Wikipedia talk:Arbitration Committee/Noticeboard

Wikipedia:Arbitration/Requests/Case/Nightscream closed

Original announcement

• For the sake of transparency I would like to make it clear that I did not discuss this case in any way, shape, or form with the other arbs during the entire time it was underway in any place other than on the publicly visible case pages. One of our secondary mailing lists was used for any off-wiki discussion, and I had myself unsubscribed from it for the duration of the case, so not only was I not involved in the discussion I do not even know what, if anything, was discussed and who may have said what about it. Beeblebrox (talk) 22:03, 18 January 2014 (UTC)

Arbcom and/or its various functionaries arranged to have a separate mailing list to ensure appropriate behaviour during this case. An off-wiki criticism site seems to disagree that this was effective. Off-wiki nonsense has no standing here, and I congratulate the people involved for making sure that this case was conducted appropriately. --Demiurge1000 (talk) 22:42, 18 January 2014 (UTC)

If off-wiki nonsense has no standing here, why is Demiurge1000 publicizing their activities here? NE Ent 22:46, 18 January 2014 (UTC)

Speaking as someone who hasn't participated (now or ever) in any off-wiki criticism site, it is a little bit troubling that the current ArbCom mailing list setup 'recycles' its secondary mailing lists. If I understand things correctly – and please correct me if my understanding is mistaken – members of ArbCom who are recused from parties to a case are unsubscribed from one of the secondary lists (arbcom-en-B or arbcom-en-C, etc.) only for the duration of the case. They have full access to what their colleagues on the committee said about their actions and testimony, and full access to the case's private deliberations after the fact—a privilege that the vast majority of parties before ArbCom do not enjoy. This does strike me as something that we ought to fix. TenOfAllTrades(talk) 22:55, 18 January 2014 (UTC)

That is the reality, and, I agree, something that should be fixed. For this case, though, there was no mailing list discussion of any significant consequence. Courcelles 23:00, 18 January 2014 (UTC)

Well, this would still apply to discussions held on the main arbcom-l; we can't predict who will be elected to ArbCom in the future. --Rschen7754 23:05, 18 January 2014 (UTC)

True...and that does point to a(nother) flaw in the way the ArbCom manages its mailing lists. We probably shouldn't be opening the private deliberations of old cases to former parties who are newly-seated, though at least that practice has two somewhat ameliorating features. First, in most cases, the case and the Arb being seated are going to be well-separated in time; the new Arb isn't going to be working today with the Arbs who were secretly critiquing his testimony yesterday, and in many instances there will have been turnover on the ArbCom in the meantime. Second, in principle it could be argued that the community taking the decision to seat a past party to Arbitration amounts to an endorsement of giving them access to those past case files (though that's a stretch).

Both of those points break entirely when we deal with a case involving a sitting Arbitrator, though. The voting Arbs are left in the awkward position of knowing that they are going to have to work closely with the party to the case for months or years to come, to persuade and cajole and extract votes. The voting Arbs will be under pressure to be much more...taciturn...in their discussion of their temporarily-recused colleague, in a way that they will not for any other party to Arbitration. It's an imbalance that we could and should be avoiding. At an minimum, there should be a sensible cooling-off period – perhaps 24 months (one ArbCom term)? – before case records are unsealed to a party. TenOfAllTrades(talk) 23:23, 18 January 2014 (UTC)

(edit conflict) As Courcelles' comments demonstrates, this is what we were always told. However, a cursory Google search tells me that deleting threads is indeed possible[1]; or more accurately, replacing the contents of messages relating to recusal-laden cases with "Message body deleted" is possible. I will try to arrange with the WMF at some point this month to have this done for the Nightscream threads. AGK [•] 23:07, 18 January 2014 (UTC)

What problem are you trying to solve here, AGK? The purpose of recusal and separation of recused arbitrators from the relevant mailing list, while maintaining the archives, is so that the recused arbitrator cannot affect the specific decision being made, while future arbitrators may benefit from reviewing the discussion that was contemporaneous to the decision. The Nightscream decision has been made. Are you suggesting that Beeblebrox is so untrustworthy that he will read the archived discussions and somehow behave unethically with respect to them? Risker (talk) 18:01, 19 January 2014 (UTC)

That is in no way a reasonable interpretation AGK's comment. Beeblebrox raised the issue in the first place and Courcelles endorsed agree with ToT's suggestion that there is something to be fixed. (And what's with the toppost?) NE Ent 18:28, 19 January 2014 (UTC)

Sorry about the top post, it was meant to be further indented, and I've fixed that. But again, what is the problem that messing around with archives is supposed to fix? Really, the only one that seems to be described is that one individual who recused from the decision-making process may potentially review the archives and....what, exactly? Recused arbitrators are bound by the same expectations that they not release information from the mailing lists as every other arbitrator. Recused arbitrators are expected to "stay recused" for future discussion of the same topic/user/situation, so having access to archives should not affect future decision-making either. What problem is being solved? Risker (talk) 18:55, 19 January 2014 (UTC)

Don't know but it appears the issue you might not have considered is that one of the general reasons for keeping private long after the communication is over, is to promote full and candid discussion at the time about people and things not party to the communication (see, privileged communication). But maybe a time limit on that makes some sense. Alanscottwalker (talk) 19:09, 19 January 2014 (UTC)

Yes, exactly that. TenOfAllTrades(talk) 03:34, 20 January 2014 (UTC)

Unless I am mistaken, because I did not receive these messages at the time I would have to deliberately go into the archives and search for them in order to see them. I have yet to even set up that ability as it seems technically complicated and I have not had reason to do so yet, but I can certainly see the issue here.

Having not messed with the archives yet I am unsure of their functionality, but I wonder if there isn't an easier way to do this. If who accessed what in the archives was logged and arbs were instructed not to peek at matters from which they were recused or were parties to we would have a mechanism for making sure no one was doing so while still leaving the material there should it need to be reviewed by arbs at some future date. I don't know if that is technically feasible but if it is it would be a good idea. Logging would also (hopefully) prevent arbs from looking at stuff out of mere curiosity as opposed to material they have an actual reason to review.

