Hi again Luis, Thank you for commenting my open letter to Lila. I guess if I send an open letter I should expect open responses, however I surely hope Lila will speak on the matter, "yea," "nay," or "not of concern to me," as I asked. Yes, I recall your previous response to my previous email (which was actually larger in scope, criticizing the now-effective overall privacy policy, whereas I now focus on the access-to-non-public information sub-policy, not yet in effect). In it you said the policies would never attain "perfection." Below you assert "there is no magical answer." These are examples of thought-terminating cliches. Presented with reasoned criticism of the policies, you attempt to stop discussion by saying they can never be perfect or magical. To give you credit, a lot of times thought-terminating cliches are effective in debate with non-lawyers. I'm going to go ahead and answer your "perhaps when we next look at the question in a few years" with the obvious observation that the procedures the policy lays out now are going to affect contributors mightily within the next few years. The access policy is not effective yet and can still be amended. So I'm going to resist your kicking the can down the road a few years. Now, to dig into the actual merits of what you say, I respond that these policies were not "discussed extensively with the community." You obtained input almost exclusively from the *administrative subset* of the community, and none no more so than the individuals that currently have or stand to obtain the accesses in question. Should we be surprised that they prefer anonymity for themselves, as they explore the IPs and browser signatures and so on of the rank and file content editors? No. "The community" according to Lila is *all* the editors, a mere fraction (though powerful) of which are the insider and involved administrative types that commented on the policy drafts. I'm confident you'll agree that this distinction is more or less accurate, that in fact it is the administrative participants particularly that tend to comment this stuff, and not so much representatives of the great masses of content editors that actually built Wikipedia. Please do not gloss over this distinction in the future when claiming immense "community" participation. I'm not saying it's your fault that the discussion wasn't representative though. I'm just saying that's how it is. Neither am I faulting, or at least I shouldn't fault, anything about Michelle Paulson's hard work on the matter. I think the bad decision to accord anonymity to the checkusers and so forth was made higher up. In fact it's interesting to look back in the discussion to see what she said: "1) We do not believe that the current practices regarding collection and retention of community member identification are in compliance with the Board’s current Access to nonpublic data policy and hoped to bring the policy and practices closer to fulfilling the original intent of the policy" (http://meta.wikimedia.org/wiki/Talk:Access_to_nonpublic_information_policy/…). What she's saying is that WMF Legal became uncomfortable with the fact that what the responsible individuals were doing with the identifications (shredding, deleting) was at odds with what the policy clearly stated to editors was the case (identifying). Faced with this problem, there were two ways to go: 1) change the practice to conform with the policy (i.e. start securely keeping the identifications), or 2) change the policy to conform to the practice (i.e. grant anonymity to those granted access to non-anonymous information of others). What I am saying here, and if Lila is reading this far, is that you chose the wrong option. This email is already long, and I am not going to start commenting again why I think the administrative culture has attracted exactly the wrong kind of people, cyber-bullies, MMORPG players, creepers, and that this change to the policy is going to magnify that. I guess I'll just close by saying that it is not that hard to buy a secure file cabinet for the identification faxes and, say, the removable hard-drive containing the identification emails. There aren't all that great many checkusers and oversighters and OTRS volunteers and so forth, and they're not being added that fast. The existing ones can be accounted for in stages. So these "practical difficulties" you refer to Luis, I don't see them as so severe. As for the "risks to volunteers" what are you saying? Are you saying the WMF cannot securely keep some copies of identifications? The real volunteers at risk are those rank and file editors you propose to expose to a group of anonymous and unaccountable administrative participants. Trillium Corsage 27.06.2014, 01:48, "Luis Villa" &lt;lvilla(a)wikimedia.org&gt;rg>: > Hi, Trillium- > > As I pointed out to you the last time we discussed the privacy > policy[1], this issue (and the rest of the policy) were discussed > extensively with the community, with the board, and with the previous > Executive Director. It was then approved by the Board. > > This particular topic was discussed particularly thoroughly, with a > separate consultation and additional discussion with the Board. We did > all that because, as we said in our blog post on the topic[2], this > was a tough question that required everyone involved to balance > difficult privacy concerns with the risks and practical difficulties > of identifying volunteers. There was no magical answer that could > please everyone, despite sincere efforts to find creative solutions > informed by several years of experience building and operating the > previous policy. > > Since we made that post (and since the Board approved the decision) > nothing has changed. The factors being balanced are still difficult, > and Legal would still come down the same way we did in February (when > we finished the public consultation) and April (when we presented our > recommendation to the Board). > > Perhaps when we next look at the question in a few years the facts > will have substantially changed and it will make sense to revisit this > decision and tighten the requirements. But right now, within months of > board approval after a lot of discussion, is not that time. > > For what it is worth- > Luis > > [1] https://www.mail-archive.com/wikimedia-l@lists.wikimedia.org/msg12552.htm > [2] http://blog.wikimedia.org/2014/02/14/a-new-access-to-nonpublic-information/ > > P.S. Tangentially, and speaking mostly for myself, I want to thank the > many Wikimedians I've talked with in the past ~18 months who have been > patient and supportive as we try our best to talk with you, weigh > costs and benefits with you, and make difficult decisions - not just > about privacy but also about many other things large and small. We'd > love to be perfect, have infinite time and infinite resources and > infinite patience, or no hard problems. Since we don't, we have to > just try our best. I'm grateful for and deeply appreciate all the > people who understand that and have worked with us in patient good > faith to move ahead the mission we all share. Corny, I know, but true. > :) > > On Thu, Jun 26, 2014 at 9:06 AM, Trillium Corsage > &lt;trillium2014(a)yandex.com&gt; wrote: >>  Dear Ms. Tretikov, >> >>  Would you please speak on the new revision of the "Access to Non-Public Information" policy? Can you express your objection to it? Can you express your support of it? You'll find it here: >> >>  http://meta.wikimedia.org/wiki/Access_to_nonpublic_information_policy >> >>  This governs the conditions by which the WMF grants access to potentially personally-identifying data such as IPs and web-browser profiles of Wikipedia editors. It grants these to particular administrative participants, for example checkusers and oversighters and arbitrators, of the various "communities," for example the Wikipedias of various languages. >> >>  Under the terms of the prior access policy, those administrative participants were required to send a fax or scanned copy of an identification document. Editors were led to believe that the WMF kept record of who these people actually were. It was repeatedly claimed that they had "identified to WMF." This soothed the concerns of editors like me that thought, okay, well at least someone knows who they are. The truth was that a WMF employee marked a chart of usernames only that the administrative participant's ID showed someone 18 or over, and then shredded or otherwise destroyed those records. The phrase that so-and-so "has identified to WMF" or "is identified to WMF" was so commonly stated, including by the WMF, that I regard it as a great deception and betrayal that it really was shredding and destroying the identifications. >> >>  The new policy is even worse. It abandons the mere pretense of an identification. So while it goes the wrong direction, at least it ceases to deceive. All it calls for now is an email address, an assertion that the person is 18 or over, and an assertion that the owner of the email account has read a short confidentiality agreement. The person need not provide a real name. You are well aware that various web-email services offer basically untraceable email addresses. You are well aware that only a named person can enter into agreement on confidentiality. An agreement by a Wikipedia username with an untraceable email address is not only unenforceable, it is a ludicrous proposition. >> >>  The webpage says the policy is not in effect yet. I urge you to reject it as written and instead have it amended to actually require identification for those faceless entities you prepare to turn loose with potentially cyberstalker tools. >> >>  Whatever your stance, I do call on you to speak on the question. Say "yea," say "nay," or say "not my concern," but at least speak. >> >>  Trillium Corsage >> >>  _______________________________________________ >>  Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines >>  Wikimedia-l(a)lists.wikimedia.org >>  Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe> > > -- > Luis Villa > Deputy General Counsel > Wikimedia Foundation > 415.839.6885 ext. 6810 > > This message may be confidential or legally privileged. If you have > received it by accident, please delete it and let us know about the > mistake. As an attorney for the Wikimedia Foundation, for > legal/ethical reasons I cannot give legal advice to, or serve as a > lawyer for, community members, volunteers, or staff members in their > personal capacity. For more on what this means, please see our legal > disclaimer. > > _______________________________________________ > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines > Wikimedia-l(a)lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

