Wikipedia:Arbitration/Requests/Case/Media Viewer RfC
Case clerks: Lord Roem (Talk) & Callanecc (Talk) Drafting arbitrators: Carcharoth (Talk) & GorillaWarfare (Talk)
Wikipedia Arbitration |
---|
|
Track related changes |
Case opened on 03:26, 18 July 2014 (UTC)
Case Suspended by motion on 23:56, 31 August 2014 (UTC)
Case closed on 00:16, 5 November 2014 (UTC)
Do not edit this page unless you are an arbitrator or clerk, or you are adding yourself as a party to this case. Statements on this page are copies of the statements submitted in the original request to arbitrate this dispute, and serve as verbatim copies; therefore, they may not be edited or removed. (However, lengthy statements may be truncated – in which case the full statement will be copied to the talk page. Statements by uninvolved editors during the Requests phase will also be copied to the talk page.) Evidence which you wish to submit to the committee should be given at the /Evidence subpage, although permission must be sought by e-mail before you submit private, confidential, or sensitive evidence.
Arbitrators, the parties, and other editors may suggest proposed principles, findings, and remedies at /Workshop. The Workshop may also be used for you to submit general comments on the evidence, and for arbitrators to pose questions to the parties. Eventually, arbitrators will vote on a final decision in the case at /Proposed decision; only arbitrators may offer proposals as the Proposed Decision.
Once the case is closed, editors should edit the #Enforcement log as needed, but the other content of this page may not be edited except by clerks or arbitrators. Please raise any questions about this decision at Wikipedia:Arbitration/Requests/Clarification and Amendment, any general questions at Wikipedia talk:Arbitration Committee, and report violations of the remedies passed in the decision to Wikipedia:Arbitration/Requests/Enforcement.
Case information
editInvolved parties
edit- 28bytes (talk · contribs · blocks · protections · deletions · page moves · rights · RfA), filing party
- Armbrust (talk · contribs · deleted contribs · logs · filter log · block user · block log)
- Fabrice Florin (WMF) (talk · contribs · deleted contribs · logs · filter log · block user · block log)
- Peteforsyth (talk · contribs · blocks · protections · deletions · page moves · rights · RfA)
- Eloquence (talk · contribs · blocks · protections · deletions · page moves · rights · RfA)
Prior dispute resolution
edit- the MediaViewer RfC, as closed by Armbrust
- response to the results of the RfC by Fabrice Florin
- discussion of how to implement the results of the RfC
- AN discussion
Scope
editWhen submitting evidence, please bear in mind that the locus of the dispute is the Media Viewer request for comments (RfC), and the actions that followed the RfC. If your evidence is outside this scope, please include supporting statements or background context to explain the relevance to the case. If an extension of evidence lengths is needed, please ask one of the drafting arbitrators.
Preliminary statements
editStatement by 28bytes
editLast month, User:Pine started an RfC over whether the new MediaViewer feature should be enabled or disabled by default on Wikipedia. Yesterday, it was closed by User:Armbrust, with the result that the feature would be disabled by default for both logged-in and not-logged-in users.
Following the closure, User:Fabrice Florin (WMF) commented on the talk page of the RfC with a "recommendation" that the MediaViewer continue to be enabled, despite the results of the RfC. Later in the thread, users discussed how to implement the results of the RfC, and administrator User:Peteforsyth made a change to MediaWiki:Common.js that disabled the feature.
Administrator and staff member User:Eloquence reverted the change, and threatened to temporarily desysop Peteforsyth or any other admin who reinstated the change. Eloquence characterized the revert and the threat of desysop as a "WMF action", but it is unclear to me by what authority WMF staffers can overrule the legitimate consensus of a local community, outside of WP:Office actions, which clearly do not apply in this case:
Office actions are official changes made on behalf of the Wikimedia Foundation, by members of its office. These are removals of questionable or illegal Wikimedia content following complaints. Office actions are performed so that the end result is a legally compliant article on the subject.
Now, the WMF does, of course, have "the keys to the server", which means that they have the power to do whatever they like, including desysoping and/or banning anyone for any reason, good or bad; we essentially have no recourse, other than, as is sometimes said, our "right to fork" and our "right to leave."