Again, I don't know how feasible this is but if it isn't difficult to set up additional lists we could simply have a new list for any case in which any current arb is recused or a party and they would just never be subscribed to it. That might get complicated but it may be worth looking into.

For the moment, I can only offer my promise that I will not peek at them. (doesn't seem like it would be very interesting anyway, this does not seem to have been ac case where there was significant disagreement and as far as I am aware there was no off-wiki evidence involved) Beeblebrox (talk) 23:52, 18 January 2014 (UTC)

I'm inferring from AGK's link mailman is used for the lists. It is (or should be) trivially easy to set up a new list for cases with recused arbs. (I was thinking that lists could be set up for arbcom-l13, arbcom-l14 and arbcom-l redirected to the current year, but that'd probably make dealing with an overlap case problematic.) NE Ent 23:58, 18 January 2014 (UTC)

If it's easy to set up new lists, would it not make sense to have one list per case? It would need someone who was able to admin the lists so that access could be given to future arbs if an old case needs to be reviewed. Imho at least I wouldn't have any problem with that person being a WMF employee as the duties shouldn't be onerous. Thryduulf (talk) 08:45, 19 January 2014 (UTC)

We already have to file a request on Bugzilla to open new mailing lists, and from my experiences, the turnaround time is about 2-3 weeks to get one list made, even for something as uncontroversial as handling the oversight requests for the Portuguese Wikipedia: bugzilla:46348. I highly doubt that WMF would be willing to do this. Also, in response to AGK, that's probably a WMF issue, not a mailman issue. --Rschen7754 08:53, 19 January 2014 (UTC)

Ask for a fish, you eat for today, ask for a fishing pole, eat forever. I've asked our WMF liaison Mdennis (WMF) to liase for us here. NE Ent 12:55, 19 January 2014 (UTC)

Thanks very much, NE Ent. AGK [•] 19:28, 19 January 2014 (UTC)

• Okay, so, to review. The purpose of maintaining archived lists is so that future arbitrators can review any discussion about cases that were conducted in the past, because that information can be significant for rendering clarifications and amendments to cases. It's also very valuable in looking at unblock requests (to see if there were any previous ones and what issues were flushed out during previous reviews), and it's pretty much the only way to review whether or not someone actually sent in their self-report of "legitimate" alternate accounts. Most of the case-related discussion on mailing lists focuses on either the bureaucratic issues related to cases (who will write it, why it's overdue, whether any special issues should be considered); it's the less-common cases where something specific *is* discussed where it's important that future arbitrators have access to those discussions. Ensuring that relevant information remains accessible to future arbitrators is a huge make-work project; the OTRS queue that was used for 2010 applications for checkuser/oversight was closed and deleted at the end of last year, but just ensuring that all significant information was transferred took me about 30 hours of comparing data. And while it is correct that it is technically possible to delete individual emails from the archives, every deletion messes up any links to specific emails throughout the entire archive and requires direct WMF operator action.

  It's not particularly relevant that arbitrators recused on cases have the ability to see discussions after a decision is made: the purpose of recusal is to take that individual out of the decision-making process, and is done by their not participating in the discussion. The same standard of discretion and confidentiality of mailing list discussions applies to arbitrators whether or not they agree with any decision made by the committee, and applies whether or not the arbitrator is recused on a case; there is no reason to believe that recused arbitrators who subsequently read a discussion relating to a case on which they recused will behave in a way that abrogates those standards. If they did so, I'd suggest that is a reason for removing that individual arbitrator from the committee - mailing list leaks are mailing list leaks - not a reason for creating a rabbit's warren of otherwise useless mailing lists.

  My position for the last few years I was on the committee, and continuing today, is that the real issue is that the committee should move to a CRM system and stop using Mailman, which will appropriately permit changes in practice in the handling of future data; and that the older archives should be first sealed off, and then only significant information forwarded to the new system, or alternately stored on arbwiki, prior to the deletion of those archives. I'll note, however, that it is widely known that past and current arbitrators have full-scale copies of mailing list archives up to the time that they were/are no longer on the list, and that prior leaks have shown that it's important that *someone* be in a position to verify whether or not the "leaked" information is actually from the archives; we were aware that there was some pretty significant modification, abridgement, and deliberate omission in past leaks. Risker (talk) 16:09, 19 January 2014 (UTC)

• I have to agree that just deleting the whole thread serves no purpose. I filed an RFAR last year and commented on some cases as well, should we now go back and delete any emails related to those as well? And We have an arb who recused from a current case as well, so all those emails would have to be deleted as well or Brad might go look at them later. I realize this is an unusual case in that I was not only recused but was the filing party and became an arb in the middle of the process, but deleting the emails would set a bad precedent that could get very messy very quickly. Beeblebrox (talk) 19:33, 19 January 2014 (UTC)

• I think we're just establishing whether deleting threads is technically possible. Deleting the threads would require a committee vote, and no such vote has been held, so this thread to my mind is just looking at whether this is an option available if we ever needed it. I tend to agree with you that deleting the threads for Nightscream is unnecessary, though some community members took a different line earlier in this thread. AGK [•] 19:40, 19 January 2014 (UTC)

• Perhaps I'm missing something here, but it seems to me that a member who is recused must not in any way influence a case, but there isn't a priori a harm of that person being aware, or later becoming aware, of what went on, so long as privacy is properly maintained. So I'm not particularly bothered if non-viewing of mailing list content by recused members is voluntary instead of software-enforced, just as long as the viewing does not lead to influencing. --Tryptofish (talk) 20:03, 19 January 2014 (UTC)