Trillium, Let's be clear about a few things. The only data that checkusers get is a subset of the data that the WMF webservers (and all other webservers throughout the Internet) collect on all visitors. This is data that is voluntarily disclosed by readers (although they may not all be aware of it). The checkusers get substantially less information than is actually available, and only on those users who *edit* and not those who simply view. That means that while you are correct, the Wikimedia community at large certainly includes all readers, only editors are stakeholders in the exposure of certain data to checkusers. There is no legal requirement in the U.S. to make this information invisible (AFAIK). The only limitations are those imposed by the Terms of Service. The previous privacy policy referred to the identification of volunteers to whom certain limited information is exposed, but when Michelle and others said that the policy itself wasn't being effectively enforced more was at issue than how (or if) the IDs were stored. The WMF has never had a method of verifying received identification. Because of the international nature of the movement, IDs were submitted in languages no one at the WMF speaks, from countries and authorities around the world. As a result, anyone could easily submit a false, altered or misleading identification. The identities provided by users with advanced permissions could never be relied upon. So if you want to argue that such users should be positively identified, then please make some practical suggestions (which you have conspicuously avoided doing so far). How should identities be confirmed? In what circumstances should the ID information be disclosed, and to whom? What, fundamentally, is the usefulness in collecting this information to begin with? What are the use cases in which it is necessary? Thanks in advance for providing us with such useful advice!

