I have posted some questions on my talk page, and should be grateful for some answers. If you don't want to speak publicly, there are a number of people who can pass messages along.

I would like to understand how this has happened. In particular, I am concerned at the privacy implications of "reports" being compiled on editors and circulated amongst closed lists of "senior editors". We have an ArbCom and a limited list of editors with checkuser and oversight privileges for a reason, and this seems to cut right across it. -- !! ?? 00:23, 19 November 2007 (UTC)[reply]

I've hard blocked the above IP address, per instructions from Dmcdevit, because it is a Tor node. There is a strong possibility that it is posted by a banned editor. - Jehochman ^Talk 00:33, 19 November 2007 (UTC)[reply]

Yes, and I've deleted the post per WP:BLOCK and WP:DUCK. I've no doubt that banned sockpuppeteers would love to see me give up sock investigations. They've forfeited the right to post those opinions here, though, and (if anything) the attempts amount to an endorsement of my overall work. Thanks for the block, Jehochman. And I'll reply to !! right away. Durova^Charge! 00:37, 19 November 2007 (UTC)[reply]

I'm half tempted to unprotect this talk page. It's becoming a good honeypot for TOR nodes. Durova^Charge! 01:12, 19 November 2007 (UTC)[reply]

Hi Durova. I appologise in advance for not doing this in an email, I just don't think it's the correct venue given that today's events all seem to be surrounded by things happening off wiki. What I want to do is make a couple of suggestions to you when conducting things like this in the future. The only reason that you wouldn't release any information about the block is that you feared your techniques of finding socks would be found out, I really don't believe this is a very good excuse for keeping things from the community, nor do I believe it is a good reason for only ArbCom to scrutinise the block. I think you've stated that in the future you will pass the information to the arbitration committee - well how about you actually post this on-wiki when there aren't any privacy concerns? There were plenty of people who wanted to take a closer look at this block but for obvious reasons, were unable to do so.There was no need whatsoever for this secrecy. I also see no explanation as to how the mistake happened. I saw a two line appology and little commentary as to how you arrived at the wrong conclusion. In cases such as this, I would suggest you owe the community an explanation for why a respected editor was blocked on what obviously must have been rather dubious evidence with only a slective few being able to view it. My final concern here is that your unblock message to !! was poor to say the least. I don't believe "false positive" gives any true indication as to the scale of your mistake here and doesn't really clear !! of any wrongdoing. I would expect that in the future, you would make more effort to clear the name of the condemned in logs which are unremovable. I'm sorry if this sounds like I'm having a go at you, it's not supposed to in any shape or form, I'm just a little concerned at how things were handled today. Ryan Postlethwaite 01:14, 19 November 2007 (UTC)[reply]

Ryan, I respect your thoughtful response. However I disgree quite strongly with the supposition that there's no need for secrecy here. Nearly everyone who's seen my sock investigations agrees that some of the methodologies are sensitive and should be kept confidential, and among people who know my work best I tend to get advice that I err on the side of openness. If I wanted to make this all much simpler I'd abandon the I'll tell as much as I can and take the rest offsite approach and then, in a large measure, I'd evade this criticism because no one would be the wiser. I don't think that would be a good solution because I do appreciate intelligent feedback and challenges. At any rate, my approach in the future is to route this directly to ArbCom, and if I happen to be on the Committee I'll let another member act as demonstration that I'm not trying to be the Lone Ranger. This site does have a problem with banned editors who engage in long term sockpuppetry. These people operate as a team and share strategies. It takes a lot of work to uncover them. I've expanded upon my earlier apology at !!'s talk page. If the ones I previously offered look hastily written, they were. My principal aim when I wrote them was to correct the problem and clear the air as swiftly as possible, once I confirmed the new evidence that had come to my attention. My goal was to minimize the harm I had inadvertently caused and to restore people's confidence. I was fielding quite a few queries at once and didn't want to keep anyone waiting. So the first apologies were sincere and simple. In comparison to other mistaken blocks that have occurred at this site, few administrators have been as swift in correcting their own error, as forthcoming with apologies, or as heavily criticized for the effort. If there's something more I can do to set things right I'll readily do so, but full disclosure of my investigative techniques would cause far more harm than good. Durova^Charge! 01:34, 19 November 2007 (UTC)[reply]