So my questions for the committee do not include "do they have the power to do this" (yes, they do), but rather:
- Are WMF staffers, who are also Wikipedia editors and/or admins, violating our policies and community norms if they do this?
- If I or another administrator implements the result of a validly conducted and closed RfC over the objections (or "recommendations") of the WMF, are we following policy and community norms, or violating them?
- Was the MediaViewer RfC, and its closure, a valid exercise of Wikipedia's consensus-forming process?
- If, after further discussion, an administrator re-implements the result of this RfC (either by restoring Peteforsyth's change, or via another method), will the committee support that administrator for following the expressed community consensus, or sanction them for wheel-warring?
This is the second time in recent memory that a WMF staffer has threatened to desysop (temporarily or otherwise) a Wikipedia administrator for implementing community consensus against WMF wishes, but in that case, WP:Office was relevant. As the WMF rolls out more and more features (many of which are great, but some of which the community may decide they don't want) this is likely to continue to be an issue. Administrators need to know where we stand when community consensus conflicts with WMF preferences outside of the bright line of "office actions". 28bytes (talk) 14:38, 11 July 2014 (UTC)
Statement by Armbrust
editStatement by Fabrice Florin (WMF)
editThanks for inviting me to participate in this discussion. I am following this case, and monitoring comments made on this thread. I believe that Eloquence has described factually the circumstances and rationale behind his action, as well as adequately presented the Wikimedia Foundation's position on this matter. He has also cited me accurately in his own statement, linking to a number of comments I have made relating to this case. And I find Risker's observations to be factual and well-reasoned, matching my own perspective on this topic. So I don't have anything else to add at this point, but am happy to answer any specific questions relevant to this case. Please let me know if there is anything I can do to help resolve this request for arbitration. Respectfully, Fabrice Florin (WMF) (talk) 19:28, 13 July 2014 (UTC)
Statement by Peteforsyth
editPlease consider the Wikimedia Foundation's understanding of the values and processes that drive Wikipedia, and WMF's ability to engage respectfully and effectively with those processes; to promote harmony over drama; and to promote mutual understanding and shared purpose among stakeholder groups. My position: WMF has lost touch with the values and processes that have driven Wikipedia's growth. WMF has tended toward divisive actions. This damages our shared strategic goals: it drives down productive participation, damages our ability to improve quality, damages our ability to convert readers (reach) into contributors.
- Background reading: Sue Gardner's "Narrowing Focus" memo (2012).
- English Wikipedians agreed to require "autoconfirmed" status for article creation. WMF developers overruled that consensus. (2011)
- WMF deployed the Visual Editor; English Wikipedia community determined it wasn't ready yet. WMF respected this consensus. However, from June-September 2012 there was confusion. Disrupting new and experienced English Wikipedians' workflow for months wasn't good. (2012)
- WMF created "Article Feedback Tool". English Wikipedia RFC removed the tool. (2013)
- WMF upgraded Wikipedia's notification system ("Echo"), but deployed it sloppily. Wikipedians determined it should be disabled (temporarily). WMF staff overrode consensus, without asserting they were doing so. (2013)
- WMF staff proposed hosting non-free video formats. Proposal was strongly rejected. A better understanding of the values driving Wikimedians might have avoided a divisive public debate. (2013)
- On the Media Viewer RFC, two points:
- WMF staff participated in RFC, but in spite of being asked how they would respond to the RFC on June 16, chose to wait until a day after the RFC's July 9 closure to comment on its legitimacy. This approach disrespects a community that considers RFCs meaningful and binding.
- WMF was roundabout in its decision. WMF announced a "recommendation", but clarified it was a decision only upon my (erroneous) Javascript change. Absent bold volunteer action, we might never have known WMF had made a decision. A "pocket veto" (similar to the Autoconfirmed and Echo incidents noted above) is unbecoming of an organization committed to transparency.
WMF makes bad predictions. There's no easy fix for this; better hiring and training might have an impact. A better understanding of values and processes cannot be legislated into existence by ArbCom. But formal recognition of the problem is a vital ingredient for a solution.