This is an interesting discussion, because arbitrator involvement and recusal can work in a number of different ways. In this case, Beeblebrox was recused because he was (IIRC) the filing party and acting in, essentially, a prosecutorial role (I can confirm that he has acted entirely properly throughout). Recusal can be for different reasons, though. I recused in a case that was heard at the tail-end of last year (the Ottoman Empire-Turkey naming dispute), not because I was involved in the dispute itself, but because that dispute was part of a much wider topic (World War I) in which I have edited and intend to edit in more heavily over the coming years. I wanted to, essentially, be free to speak my mind on the matter as an editor if needed. I didn't pay much attention during the case (that was a mistake, I should have done), but as matters developed after the case I saw some things that gave me cause for concern, and I've found myself as an editor with knowledge of how arbitration works giving advice to one of the parties in that case on how to proceed. It is a complex matter and I would prefer the details not be discussed here. But is this 'defence' role (even if it is entirely after the case closed) any more appropriate than a prosecutorial role? What I've essentially found myself doing is trying to maintain an awkward balance between giving appropriate advice, reconciling my concerns with those of others, and getting a fair hearing for someone from those who may look askance on me giving this sort of advice. I'm not sure it is a balance that it is possible to maintain. Carcharoth (talk) 01:34, 20 January 2014 (UTC)

I would point out that while creating lists isn't hard, nor is removing content from them, they are a quite costly thing to do in terms of staff time. It means context switching for someone every time they need to do it, and my understanding is that removing the content of particular archived posts is quite dangerous, and prone to creating problems with the archive. Our tech team is highly capable of creating lists - in fact, that part is so easy that they've delegated it (and I'm one of the people who does it) - but the expensive part truly is the context switch, and the staff member who has to take the time to create and configure each list, double-checking the setup because of the sensitivity of Arbcom cases, etc. I strongly encourage us to not go down any path that involves that sort of work. At that point, I'd rather look back toward workforce management and/or CRM products, as the Committee did last year, a conversation that has appeared to be stalled (after the WMF invested considerable resources in creating three demo instances). Let's not perpetuate the problem by using technology that isn't a fit - let's fix the root cause. Philippe Beaudette, Wikimedia Foundation (talk) 02:27, 20 January 2014 (UTC)

Removing content from a mailing list's archives - after it has been sent - is quite a difficult task and can totally goof up an archive.--Jorm (WMF) (talk) 03:07, 20 January 2014 (UTC)

• "after the WMF invested considerable resources in creating three demo instances" is unwelcome passive-aggressive nonsense.

  The fundamental problem with CRM software was that none of the packages trialled could easily and entirely replace Mailman as we use it. The demos were of customer-facing ticketing software (one of them was indistinguishable in substance from OTRS), not internal email discussion software. Switching at this stage, and because we can't delete threads, would be overkill.

  In any event, it does seem like deleting the body of sensitive messages after the fact is possible, but the Foundation are reluctant to make it a habit - for understandable reasons. I suppose this means the community has its answer, and a further question to consider: as you elect arbitrators knowing they might gain access after-the-fact to deliberations on cases they are a party to, is this worth it? I'm inclined to think not, but don't much mind: it's not me that would do the fiddly deleting of threads. AGK [•] 09:34, 20 January 2014 (UTC)

AGK, I find this comment (""after the WMF invested considerable resources in creating three demo instances" is unwelcome passive-aggressive nonsense. ") ... regrettable. I'm not sure why the hostile tone came out there, but I said nothing that wasn't true. The WMF did invest substantial resources in putting up three demo instances. I'm sorry none of them were to your liking, but the Arbcom had plenty of time to propose alternative solutions, which we would happily have looked into. I regret your characterization of that phrase as "passive-aggressive nonsense". It's not the collegial tone that has been employed between the Arbcom and the WMF in the past. Philippe Beaudette, Wikimedia Foundation (talk) 03:19, 21 January 2014 (UTC)

Well, Philippe, if a polite request and a collegial tone are all that's necessary, I respectfully request that the WMF start taking a more active role in policing and sanitizing all-languages Wikipedia BLPs, including establishing and consistently enforcing strict notability guidelines, and quickly and effectively taking care of any and all child-protection issues. I humbly submit that both of these areas are as appropriate for the expenditure of substantial resources as the three demos mentioned above. Cla68 (talk) 06:26, 21 January 2014 (UTC)

• • If none of the packages were suitable, that's a good reason not to use them. It isn't a good reason to stop looking for one that is though. Has anyone drawn up a specification? Has anyone detailed why the trial packages weren't suitable? What features were missing? Has anyone done any research into what other packages are available that could/should be trialled? Thryduulf (talk) 11:49, 20 January 2014 (UTC)

The RT package was indistinguishable from OTRS, which was not a good fit. However, either of the two CRM systems are a very considerable improvement over the Mailman system; they just failed to be perfect, and required changes to the way some things are done. They were quite suitable, and a test use was proposed by moving the ban/block appeals there (which would have provided very significant tracking improvements, logging, and ease of use of standard responses), but never happened. It might be because last year there was a dearth of arbitrators with a technical bent; that isn't a problem this year. The current arbitrators have the notes on the three systems that were compared, and I believe the test instances are still available. Risker (talk) 12:29, 20 January 2014 (UTC)

• It looks as though there may be some appetite on the new incarnation of the committee to review these possibilities again. Beeblebrox (talk) 19:06, 20 January 2014 (UTC)

I am absolutely thrilled to hear this, Beeblebrox, and will gladly provide background information and assistance wherever the committee may find valuable. If someone has specifically accepted the task of working on this, he or she should feel free to contact me. Risker (talk) 19:19, 20 January 2014 (UTC)

• Uninvolved comment here. Any CRM worth a damn would have auditing capabilities and ACLs; even the Windows default file system (NTFS), which is hardly a CRM, has both of these, so really, just share a Windows folder via SMB over VPN in case of desperation. Better CRMs might even have the more sophisticated RBAC making both proactive security and auditing rather easy to handle... assuming you can trust someone to manage it (i.e. not purge the logs, etc.) Quis custodiet ipsos custodes? still applies no matter the level of technical sophistication. I should add that the Wikipedia articles on these technical topics read far more scary than these things actually are to understand and use in practice. Someone not using his real name (talk) 11:08, 22 January 2014 (UTC)