I see the predicament here... On one hand we know the harm done by these SPs that go unnoticed and cause massive damage to hundreds of articles before we they get caught as well as SP gangs that know WP well enough to game the system and undermine the project; on the other hand we have people like you that have the time, patience and skill to unravel these SPs' shenanigans and get them blocked. The tension between the need and the method, and the eventual "false positives" makes this a hard call. IMO, bringing the evidence to the ArbCom will do no good, as they are usually busy and will be unlikely able to review the evidence in detail. You said you shared your evidence with a dozen trusted members of the community, but did you get their support for a block, or just a no reply? ≈ jossi ≈ (talk) 01:55, 19 November 2007 (UTC)[reply]

Yes, that's pretty much the predicament. I got about 5 responses from editors I trust. None of the ones who discussed it in detail were arbitrators or checkusers, that I recall. The original account of this user seems to have been an open secret in some circles, so in retrospect I'm still surprised that a cursory glance hadn't sent up any red flags. Chalk that up to experience, and something I'll work harder to handle with discretion and respect in future. Durova^Charge! 03:38, 19 November 2007 (UTC)[reply]

I do respect you sorting the problem out as soon as it came to your attention that the evidence wasn't quite right, and I fully understand why the apologies may have appeared short as you were actively trying to fix the situation. I think the problem here is we obviously have two very different schools of thought about what things should and shouldn't be discussed on-wiki. I strongly believe that we should only resort to offering private evidence to ArbCom when we really have no other option (we should always attempt to discuss issues ourselves) - I don't think this was a situation that required the level of privacy that you gave it. I agree that we have an obligation to look out for banned editors and do everything we can to remove them from the project, I just think that when a respected user is accused of such an offense, we should look to discuss the issues in house and offer the opportunity for the community to review such actions. I do think it's a good idea for you to refer future evidence like this to the committee first before blocking as it would most likely stop the drama which has followed in this instance. I know you do a lot of hard work in your investigations and more often than not come up with the right answer, but please try not to be complacent - there's plenty of trusted people willing to review evidence here, and I think now more than ever we would expect you to seek more advice than you did this time before blocking. Having said all of this, I want to make it clear that I still have a lot of respect for you, and although I think you made a serious error of judgment with this one, you obviously do a lot of hard work for the project. Ryan Postlethwaite 02:42, 19 November 2007 (UTC)[reply]

Pardon if you've posted this elsewhere, but it would be useful to know who you sent this "report" to and who responded positively that the user should be blocked on the basis of that information. Or at least, the numbers in each group. It seems clear to me that providing this information would expose nothing about your methods not risk exposing any private information. If you're uncomfortable about releasing this information it would be useful to know why you feel this needs to be kept secret. Christopher Parham (talk) 02:44, 19 November 2007 (UTC)[reply]

You said "given the amount of time my report circulated and the people who had access to it ..." It has been my experience that most arbcom members and most others do not actually investigate much or even read all the evidence; but as this is an unpaid time drain, rely mostly on the opinions of trusted others. This can produce a situation where everyone thinks someone else checked something out. I had a situation decades ago where a secret government report was being "verified" by three separate US government agencies and the question was raised by someone about to act on the intelligence assessment "Is this three separate verifications or are they all relying on a single classified source?" After an investigation it turned out that all three were indeed relying on a single classified questionable source that had been judged reliable solely because three separate agencies were reporting it (but because the source was secret, none of the three revealed that source - a "need to know" categorization issue even though all involved had top secret security clearance). I hope you can see the issues and problems involved here. WAS 4.250 (talk) 02:54, 19 November 2007 (UTC)[reply]

To ask one additional question, I think it's worth noting that without you ever revealing the information that led you to make your report, as soon as the block was opened for public discussion evidence was rapidly brought forward to rebut your claims. Would it be possible in the future to open your suspicions for public discussion before making a block? That is, instead of saying "I have blocked..." what about "I intend to block.... Is this idea ludicrous?" This would certainly have prevented this incident, and again, there would be no need to reveal your proprietary investigative techniques. Christopher Parham (talk) 02:59, 19 November 2007 (UTC)[reply]

(edit conflicted)

It's very difficult to convey this in purely theoretical terms. I've got a trusted-user version of the evidence I submitted to the Alkivar case. Maybe that will help clear up the basic issues about confidentiality: everyone who's seen it agreed that it was substantive and that it needed to be handled with care. The specific report I circulated for this case was shorter: about 2 printed pages with 28 diffs and a 7 point methodology. Obviously that second report wasn't as good as I thought it was, since it swept up a false positive. So I'd like to study that more to refine the techniques rather than circulate it further. I'd rather not say specifically who I circulated it to, and ultimately the responsibility rests with me for not following up better than I did. I got roughly 5 responses, all positive. That seemed like reasonable diligence when I acted this morning. Now I have to call that another lesson learned.

