach a final conclusion is up to arbcom and community consensus. -- -george william herbert george.herbert(a)gmail.com