Trillium, I am having difficulty understanding how retaining copies of possibly forged identification documents helps anyone with holding accountable any rogue functionary or OTRS user. Can you explain that please? Surely someone who intends to misuse the tools will be smart enough to forge an identification document. Even in the United States, forging identification documents is not impossible, and the police occasionally catch people creating such documents. Pine On Fri, Jun 27, 2014 at 7:42 AM, Trillium Corsage &lt;trillium2014(a)yandex.com&gt; wrote:

Pine, An analogous argument to the one you're making is: someone who intends to rob your home will be able to get in one way or other, so why bother locking the doors when you go out. This is not a good argument. You're calling into question the reliability of every identification document copy ever presented to the WMF by an advanced-rights-seeking administrator because a really sophisticated wrongdoer (I dunno, Chinese military intelligence, with whom arbitrator Timotheus Canens is said by some to be associated?) could make a masterful forgery that beats the system. The fact is that 95% of them, I'd suppose, are going to be okay and the identification requirement is going to be an effective deterrent to at least the casual among the bad apples. And of course, once they've truly identified, the personal accountability aspects of it are going to keep in line once well-intentioned administrators that might be tempted to go bad for some reason. "Forging identification documents is not impossible" is another variation of the "perfection is not attainable" and "no policy can be a magical solution" arguments put forth previously on this mailing list by the WMF's deputy general counsel Luis Villa. I've attempted to answer those by explaining that you can have a pretty good and effective policy without having an infallible one. Trillium Corsage 29.06.2014, 07:32, "Pine W" &lt;wiki.pine(a)gmail.com&gt;om>:

On Sun, Jun 29, 2014 at 4:18 PM, Trillium Corsage &lt;trillium2014(a)yandex.com&gt; wrote: Seriously? I think you've gone on long enough for now. You can come off moderation when you contribute something to the discussion rather than attacking others and, dare I say it, just plain ranting. Austin

On 06/29/2014 03:19 PM, Pine W wrote: I'm not privvy to that discussion, but I'd expect that "[...] that does not unduly exclude valuable volunteers" is also an implicit requirement of any identification method considered. Even if you /could/ develop a mechanism by which we had safe and reliable identification of functionnaries, it'd be worthless if most (or even just many) of the volunteers we had were unable to avail themselves of it because of social or geographical constraints. -- Marc