The main reason why I handle certain evidence offsite is because I know my edits get watched by precisely the people I work so hard to foil. More than 90% of the material I use is public information, but my methods of parsing it seem to be better than most. It's those methods I want to keep out of the hands of the people who try to exploit this site, because they'd get better at sockpuppeteering, proxy editing, and general exploitation if they could. Again, it's easier to convey this with reference to specific examples. In fact, once a trusted editor sees the specifics they usually understand the rest without any more explanation from me. Durova^Charge! 03:05, 19 November 2007 (UTC)[reply]

Actually, Christopher, that's an interesting point. I've pledged to route this through ArbCom in the future in order to assure proper double checking. Your solution might work also. It's something to chew on, at least. I've always been uncomfortable with the community banning formulation of, indef first, discuss afterward. Since I've made a public pledge I'll uphold it. I'd also be willing to discuss this alternative solution, at least theoretically. These are difficult blocks to implement. Some of the hardest of all, in fact, since when they're right they're non-obvious. And the non-obvious part is highly exploitable if it gets discussed onsite. I've been looking for a way out of that Catch-22. Durova^Charge! 03:09, 19 November 2007 (UTC)[reply]

The fact that 6 people were fooled in this case doesn't make me optimistic that 15 (minus whatever Arbs are inactive) won't be fooled in a future case; contacting ArbCom on a private basis, in my view, isn't a meaningful improvement compared to your current process of due diligence.

In this case, the community was able to provide you some new evidence in less than an hour that convinced you to change your mind. I think that opening your suspicions (but not your evidence) for community discussion before blocking would materially improve the success rate of your investigations without revealing anything about your methods. The community doesn't and shouldn't have the final say when there is significant private evidence. But excluding the community's wealth of information serves only to make it less effective. Ultimately, the long-term success of your work depends on your blocks being correct, so I think it would behoove you to open your conclusions for community comment before acting on them. Christopher Parham (talk) 03:32, 19 November 2007 (UTC)[reply]

That's the kind of assumption that I think a more hands-on look would settle. My evidence here was about 2 pages long and 28 diffs. My evidence to the Alkivar case was originally submitted in segments. The Alkivar portion itself had been about 6 pages in the text document with several times that number of diffs, distilled from about 30 pages of text notes and diffs. I had a separate presentation of about the same size and depth for Burntsauce. I really don't want to say any more than that onsite. If you're earnestly curious I'll see (and respond to) that curiosity at my inbox. Durova^Charge! 04:03, 19 November 2007 (UTC)[reply]

Sorry, I don't see how that's a response to what I wrote. To what assumption are you referring? The extent of the research you provided in this or other cases is not relevant, nor are the contents of your inbox. The question is whether or not you intend, in future investigations, to refer blocks to the community before implementing them. There can be no doubt that had you done that in this case an error with potentially serious consequences would have been avoided. Nor is it reasonable to believe that you will ever, as a single individual, be able to investigate so thoroughly that the 1000+ member community could not add any potentially important information to inform your conclusions. I don't really see a downside here and I think there is significant upside in terms of your success rate in making these blocks. Christopher Parham (talk) 04:15, 19 November 2007 (UTC)[reply]

The pledge I've given is to route this type of thing through ArbCom in the future and let them examine it and act. If I happen to be on the Committee I'll route it through formally and let another arbitrator act. That way the community will have reasonable confidence that my evidence has undergone proper scrutiny. In my experience ArbCom is very good at distinguishing good evidence from flawed evidence. Most people seem to be satisfied by that pledge. Do you have a better suggestion? In addition, I'll be tweaking my methodology in light of this and I'll be much more careful about making sure a checkuser actually gets run. My report was in the hands of some people who had that ability, but I wasn't actually certain whether any of them had checked it out that way. It's my responsibility that I didn't follow up on that explicitly, so I'll learn and improve from that. Durova^Charge! 04:25, 19 November 2007 (UTC)[reply]