• The stuff discussed in depth above (someone not being both arb and party in the same case) is technically called [simple] dynamic separation/segregation of duties (DSD) in the RBAC (and other) security model(s). Plenty of commercial software has support for DSD because it's needed for SOX compliance. (You can do it "manually" with ACLs, but it's an obvious PITA when role holders change, whereas a RBAC model handles this seamlessly.) I don't know of any forum/wiki-type software which has support for DSD though (not normally needed there). On the other hand, I can't think of the top of my head of XRM software that supports out of the box the kind of workflows that ArbCom has, even just limiting the discussion to the public ArbCom workflows. In the $$$$+ world, stuff like this is made with semi-custom software for various XRM core products (which may themselves be FOSS). Still the custom workflows will cost you, either money or own/WMF developers' time. Someone not using his real name (talk) 14:00, 22 January 2014 (UTC)

• I've looked at SugarCRM, which seems to be what the WMF uses, but being a FOSS product it doesn't seem cater much for more serious corporate uses. In particular it doesn't seem to have support for separation duties in its RBAC implementation; that's level 3 RBAC ("constrained RBAC") in the NIST standard. Someone not using his real name (talk) 09:21, 29 January 2014 (UTC)

Promotion of clerk Rschen7754

Original announcement

• Well deserved. --Guy Macon (talk) 23:35, 25 January 2014 (UTC)
• All I am going to do is keep my mouth shut and shake my head. He wouldn't have been my choice though. Kumioko (talk) 23:56, 25 January 2014 (UTC)

This is apparently some new form of keeping one's mouth shut that I was not previously aware of. ArbCom clerking is not some position of power, they are users who volunteer to help the committee with what are quite frankly the more boring and beuracratic aspects of its proceedings. He's been a trainee clerk for some time and I am not aware of any problems with his work. Beeblebrox (talk) 01:42, 26 January 2014 (UTC)

Yeah I could have said worse and I know he/she doesn't think any better of me. Kumioko (talk) 02:31, 26 January 2014 (UTC)

Is there any chance you could start your comments with your signature so we know to ignore them? Johnuniq (talk) 03:30, 26 January 2014 (UTC)

Kumioko (talk) 03:58, 26 January 2014 (UTC) Sure! Anything for you John!

I previously had serious doubts, but Rschen7754 and I talked it over like adults and I became convinced that he would do an excellent job as a clerk or any similar position. I think that his record so far confirms my opinion 100%. Like I said, well deserved. --Guy Macon (talk) 18:22, 26 January 2014 (UTC)

Clerks don't do anything that requires much though anyway so it should be fine. They just post stuff on behalf of the committee and learn the ropes to be committee members themselves. I don't really have a problem with him being a clerk but I can see in his actions the asperations to be an Arb and I don't think he has the right demeanor. I also don't believe he has the projects best interests in mind and I can see in many of his actions self serving actions designed to futher his own gains and goals. 138.162.8.59 (talk) 14:23, 30 January 2014 (UTC)

Wikipedia:Arbitration/Requests/Case/Kafziel closed

Original announcement

• I'd just like to say that some of the way this case was conducted was throughly vexing, such as choosing the punishment and then and only then looking for the crime. I haven't really followed ArbCom cases before so I sincerely hope this isn't the way the body regularly does its business. KonveyorBelt 00:33, 29 January 2014 (UTC)

That isn't entirely accurate. If you look at the sequence of events you would see that several arbs had a crime proposed to go with the punishment. It was because other arbs were not happy with the verbage that it had to be changed. Yes, for a brief period a punishment was on the table without a proper crime to go with it, but the punishment still wasn't first. If anything the crime they settled on was basically the same as the original proposed, they just split hairs about how they should articulate it. 204.101.237.139 (talk) 17:45, 29 January 2014 (UTC)

Of course it was mandatory to get Katziel desysopped. You can't have admins disrespecting bureacrats, arbitrators, foundation stuff, stewards, Jimbo or any other of their various wiki superiors. Imagine what would happen if commoner class wikipedians would start showing bad attitude to admins, what good would that do? Whole power-structure of Wikipedia could crumble. The arbs surely did split hairs about what the crime was, for the crime was really "you did not respect our authoritat!" and how to articulate their predecided attitude desysopping, not just the presentation in the case pages. jni ^(delete)_{...just not interested} 07:33, 30 January 2014 (UTC)

• Anyone who has watched ArbCom knows that the proposed decision is a draft by one or two Arbs, and sometimes groups or even the majority of arbitrators disagree with parts of it. In this case, a review of the evidence had led most arbitrators to a desysop conclusion but only after the draft was posted did it become clear that several different reasoning paths had been adopted. In an appellate court, this would result in a series of concurring judgements resolving the case but providing little useful precedent. ArbCom does not use this approach, preferring a single set of principles, findings and remedies, which in cases like this one leads to the unedifying display of how the sausage was made. It is more common for the findings to be agreed and the remedies to be the subject of disagreement, but the events of this case were neither unprecedented nor sentence-before-evidence as has been implied. ArbCom has had cases of sanction without evidence - one notorious example had a sanction posted before the evidence page was used, IIRC - so there have been much objectionable cases than this. By the way, as an outsider observation, Kafziel's behaviour on and off wiki made the outcome inevitable, and no I don't mean the incivility. I don't think anyone should have to kowtow to ArbCom and I have certainly criticised mistakes and unreasonable actions. The outcome for Kafziel was neither, and the one for Hasteur was within reasonable discretion. By all means, trout ArbCom when they deserve it, but stay with accuracy when it comes to facts, and the fact is this was not a punishment before evidence case. EdChem (talk) 08:00, 30 January 2014 (UTC)

• Thank you very much for offering your thoughts; I agree with you almost entirely. AGK [•] 22:32, 30 January 2014 (UTC)

• No problem, AGK. I believe in criticising fairly, which means not only being critical when it is warranted but also saying when criticism is unwarranted and acknowledging positive actions. It's a matter of principle for me, acting reasonably, and it has the benefit (at least, I think it has the benefit) of having some credibility when offering criticism - after all, if one can't recognise what is reasonable, why would one be trusted when pointing to something unreasonable? (By the way, on the criticism front, this case was a good example of why better control of case pages is needed...) EdChem (talk) 02:01, 31 January 2014 (UTC)

• Yep, ArbCom, you still deserve the barnstar of decapitation for such bang-up job at rooting out the evils of another Rubashov. Whatever would we do without the High Court of Injustice squashing dissent, punishment first, for all to see. Yay!--ColonelHenry (talk) 15:28, 30 January 2014 (UTC)

• A tad melodramatic, if not downright silly. AGK [•] 22:32, 30 January 2014 (UTC)

Arbitration motion regarding Increase of protection on article protected under WP:OFFICE action

Original announcement

Seriously disappointed that Philippe's behaviour was not given more attention. A few arbs made some comments, which was positive even though he may not even read them, but I hoped for more. EdChem (talk) 08:04, 30 January 2014 (UTC)

Totally. Is it as difficult as admonishing Jimbo was in 2009? Bishonen | talk 10:12, 30 January 2014 (UTC).