Erik's threat toward me was insignificant as an isolated incident, and doesn't demand strong reaction. But over the years, senior WMF personnel have shown poor understanding of our values and processes, sometimes acting with hostility. This has stoked unneeded drama. This pattern invites reflection on how Wikipedians should engage with WMF staff. Two general points (not about the present case):
- If/when staff abuse community-granted tools, tools should be removed -- even if there's no immediate practical difference. (Precedent exists.) Staff, and staff accounts, come and go.
- ArbCom should admonish staff who misuse staff accounts, even if it carries no actionable consequence. WMF should be accountable to Wikimedia communities; a clear statement when a line is crossed is helpful in itself.
-Pete 14 July 2014/edited 15 July
Statement by Eloquence / Erik Moeller / WMF
editHi folks,
A couple of notes:
1) As Risker pointed out, this RFC forms an insufficient basis for a decision affecting all logged-in users due to its small participation by logged-in users (much smaller by a factor of >100 than the number of people who'd previously enabled Media Viewer!), and certainly forms an insufficient basis for a decision affecting all readers. Generally, WMF treats such RFCs on a case-by-case basis; see also Limits to configuration changes for historical examples.
Fabrice Florin, the Product Manager for Media Viewer, has started a conversation here about a process that we could use to get better, more representative information from readers and editors alike. I trust him to continue this conversation, and would encourage others to participate in it.
2) Regarding site stylesheets and JavaScript, we regard them as subject to the same development policy that governs code that is executed server-side; i.e., WMF makes the final call regarding software deployed to sites hosted by the Wikimedia Foundation.
English Wikipedia articulates this in the WP:CONEXCEPT policy, but we’ve stated this independently in other forums as well, e.g. here.
In this instance, a core site feature was disabled on grounds we consider insufficient to do so. The reasons for our decision not to implement the RFC were previously communicated in Fabrice’s response.
I apologize for the unduly stern warning to Pete, who performed an action that he felt was acceptable and warranted, and who did so in good faith. However, the decision to reinstate the feature is one we maintain. Generally, we would ask users to request such configuration changes through Bugzilla in the future, where WMF will always seek to provide a response in a timely manner.
To be clear, we understand that we need to work together in these matters. As noted above, we’re prepared to discuss constructive paths forward. Fabrice will continue to take the lead on that from WMF. Above all, we look forward to further improving the experience for uploading media, viewing them, and curating file metadata, all of which is part of the multimedia team's roadmap.
For the complete avoidance of doubt, this response is in an official capacity.
Sincerely,
Erik Moeller
Vice President of Engineering and Product Development, Wikimedia Foundation
- Clerk note: User:Eloquence is the account used by Erik Moeller, this is a little unclear above.
- A couple of users (notably Gwillhickers and Jayen466) have commented on whether data supports the deployment of Media Viewer for all users.
- First, as per my original statement, our primary concern here was with the decision to disable Media Viewer on the basis of the RFC, given the very low participation rate by a small subset of the English Wikipedia community (64 users voted to disable for an active editor population of >30,000 people who make 5 edits/month). Even with significantly higher participation, when it comes to defaults for readers, the RFC process is unsuitable because it's not designed to take their preferences into account.
- There was a survey built into Media Viewer itself to get user feedback (accessible via a megaphone icon). Regarding the survey data, a few comments:
- The Media Viewer Survey Results breakdown as of 6/20 (15,736 respondents) shows the following approval across all wikis combined: 56% for all users, 65% for readers, 45% for editors.
- It's true that the subset of English Wikipedia survey responses show significantly lower approval, while most other wikis show significantly higher approval. Results have improved since the 6/20 breakdown, see the live dashboard for English Wikipedia (total approval across the sample increased from 28% to 36% -- that's for the entire sample and time period). The data trends tab shows that that the average of daily responses increased from 23% daily approval (6/4, after deployment) to 47% daily approval (7/8, the last day responses were actively invited) across both readers and editors combined. While daily response rates are comparatively low, the increase is consistent over time.