I've already suggested what I view as a better suggestion: asking the community for input before making a block (without revealing any of the information you wish to keep private -- all you really need to say is the name of the editor you are suspicious about). Again, I just don't see a downside to this. It's also a constructive step to ensure that checkuser evidence is also introduced into your decision-making. But I don't think this will eliminate the capacity of the community to add fruitfully to the process. Christopher Parham (talk) 04:37, 19 November 2007 (UTC)[reply]

That's a very sensible idea. It may take a little while to wrap my head around how I'd square that with the pledge I made about routing this stuff through ArbCom. And in general, I do like to discuss things proactively. As a challenge here, what would you say about the very legitimate concern this editor has about potential damage to his reputation? I wouldn't want to raise clouds of suspicion or drama unnecessarily, and as I think about this I'm uneasy about posing this kind of question to people who won't be seeing the best evidence I've got. They can't view the query in context. Durova^Charge! 05:04, 19 November 2007 (UTC)[reply]

In my view, this would be something to do just before blocking, after taking your private investigation as far as it can go and deciding that on that basis you would block. So there's zero additional risk of damaging reputation -- surely an announcement of suspicion, swiftly clarified, is better than an indefinite block, swiftly reversed. At least, it does not appear forever on the block record. I think there would be a meaningful net gain in the drama department -- historically, inappropriate blocks of established contributors have been the single biggest cause of drama and avoiding them is probably the most important consideration in improving your process.

As for the fact that the public won't have access to the private information you have, it will still be up to you to take all that information and decide whether or not to block. I view you taking things to the community as more of an information-gathering exercise than a search for consensus. The community doesn't have a final say on blocks that reference private information; if nobody can provide information that satisfies your concerns, you can still block subject to ArbCom review. Christopher Parham (talk) 05:21, 19 November 2007 (UTC)[reply]

Something like that has worked pretty well in reverse. I have a standing offer to community banned editors that I'll discuss an unblock with them after six months if they do some simple things such as respect WP:SOCK. When someone approaches me about that I open a thread at WP:AN and see whether anyone has an objection to restoring their editing privileges. Mind if I chew on this one? Durova^Charge! 05:33, 19 November 2007 (UTC)[reply]

Obviously the decision is up to you. I don't think it needs to be an explicit decision, nor need it be applied in every situation depending on the severity of the evidence in question. Christopher Parham (talk) 05:42, 19 November 2007 (UTC)[reply]

To avoid perpetuating any errors of fact, am I correct in saying that your block message was your first attempt to contact User:!!? Christopher Parham (talk) 03:43, 20 November 2007 (UTC)[reply]

Yes, and that was my error. I'm quite sorry for it. Durova^Charge! 03:45, 20 November 2007 (UTC)[reply]

Hi Durova. Just a quickie about your blocks of these Tor nodes - 24.242.78.69 (talk · contribs) and 134.48.216.41 (talk · contribs). I presume you meant those to be hardblocks given present open proxy policy, but the software doesn't seem to be allowing me to unblock them in order to reblock. Could you have a look and see if it'll work for you? WjBscribe 02:58, 19 November 2007 (UTC)[reply]

Dmcdevit has instructed me to hard block Tor nodes for 5 years and add {{openproxy}} to the talk page as the block message. - Jehochman ^Talk 03:05, 19 November 2007 (UTC)[reply]

Actually I've been working at high speed all day. I was looking for a policy page on open proxies, but didn't find one in my haste, so I simply implemented standard 1 year blocks. There was another such proxy I blocked at this talk page before I semiprotected it. If either of you could help out with that I'd be much obliged. If there's a silver lining to this drama, it's that it became a rather good honeypot for TOR nodes. Durova^Charge! 03:12, 19 November 2007 (UTC)[reply]

No problem with helping, except that I get an error message whenever I try to unblock those IPs so I can reblock them. Dunno if one of you will have more luck. WjBscribe 03:14, 19 November 2007 (UTC)[reply]

Popping in to say that I can't unblock/reblock them either. MediaWiki claims that the "block ID" cannot be found. Strange. — TKD::Talk 03:17, 19 November 2007 (UTC)[reply]

I see that WJBscribe finally got the reblock to go through, as I was writing my message. — TKD::Talk 03:19, 19 November 2007 (UTC)[reply]

I managed to do it, for some reason the software seemed to be treating them as autoblocks rather than standard IP blocks... WjBscribe 03:20, 19 November 2007 (UTC)[reply]

Thanks all. Thank you very much. Durova^Charge! 03:22, 19 November 2007 (UTC)[reply]