Yes,  Roger Davies ^talk 10:45, 30 January 2014 (UTC)

I agree with EdChem's view. A statement about the community's expectations about the professionalism of WMF staff would have been welcome.  Sandstein  10:53, 30 January 2014 (UTC)

You may find this surprising but in this instance agree in all respects. This is typical of the WMF's feelings towards the community though and I really think something needs to be done about it. Not sure what that is though. The WMF seem to think the community members are expendable and treat them with contempt which I find highly disappointing. They make decisions that affect the community in a negative way like Visual Editor with no sense of responsibility for the fallout and expect the community to fix the problems themselves. Its really a shame the WMF doesn't seem to want to work with the community but then again its our own fault because the community generally fails to accomplish amnything resembling change. So its a double edge sword. They don't trust us because we have shown we can't be trusted to do the right thing and we don't trust them for the same reasons. At some point though if this project is going to survive (which I frankly think is already getting to an irreversible point) both sides need to work together. Otherwise we'll all be out of a job. 138.162.8.59 (talk) 15:17, 30 January 2014 (UTC)

There may have been a salient point in your message, but it's totally lost now. Unfortunately, a side-effect of generalising a true statement is that it often ends up as a false statement afterwards. --(ʞɿɐʇ) ɐuɐʞsǝp 19:54, 30 January 2014 (UTC)

Frankly it was lost when I started it as I do not believe there is anything the community can or will do nor is there even the will to pursue it if they could. This whole episode emphasis how broken this fragile system is and how no one is stepping up to do anything about anything. Its easier to ignore the problem that to accept there is a problem and pursue a remedy to it. 138.162.8.58 (talk) 20:47, 30 January 2014 (UTC)

• About 2/3s through this [2] I asked Philippe about his style of approach. I didn't find his response encouraging, I'm afraid. Leaky Caldron 15:25, 30 January 2014 (UTC)
• The WMF can do whatever the hell it wants. --In actu (Guerillero) | My Talk 15:49, 30 January 2014 (UTC)

• That's true and they frequently do, but that doesn't make it right and it doesn't foster a spirit of cooperation with the community. They need us significantly more than we need them but they think its the other way around. I can guarantee that the community can keep this site running (and in many respects probably better) without the WMF. @Leaky, that has been my impression of multiple members of the WMF unfortunately. Not all of course, but many (and I have met quite a few IRL). 138.162.8.59 (talk) 15:59, 30 January 2014 (UTC)

• It can't really, Guerillero. It exists at the community's pleasure; in the simplest terms, without a community the WMF has nothing. Even if this weren't the case, I'd be reluctant to endorse such a servile position. AGK [•] 22:38, 30 January 2014 (UTC)

We don't own Wikipedia, but volunteers are responsible for just about everything the encyclopedia has accomplished, and acting with a bad attitude toward your volunteers do not encourage helping. KonveyorBelt 16:53, 30 January 2014 (UTC)

• What a ridiculous pantomime. Can we get back to the encyclopaedia now, or is that just a distraction form projectspace drama? HJ Mitchell | Penny for your thoughts? 17:14, 30 January 2014 (UTC)

• I realize you are one of the admins on this site HJ but your dismissive attitude about the problem really doesn't help. Ignoring the way the WMF feels about and treats the community isn't going to help it. The WMF needs the community and we need them, but the WMF doesn't feel that way because the community is so disfunctional and can't do anything other than argue. If you/we can start making some meaningful changes on this site and show the WMF that we have the projects best interests in mind and not our own petty POV it would go along way. But being dismissive is both not helpful and not flattering to you. If you care about the project than help make it better, don't degrade others for making a comment that the WMF needs to do a better job of cooperating with whom they serve. 138.162.8.57 (talk) 17:33, 30 January 2014 (UTC)

• I'm not known for my great love of the WMF, nor of ArbCom, and I generally think this thing has been blown out of all proportion. Kevin should have known he wasn't going to get a pat on the back for stepping on the WMF's toes, and Philippe shouldn't have got up on his high horse, but none of the drama that followed has done anything to improve the encyclopaedia. The effort that has gone into resolving this minor spat between two people who really should know better would have been more productive spent elsewhere. HJ Mitchell | Penny for your thoughts? 17:59, 30 January 2014 (UTC)

• I completely agree with all of that but the problem itself stems from a larger unchecked problem. The communities failure to make changes that positively affect the project and the WMF's hubris in thinking that they know better than the people on the front lines doing the work. Both problems are fixable and neither problem is going to fix itself by being ignored. Here are four things that might help although I am certain nothing will come out of them and they are in no particular order and certainly not all inclusive of the problems:

1. When the WMF does an Office action they should ensure they do so with their WMF account. Not their Wikipedia account or their admin account if its different. If its an official action it needs to be done in an official capacity.
2. 2nd the community needs to start actively addressing the various problems. It will take some time to regain our respect back but if the community starts to act in common interest again, then the WMF will liklely follow suit.
3. the WMF needs to act like adults when they interact with the community members. They can't just go flying off the handle everytime their authority is questioned. They have made some really bad choices in the past too so they also have some room for improvement.
4. The community and the WMF need to start working together. Not be fighting each other to show the other who's in charge. On the same token though the WMF needs to understand that this project will not survive without the community. If there is no community doing the work, there is no jobs for WMF employees and they all have to go down the hall and work for Wikia.