Autoblocks. Let's investigate further. What would trigger autoblocks? Could those be triggered by a blocked user attempting to login through those nodes? - Jehochman ^Talk 03:25, 19 November 2007 (UTC)[reply]

That is what causes autoblocks - but they would have to be recent blocks (within a day or two) and there don't seem to any blocks that correspond with the autoblock IDs I unblocked - #697367 and #697368 (see autoblock finding tool). Autoblocks are separate from those on the IP itself - the presence of both doesn't usually cause this sort of thing. WjBscribe 03:32, 19 November 2007 (UTC)[reply]

Jimbo wants Tor soft-blocked, in general, and "present open proxy policy" is not prescriptive of hard or soft blocks. Wikipedia talk:Open proxies reached consensus for soft-blocking but not for hard. Consistent with the policy statement that "while this may affect legitimate users, they are not the intended targets", we are telling Tor users to send email requesting account creation, so we can't be hard-blocking every Tor node. ··coelacan 03:50, 19 November 2007 (UTC)[reply]

So what's the correct procedure here? Soft block for 5 years? I'd like to be clear on this, please. Durova^Charge! 03:54, 19 November 2007 (UTC)[reply]

Jimbo is by his own admission in a minority on this one. Softblocking Tor nodes is a bad idea - accounts editing from softblocked IPs don't get autoblocked. So a sockpuppeteer can easily take advantage of a softblocked Tor node to get maximum mileage out of his socks knowing that one of them getting blocked will have no effect on the next. The real way forwards in my opinion is for a consensus to develop for us to allocate special:IPblockexempt (presently assigned only as part of the sysop user right) to be able to be individually allocated to good users who wish to edit from a Tor address, so they are unaffected by the hardblock. WjBscribe 04:01, 19 November 2007 (UTC)[reply]

Is there a short and practical answer here? I'm basically looking for an in-the-field solution to That disruptive post originates from a TOR node. So how can I close that loophole quickly without generating additional drama? Durova^Charge! 04:07, 19 November 2007 (UTC)[reply]

My reading of the open proxy policy talk page shows neither side as a clear minority. I say softblock for 5 years, with account creation disabled. This remains controversial, because consensus for hard-blocking was never reached, and some admins do it anyway. A significant set of soft-blockers would probably become hard-blockers if bugzilla:9862 was finally implemented. In the meantime there is no universally agreed-upon procedure. I say soft. Dmcdevit says hard. Jimbo says soft. Discussion trailed off a couple of months ago and it's all unresolved. The particular loophole you mention will be sufficiently plugged with a soft block disabling account creation. ··coelacan 04:10, 19 November 2007 (UTC)[reply]

I disagree that disabling account creation is sufficient - it presumes the sockpuppeteer wouldn't already have sleeper socks. Experience teaches us to presume the opposite I think. Durova, I recommend hardblocks and I believe the majority favour that approach, Coelacan disagrees both with that advice and my assessment of consensus. You aren't going to get a concrete answer on this I'm afraid. WjBscribe 04:36, 19 November 2007 (UTC)[reply]

Geez, the whole topic seems like a minefield. Should we work toward a consensus on this one? Seems like anything I do could be raised against me by a politically motivated editor who's determined to assume bad faith. Durova^Charge! 04:27, 19 November 2007 (UTC)[reply]

Consensus may be hard to find, I believe the options are:

1. No new features - softblock Tor
2. No new features - hardblock Tor
3. Have a new type of softblock that doesn't make accounts using the IP imune from autoblocks - use that for Tor
4. Make special:IPblockexempt individually assignable - hardblock Tor
5. Withdraw special:IPblockexempt from admins - hardblock Tor

Option 3 still requires developer work. As to Option 4, the mediawiki extension for it has been written it seems - I think it would just need consensus. But I think there will be quite a range of views as to which is the best option. WjBscribe 04:36, 19 November 2007 (UTC)[reply]

(ec) Durova, anything any of us do is subject to that kind of criticism. ;) I think it's safe to say that whether you pick hard or soft blocks, admins are going to defend you. Nearly everyone is supportive of softblocks at least. I have technical reasons for recommending a limitation on hardblocking, but I'm not going to scream about it. I agree with WJB that preferable to reopening policy discussion is to push for ipblock-exempt as a technical measure. ··coelacan 04:37, 19 November 2007 (UTC)[reply]

Thanks. I'll do my best to stay within the realms of what's reasonable and customary. Durova^Charge! 05:07, 19 November 2007 (UTC)[reply]

You wish to improve your investigative methods. Society has a long history of trying to perfect its ability to ascertain right/wrong truth/fiction. Logical argument based on evidence between opposing sides has been established by science and modern society as the best that can be achieved. The opposite of that is a one-sided star-chamber evaluation where group-think reigns unopposed. WAS 4.250 (talk) 07:18, 19 November 2007 (UTC)[reply]

Your investigative method lacks a Devil's advocate. WAS 4.250 (talk) 07:20, 19 November 2007 (UTC)[reply]

Where do you get these ideas about my investigative methods? Durova^Charge! 08:08, 19 November 2007 (UTC)[reply]

You say that your response to the failure of this strategy:

I have placed an indefinite block on this account as a disruptive sockpuppet. Due to the nature of this investigation, our normal open discussion isn't really feasible. Please take to arbitration if you disagree with this decision. Thank you. DurovaCharge! 16:48, 18 November 2007 (UTC)[reply]

is this strategy:

Iamunknown, I've already pledged to route future investigations of this type through [arbcom] and let them handle it. Or, if elected, I'll let another arbitrator act upon my investigations as proof to the community that I'm not being the lone ranger. Incidentally, TOR nodes keep posting cricitisms of my actions to this thread. It's become a rather good honeypot for that purpose. ;) DurovaCharge! 01:38, 19 November 2007 (UTC)

which, in my opinion, lacks the free Devil's advocate assets of this strategy:

I've already suggested what I view as a better suggestion: asking the community for input before making a block (without revealing any of the information you wish to keep private -- all you really need to say is the name of the editor you are suspicious about). Again, I just don't see a downside to this. It's also a constructive step to ensure that checkuser evidence is also introduced into your decision-making. But I don't think this will eliminate the capacity of the community to add fruitfully to the process. Christopher Parham (talk) 04:37, 19 November 2007 (UTC)

WAS 4.250 (talk) 09:59, 19 November 2007 (UTC)[reply]

I see what you're driving at. In practice, one reason I've opened my investigations to any trustworthy Wikipedian is to encourage precisely that kind of independent feedback. The basis I've used for responding to queries has nothing to do with whether the editor normally agrees with me and everything to do with whether I think they'll handle the information with due discretion. There is, of course, some potential for inadvertent selection bias in that approach. To the extent that it's feasible this semi-open method counteracts the tendency you warn about and I accept a level of risk with the information in order to maintain that degree of checks and balances on my research. The downside of this approach is that it also exposes me to a high degree of criticism. Some people expect all their questions to be answered onsite, so in good faith I explain repeatedly why there are limits to that. Then another set of people jump to conclusions and suppose I haven't used one or more rather basic quality checks. So that leads to a different set of conversations where I do my best to reassure people that those bases are already covered, without tipping my hand too much for the people who follow these discussions in the hope of gleaning information to become better sockpuppeteers. I know there's an active interest because, among other things, both this talk page and ANI had to be semiprotected today to seal off disruption from TOR nodes.

So I weigh this advice to "introduce" checkuser evidence to my decision-making against the fact that it's been part of my procedure for a long time, not used universally but perhaps worth it all the time as a good just-to-be-sure thing, and also useless against the folks who are using these TOR nodes. I'm more than just a step or two ahead of that level of advice. The people I've been foiling know I'm often ahead of them too, farther ahead than I let on, and they're trying like mad to figure out where I am and how I got here. I make no pretense of being perfect, which is why I invite criticism. Often it arrives in redundant or superfluous form. Durova^Charge! 10:32, 19 November 2007 (UTC)[reply]

I feel that I'm not being clear, so let me be blunt: Before blocking a username based on complex analysis, secret or not, say "I'm planning on blocking [so and so]" at AN and give it a day to a week depending on the response. Give as much extra data as you feel comfortable with knowing that if you give too little, others may unblock. But the idea is for devil's advocates to have their day. You don't get that by asking angels. Your friends will pat you on the back, thinking "I trust her", and tell you what you want to hear. Friends of so and so will speak up with truths and/or lies; but they will give you data others don't have and can't give. It is not about others evaluating your data; it is about others adding data you don't have and can't get from your friends because they don't know it either. WAS 4.250 (talk) 12:30, 19 November 2007 (UTC)[reply]