A lot of these could be solved at the next Wikimania by the way. Determine what some of the critical problems are and then create some working groups to adderess those. The groups should have members from WMF and Wikipedia and Wikia where applicable. We need to start working together in order to break the us and them mentality that currently exists (like the one between admins and editors). 138.162.8.57 (talk) 18:32, 30 January 2014 (UTC)

• Why isn't anyone commenting upon the fact that an OFFICE action overrode the English Wikipedia policy on PC2 when it appears that policy-compliant options were available? I challenge anyone to come up with any scenario, however implausible, where legal or other considerations require PC2 instead of full protection.

Also, Jimbo's original 2006 announcement[3] made it very clear the office actions were to be "a short-term action" and that "This quick action is in no way meant to override or replace the process of community consensus. There is still plenty of time, and there are still plenty of places, for the community to discuss [the office action]." (Emphasis in original). When was this changed?

Finally, a procedural question; as I read it, to appeal one should contact Philippe, but of course before doing that it would be a good idea to see if anyone else has appealed and whether the appeal was rejected or is under consideration. How would I find this out? --Guy Macon (talk) 19:46, 30 January 2014 (UTC)

To appeal the office action? IINM, Phillipe has specifically stated the only appeal would be to file a DMCA counter-notice on that article, which I don't see happening as it'll out whomever makes it due to the amount of personal information required for a DMCA counter-notice. It's a Catch-22.

As for the issue with PC2, I cannot speculate on the OFFICE's actions, only except to assume that the intent was to keep the target of the takedown notice (i.e. the URLs) from being reposted and made visible, while at the same time allowing the article to be edited - a significant concern given that the general consensus amongst the editors seems to be that the DMCA takedown notice is illegitimate. That said, the WMF using PC2 once more raises questions that were first raised in the 2012 PC RfC, where the developers more or less gave the closers no choice by stating they'd refuse to work on it unless it was turned on (a lot of people on that RfC wanted PC's most onerous issues fixed first). To me, the WMF is forcing PC2 on us, just like they forced PC on us and tried to force VE on us. —Jeremy v^_^v ^Bori! 20:44, 30 January 2014 (UTC)

Pardon the digression (I'd put this on your talk page, since it's only partially relevant to this discussion, but your header says you'll delete it on sight), but speaking as one of the closers of the 2012 RfC: I have no idea where you got the idea that the WMF "forced PC on us" in that RfC. We weren't "given no choice" about anything, we did not take orders from the WMF, and the issue of asking the WMF for further tool development was way, way down the list of things that were considered important by either closers or commenters at the time of that RfC. A fluffernutter is a sandwich! (talk) 21:16, 30 January 2014 (UTC)

"Phillipe has specifically stated the only appeal would be to file a DMCA counter-notice on that article"? That can't be right. I can see why a DMCA counter-notice would be the only way to allow removal of PC2 protection, but I can't see how a a DMCA counter-notice in relevant to strengthening it to to full protection. What were Phillipe's exact words? --Guy Macon (talk) 23:53, 30 January 2014 (UTC)

No, the DMCA counter-notice is for overriding the office action on that article, including altering the protection level. Office actions, per the policy, can't be countermanded by rank-and-file, so adjusting the protection level in any way either has to be done by the WMF or after a successful DMCA counter-notice. —Jeremy v^_^v ^Bori! 05:01, 31 January 2014 (UTC)

Addendum: the specific quote from Phillipe is thus (Talk:Conventional PCI#OFFICE action), relevant parts bolded:

For those who are interested, I've posted the DMCA take-down notice at this location on the Foundation's wiki. It specifically mentions the reference section and requires that we remove the specifications. At this point, until or unless we receive a DMCA counter-notice, those standards can not be reposted. If we were to allow that to happen, we threaten the immunity under which we operate through the Safe Harbor provisions of the Digital Millenium Copyright Act.' I'd be happy to provide someone with instructions regarding where to send a counter-notice if someone has a valid couter-claim, but be aware that it would be a statement under penalty of perjury that you have a valid copyright counter-claim. Philippe Beaudette, Wikimedia Foundation (talk) 01:24, 24 January 2011 (UTC)

Hope this helps. —Jeremy v^_^v ^Bori! 05:10, 31 January 2014 (UTC)

Interesting, but clearly an answer to a different question. Nobody here is even hinting that those copyrighted standards should be reposted. Everybody here is discussing replacing PC2 with full protection. I do not need to have a valid copyright counter-claim to ask Philippe to increase the protection and make it harder for someone to repost the copyright-violating standards.

Nobody is questioning the removal of material that is clearly infringing. we are questioning why Philippe used PC2, which violates our policies, when a non-violating alternative (full protection) was available. Yes, we all understand that Office actions are not subject to our community standards, but it appears to me that Philippe is deliberately violating our community standards, and by doing so greatly increasing the number of editors who can (purposely or inadvertently) infringe on the copyright in question. And I would like to appeal that decision. --Guy Macon (talk) 15:33, 31 January 2014 (UTC)

That's a misreading of WP:OFFICE, but unfortunately your misreading was shared by Arbcom in this case. Admins cannot reverse or circumvent office actions. The policy does not say that alteration is forbidden.—Kww(talk) 05:13, 31 January 2014 (UTC)

Good way to fight back this DMCA nonsense is to go search for:

• PCI Local Bus Specification, Revision 2.1
• PCI Local Bus Specification, Revision 2.2
• PCI Local Bus Specification, Revision 3.0

from Google and make sure the PDFs are mirrored everywhere and never get deleted from Internet. If the stupid consortium has nothing else to do than harass Wikipedia and general public with their silly lawyer letters to keep these very obsolete specifications that have been leaked to public countless times, still secret or behind a paywall, well we the editors don't have to co-operate. The Internet never forgets, so their struggle to remove these is completely futile. You can get direct links to OFFICE-censored materials from editing history of Conventional PCI and if those get rev. deleted, you can always link to the web.archive.org past copies of that article. jni ^(delete)_{...just not interested} 21:00, 30 January 2014 (UTC)