And I'll be equally blunt: look at the drama on ANI right now. ArbCom has an excellent record of parsing formal evidence efficiently. That's the solution I've pledged to adopt. I'm rather surprised by the insinuations you continue to make that some coterie of friends rubber stamps this research. It's a rather inappropriate assumption to make in the first place, and to persist at this point is clearly the wrong side of WP:AGF. Durova^Charge! 12:42, 19 November 2007 (UTC)[reply]

Thanks for being blunt. I had no idea you were assuming bad faith in my trying to help you. I'll walk carefully away from this discussion now. Good luck to you. WAS 4.250 (talk) 14:23, 19 November 2007 (UTC)[reply]

Replied at Was's talk. Durova^Charge! 17:36, 19 November 2007 (UTC)[reply]

Do you really think you're the best person to be doing this? Please stop blanking edits on ANI for whatever reason: you are in enough controversy as it is, and you are only adding to the drama. If edits need be removed, we have more than enough admins with ANI on their watchlists to do it. Physchim62 (talk) 12:24, 19 November 2007 (UTC)[reply]

Yes, these actions are appropriate. Per WP:BLOCK, posts by blocked or banned editors may be removed by any editor. Most of the IP posts I removed admitted that they were ban-evading posts and originated from TOR nodes. I also removed posts by two sockpuppets of Amorrow, who is the most thoroughly banned former editor. A lot of the sysops who know what he is would be afraid to intervene. Durova^Charge! 12:33, 19 November 2007 (UTC)[reply]

You should reread WP:BLOCK: it does not allow for reversion "on sight", only WP:BAN allows that. These posts should not have been reverted unless they were disruptive, and in no case should they have been reverted by your good self! Don't you realize that the whole point of the discussion is that many users, including several admins, appear to have lost faith in your application of your administrative discretion as regards sockpuppetry? A little bit of self-criticism would go a long way: otherwise I will be quite happy to supply you with industrial excavation equipment if you really wish to enlarge the hole that you find yourself in. Physchim62 (talk) 13:39, 19 November 2007 (UTC)[reply]

Yes, a little bit of WP:AGF would go a long way. Why this combative tone? Durova^Charge! 16:46, 19 November 2007 (UTC)[reply]

Hey, Physchim62, take it easy, would you? No need to increase the drama here.≈ jossi ≈ (talk) 17:19, 19 November 2007 (UTC)[reply]

WP:AGF is yet another of the Wikipedia policies which you would do well to read, as you have not applied it in several recent cases. Physchim62 (talk) 18:15, 19 November 2007 (UTC)[reply]

Says the fellow who has insinuated that other admins are taking bribes, with no supporting evidence. Raymond Arritt (talk) 18:22, 19 November 2007 (UTC)[reply]

I did not say they were "taking bribes", and I had more evidence than Durova has been willing to share with the WP Community in many of her allegations over the last couple of months. Physchim62 (talk) 18:28, 19 November 2007 (UTC)[reply]

That's true; you merely suggested that they were "selling their services as Wikipedia admins."[1] There's a huge difference. </sarcasm off> Raymond Arritt (talk) 18:34, 19 November 2007 (UTC)[reply]

Please, let's spread the wikilove. This conversation grows out of a thread where some individuals have obviously done their best to sow dissention. Let's rob those people of the pleasure of success. Physchim62, I trust that you came here in good faith. The kind of sysop work you usually do is very different from the kind I usually do. Once in a while, in all likelihood, each of us makes a mistake. I trust that you know far more about name redirects for chemistry articles than I do, even if one of your decisions is mistaken and you correct yourself 75 minutes later. Yesterday one editor got blocked for 75 minutes and I corrected myself. Minus drama, that's all that really happened. Physchim62, there's an adorable photograph of a stuffed elephant on your userpage with a link to the Wikipedia:No angry mastodons essay. I laughed with pleasure the first time I saw that photo and I'm very flattered that you enjoyed a list of reminders I wrote to myself about how to handle difficult moments onsite. That essay was made for moments like this one. Let's put it to use. With respectful regards, Durova^Charge! 18:51, 19 November 2007 (UTC)[reply]

[reset tabs] I'm glad you've met my political advisors, let's see what they say:

Jote, my expert on socks, believes that you are the alternate account of another well-known Wikipedia editor. Unfortunately, I can't tell you why. Jote says that he will only share the information with certain trusted users. Let's be glad that he doesn't actually have a block button (and that it's difficult to type when you're a stuffed donkey).

Harry, the Large Wet Haddock™ who oversees my own administrative actions and those of others, cannot understand why you haven't resigned your sysop bit yet, given that you are open to recall and that the unease over your actions is so great. He would like to know what if you under what circumstances you would be open to recall, as he believes that you have simply placed the category link on your user page for decoration.

Jumbo, my pet angry mastodon, is, of course, angry (as is his nature) at the comparisons of administrative actions which you make above. He believes that administrators have discretion in their actions, but are expected to use some common sense. To compare the block of a long-standing and useful user to a technical problem over redirects is, in his eyes, insulting to those who spend their time contributing to this encyclopedia.

I provide this link, should you wish to use it. Physchim62 (talk) 18:41, 20 November 2007 (UTC)[reply]

I do not know what you intend with these type of comments, but sarcasm is a trait that is neither valued, nor useful. May I suggest that you move on to more useful endeavors? ≈ jossi ≈ (talk) 19:01, 20 November 2007 (UTC)[reply]

Hi Durova. One thing I haven't seen discussed, though there has been so much I might have overlooked it, is how you got on the trail of investigating !! in the first place? Did you 'pick' !! for investigation because of something they did? Or are you running some sort of large scale analysis methodology against all users which picked up some sort of 'apparent correlation'? I don't need to know any 'proprietary hunting methods'... just whether the chicken or the egg came first. Did you decide to investigate !! and then come up with correlations or get correlations from a generic search tool and therefor decide to investigate !!? --CBD 12:52, 19 November 2007 (UTC)[reply]

Would it make much difference to you either way? One of the things I like to do is keep the sockpuppeteers guessing. If they knew the extent of the automation I use, or how things come to my attention, then that could be very useful to them. A really successful sockpuppet is one that never gets scrutinized in the first place. Durova^Charge! 17:43, 19 November 2007 (UTC)[reply]

Well, I was going to try to relate the way you go about selecting people for investigation with the way we handle checkuser requests. We don't just run checkuser on anyone and everyone... we use it only when there is already a reasonable pre-existing suspicion of wrongdoing involving multiple accounts. This is primarily to comply with Wikimedia's privacy policy, but it also serves the function of protecting people who are doing nothing apparently wrong from intrusive investigation and unwarranted suspicion. If you are doing a global search then you are investigating everyone. If you are picking people... I'm not aware of anything which would have placed !! under suspicion prior to investigation. We have rules and restrictions around when checkuser can be used, an ombudsman commission to investigate alleged mis-use, logs which can be reviewed by multiple parties to uncover any problems. Your methodology, whatever it is, seems to be just as capable of intruding on privacy and creating unwarranted suspicion... but has none of these safeguards that we place on checkuser. You send your results to a chosen group of people for review, with the unfortunate side-effect of expanding the breach of privacy, but there are no controls on who you investigate in the first place or why you do so. The act of investigation itself, regardless of what is found or done with the information, is an intrusion - which is why we restrict it to cases where there is reasonable suspicion for checkuser. --CBD 09:33, 20 November 2007 (UTC)[reply]

My methodology uses information that is already publicly available to anyone on the planet who has an Internet connection. People who run checkuser, of course, receive full and candid answers to all their questions. I've explained a bit more about this particular instance over here. Durova^Charge! 17:16, 20 November 2007 (UTC)[reply]

		The Defender of the Wiki Barnstar
		Many thanks for your tireless efforts in keeping those who intend to disrupt, off this project. Wikipedia is a better quality project because of hardworking and conscientious editors like you! Hu12 (talk) 20:10, 19 November 2007 (UTC)[reply]

Thank you very much. Cheers, Durova^Charge! 20:43, 19 November 2007 (UTC)[reply]

Keep your chin up, Durova.. you handled everything well once you got the additional evidence. Illegitimi non carborundum. SirFozzie (talk) 21:05, 19 November 2007 (UTC)[reply]

Thanks very much. Durova^Charge! 21:52, 19 November 2007 (UTC)[reply]

I'll second that. I can't imagine having to post that "nevermind" in full view of the entire community. Hang in there. :-) ATren (talk) 04:20, 20 November 2007 (UTC)[reply]

--Le Grand Roi des Citrouilles^Tally-ho! 19:04, 20 November 2007 (UTC)[reply]

Thank you very much. Durova^Charge! 19:10, 20 November 2007 (UTC)[reply]