You just encouraged off-wiki behavior that violates copyright law. You have every right to do that, but I have an equal right to point out that that is one person's opinion, not Wikipedia's opinion, and that my opinion is that we should try to change the laws instead of violating them. Putting my money where my mouth is, I have already sent my yearly donation to the EFF, and I encourage others to do the same. --Guy Macon (talk) 00:31, 31 January 2014 (UTC)

The DMCA is not a law where I live. I have zero interest in supporting someone else's stupid laws in some wiki. Even if the links get posted directly to some Wikimedia server, I think I'll just ignore it and not be the first one to enforce their removal, even if it violates our policies to post links to copyrighted material. jni ^(delete)_{...just not interested} 07:33, 31 January 2014 (UTC)

The DMCA may not be a law where you live, but the Berne Convention is. --Guy Macon (talk) 15:33, 31 January 2014 (UTC)

Not to mention US law applies to Wikipedia as it's based in the United States, so DMCA applies regardless of where an editor lives. —Jeremy v^_^v ^Bori! 19:25, 31 January 2014 (UTC)

• This ArbCom case must be the stupidest ruling ever from this silly committee. One admin raised the protection level from its invalid, against policy, state and got angry reply from some WMF bureaucrat. Yawn. Really a tempest in a teapot, no substance whatsoever. I wonder if Kww is learning anything from his "admonishment", whatever that even means. I agree with others here that the WMF needs to do a better job of cooperating with their underlings and servants^W^W^H, eh.. I mean the community of editors. jni ^(delete)_{...just not interested} 21:07, 30 January 2014 (UTC)

• Yawn indeed, but this wasn't an arbitration case; it was an arbitration request. After it was filed, we had no option but to rule on it. I wonder what you would have did if in our position, and why. (Just noticed you've been on Wikipedia an awfully long time, but don't think we've ever crossed paths. Nice to meet you, so to speak!) AGK [•] 22:35, 30 January 2014 (UTC)

• Decline it, because it's not the committee's job to be the hatchet folks for WMF. NE Ent 01:20, 31 January 2014 (UTC)
• I'd either decline it, and if that were not possible I'd rule this to be De minimis non curat lex matter requiring a minimal verdict, which could be nothing at all. Also a "trial" were the other party is untouchable by the "court" is fundamentally unjust as a dispute resolution mechanism. ArbCom has started to resemble a criminal/military court more than a civil process with two equal sides. (I have been semi-retired for years. I don't have much intention to participate on any of the wiki politics here. I liked the ArbCom of old days better, when it was helping banning trolls from wikipedia. Now it seems to be "admonishing" normal editors about trivial things or their bad attitude.) jni ^(delete)_{...just not interested} 07:33, 31 January 2014 (UTC)

I posted a question for Philippe in the discussion Leaky Caldron mentioned above, asking for his view on how Reviewers are supposed to handle the article he has OFFICE-protected. In my view, this is one of the questions that is still open from this situation. I would be hesitant to approve a pending change on the article out of concern of inadvertently treading on Philippe's or the WMF's toes and being hit with a massive over-reaction. Philippe has not edited since my question was posted, so I still hope that he might respond.

Another issue is the content of OFFICE, which suggests core policies will be respected - perhaps foolishly, I would have thought that the protection policy was core. As an Australian, I am reminded of the Howard Government which divided its election promises into "core" and "non-core" after it was elected; the government did not believe it was bound by anything it declared "non-core promises." WP:OFFICE needs to be redrafted if Philippe's actions are acceptable (and his attitude / behaviour unreviewable).

@Guy Macon: one possible scenario has already been proposed in the RfAr, though it was of the WMF's own making... if a confidential agreement were reach as a settlement in which the WMF undertook to institute PC2 protection then changing to full protection might not be recognised by a judge as fulfilling the agreement. Of course, no such agreement should be made, but given what we have seen here, I do believe that making such an agreement is something the WMF might do.

@Roger Davies (or any Arbitrator willing to comment): would you please explain why appending to the admonishment a statement that ArbCom were unimpressed with Philippe's handling of the situation and asking he be more careful, thoughtful, and assuming of good intentions in the future was not considerable possible? I understand that you can't overturn an OFFICE action, but you could state that OFFICE actions should only be undertaken with an official WMF account and that the non-WMF account falls under ArbCom jurisdiction. Kww was foolish to act without asking first, no doubt, though his assumption that using PC2 in violation of the protection policy was also understandable. If he deserved admonishment, then surely Philippe deserved more than obiter dictum observations that telling ArbCom to desysop for him was unacceptable. EdChem (talk) 23:07, 30 January 2014 (UTC)

An agreement with a judge (if it exists) simply moves the problem back one level. Instead of the question being why an office action overrode the English Wikipedia policy on PC2 when it appears that policy-compliant options were available, it becomes a question of why a confidential agreement was made overriding the English Wikipedia policy on PC2 when it appears that policy-compliant options were available. That being said, if Philippe tells us that he made such an agreement and now we are stuck, we would accept it and ask him not to do that the next time.

Not that I buy it; I have been involved in enough DMCA takedowns to be pretty darn confident that no judge was involved. I think Philippe didn't dispute the takedown, which could have wound us up in a court. I believe that instead he took down the links (as is our policy with links to copyright infringing documents) and then took action that he believed to be sufficient to make sure that they didn't come back. That's the whole point of DMCA takedowns; take it down and you are protected from being sued. If I am right, he could have chosen full protection, and he can change it to full protection now. Unless Philippe tells me differently, I am going to assume that there is no such agreement -- it is far too implausible. --Guy Macon (talk) 00:23, 31 January 2014 (UTC)

To clarify, Guy, I read the suggestion of an agreement as meaning following negotiation with the copyright holder, perhaps made by the WMF counsel who could have offered PC1 + semi but didn't and is now instructing Philippe to use only PC2 + semi. I think a judge being involved is highly unlikely. Ignorance of wikipolicy by the general counsel seems plausible to me. None of this is to defend the situation, simply to recognise it is possible; if this is the case, Philippe should really have informed the counsel that PC1 + semi was en-wiki policy compliant and PC2 is not. Philippe likes to quote the part of OFFICE about not being bound by consensus, though the opening lines about acting within core policy seems less popular. I noted above that apparently the protection policy is non-core. I wonder how ArbCom would respond to admins deciding they too are not bound by this minor non-core policy... EdChem (talk) 00:54, 31 January 2014 (UTC)

I am not asking about replacing PC2 + semi with PC1 + semi. I am asking about replacing PC2 + semi with full protection. While Kww was wrong to act instead of asking, he was correct in his conclusion that full protection is the only policy-compliant protection level that stops future infringement as well or better than PC2. --Guy Macon (talk) 01:07, 31 January 2014 (UTC)

I agree that Kww's conclusion in the light of the protection policy is unarguable. EdChem (talk) 02:04, 31 January 2014 (UTC)

• When I run for Arbcom, I always do so on the basis that Arbcom seems to fail to get the point of any matter brought before it. Nothing that happened here changed my mind. All admins, myself and all members of Arbcom included, have an obligation to take steps that ensure that things are in a state that satisfies both WMF and the community. It is beyond unlikely that WMF actually has a need that an arbitrary group of editors that were assigned "reviewer" status during the last trial of pending changes are able to approve changes to Conventional PCI. I can't come up with a single scenario that would make that true. In fact, the article was doing quite fine under semi-protection and was doing so until George Ho complained. I hadn't noticed earlier, but George Ho also complained about the PC2 and had his objections summarily dismissed with the false statement that Philippe's usage of PC2 was in accordance with the letter and spirit of the protection policy at the time. That's a demonstrably false statement: it clearly states that there was no consensus for the use of PC2 at the time PC2 protection was installed.

While Philippe is clearly immune from any form of discipline for his behaviour, WP:OFFICE has never before been read as saying that the office can dictate a maximum protection level without explanation. The legal issue facing the article is a DMCA notice. Full protection would certainly satisfy all issues with that and fall within our protection policy. Judging from history, semi-protection did so as well, and it's quite likely that PC1 would. There are a wide range of choices that would meet the needs of the office and satisfy our policies. The clarification that Arbcom should have issued was that admins are free to upgrade protection from WP:OFFICE imposed actions in the absence of a clear statement from the office that there is a legal need not to protect the article. I suspect that such statements will never be made, because in practice that will never be the case.—Kww(talk) 01:03, 31 January 2014 (UTC)

Good point; since even I seem to have this "reviewer" right -- it doesn't get any more arbitrary than that. NE Ent 01:20, 31 January 2014 (UTC)

according to your user rights logs, it was granted after you asked for it at WP:PERM in 2010,although as I recall it was around that same time that it was being handed out like candy. Beeblebrox (talk) 21:05, 31 January 2014 (UTC)

I have this question to Philippe on the reviewer topic pending. EdChem (talk) 02:04, 31 January 2014 (UTC)

• Related question: I looked at Category:Wikipedia pending changes protected pages (level 2) and didn't see Conventional PCI listed, so I added it. I also saw that Persia (the redirect, not Iran) and Vikramaditya VI are under PC2, but they all seem to have the same "[edit=autoconfirmed]" as a semi-protected page. So I have two questions. One, how do I tell from a protection log that a page is under PC2? Second, do those two pages violate our protection policy? --Guy Macon (talk) 04:25, 31 January 2014 (UTC)

• I fixed those two as well, setting Persia to full protection and Vikramiditya to semiprotection. For me, the log says "heeft publicatieinstellingen gemaakt voor "Persia" [Automatisch controleren: recht "review" benodigd]", so the log should show something like "has installed publication rights for Persia ["review" right required for automatic acceptance". I don't feel like switching to English to check the exact wording. In the section I see labeled with "Automatisch bewerkingen controleren", there's a line that reads in English for everyone: "Require review for revisions by everyone except Reviewers".—Kww(talk) 04:36, 31 January 2014 (UTC)

Persia still has a {{pp-pc2}} tag, which I assume accounts for the orange padlock and the "This is a redirect from a title that is semi-protected from editing" notice. I cannot remove the tag because of the full protection.

The protection log for Persia reads

04:30, 31 January 2014 Kww (talk | contribs) changed protection level of Persia‎ ‎[edit=sysop] (indefinite)‎[move=sysop] (indefinite) (PC2 has not been accepted by the community as a permitted protection state) (hist)

04:05, 26 September 2013 BD2412 (talk | contribs) protected Persia‎ ‎[edit=autoconfirmed] (indefinite)‎[move=autoconfirmed] (indefinite) (Per talk page; this redirect is frequently retargeted without discussion, causing substantial disruption.) (hist)

I see no indication that tells me it was ever under PC2. What am I missing? --Guy Macon (talk) 16:54, 31 January 2014 (UTC)

Observation Much like the supreme court of the United States (or at least the fictionalized account in The Short List on The West Wing) ArbCom has movements of types of cases. Previously it ranged from various nationalistic disputes, to civility, but now has landed on the role of an Administrator with respect to others. Seeing the Nightscream case, the Kafziel case, the "increased office protection" case, and what I see potential case I see brewing (User talk:WilliamJE) I expect we'll see annother case here at ArbCom shortly to further clarify the role. Hasteur (talk) 15:44, 31 January 2014 (UTC)

The WMF can do whatever they please—they own the servers—but what they frequently do is an abysmal job of communicating with the community. At the RfC du jour on pending changes, I asked Philippe a perfectly reasonable question about using PC3 as an office action, but to no avail: assuming our beloved new notification system is working, he chose not to reply. Anyway, boo hiss at Arbcom. It's disappointing to see a body elected by the community admonishing an administrator for upholding the community's own consensus. Rivertorch (talk) 17:36, 31 January 2014 (UTC)

• For anyone not already aware of it, Wikipedia:Petition to the WMF on handling of interface changes may be of interest. --Tryptofish (talk) 19:56, 31 January 2014 (UTC)