- We attribute the changing response pattern to two factors: users are simply getting used to Media Viewer (change aversion is a common initial response to user experience changes), and the user experience has improved and key concerns have been addressed. Performance of the viewer depends significantly on cache hit/miss ratio for image sizes required by the viewer, which has improved over time. On June 12 we began deploying new features and bug fixes, including a more convenient link to the full-size original version of the file [1]. This was followed quickly by providing a simple opt-out for anonymous users.
- We don't think the survey in the viewer itself is a perfect tool, because it still suffers from significant self-selection bias, initial lack of familiarity by respondents (e.g. many capabilities users expect are there, but take some time to discover, as with any new feature), and a low response rate from readers. As Fabrice pointed out here, we're considering implementing a consistent "Viewing options panel". This could be shown to all users after a few image views as a way to measure how many users actually prefer one viewing mode over another (current opt-out statistics are here). It seems clear that to the greatest extent possible we ought to base defaults on users' actual preference.
- In addition, we maintain that measured changes in user behavior are an important indicator for whether a feature is a net benefit or not. As a small example, do users find the "next/previous" navigation offered by Media Viewer useful, or do they just want to get to the File: page? For a typical day, July 12, we get 7.71M total clicks on thumbnails, and 8.7M clicks on next/previous combined. That demonstrates that users intuitively and frequently use the built-in navigation features of Media Viewer to discover other images in a page. Relatedly, the data shows that the mean load time for media viewer is significantly faster than File: pages (and near-instantaneous for next/previous images due to intelligent preloading of images).
- In sum, in considering the default state of a feature, we should:
- Take into account that initial user response to significant changes in user experience is often negative -- changes need time to "settle";
- Ensure we properly represent the interests of both readers and contributors;
- Measure actual impact of a feature on user behavior;
- Reflect users' actual preference to the greatest extent possible.
- We'll continue to engage on the RFC talk page regarding next steps; Fabrice's most recent response has a lot of useful information in it.
- Erik Moeller (on behalf of WMF) 17:27, 13 July 2014 (UTC)
Statement by Stephan Schulz
editI'm deeply concerned to read a statement by "Eloquence / Erik Moeller / WMF" above. I understand that Erik's user name is Eloquence, and that he is an WMF staffer. But what is not clear to me at all is if his statement is the considered and deliberate statement by the whole WMF, wether the WMF has delegated the task of developing an argument to him, wether it's a statement he makes based on his normal role in the WMF, or wether it's his personal argument and the "WMF" just serves to remind us of his connection with the WMF. We've previously had similar problems when one Arb made statements "on behalf of the committee", without clarifying what level of backing (s)he had. I'd had hoped that we had learned from that experience.
I'm also concerned that Erik is repeating his (or WMF's) argument for trying to force a change on the community, instead of handling the underlying problem of amicably resolving conflicts between WMF and community. He is welcome to start a new RfC and to try to convince the community of his (or WMF's) position - this is not the proper forum for that discussion. What is important here is if admins can enforce (or try to enforce) community decisions, and, even more, if the WMF really wants to impose technical and content top-down decisions using their technical control of the software. If the later is the case, this would be a reason to seriously reconsider my continued participation in the project. Enciclopedia Libre Universal should be a warning. --Stephan Schulz (talk) 00:11, 20 August 2014 (UTC)
Preliminary decision
editClerk notes
edit- This area is used for notes by the clerks (including clerk recusals).
- Recuse I commented,so will recuse from clerk actions.--S Philbrick(Talk) 11:57, 15 July 2014 (UTC)
- The case has reached a definite threshold to open. We'll be making the case pages soon. --Lord Roem ~ (talk) 15:29, 16 July 2014 (UTC)
- Posting an update here: I've asked the clerks to open the case as soon as possible (by Saturday 19 July at the latest). The provisional schedule is one week for evidence (closing weekend of 26-27 July). Workshop page to be open for two weeks (19-27 July). Proposed decision provisionally scheduled for 2nd August. The following notice is to be included on the case pages:
Carcharoth (talk) 00:33, 18 July 2014 (UTC)"When submitting evidence, please bear in mind that the locus of the dispute is the Media Viewer request for comments (RfC), and the actions that followed the RfC. If your evidence is outside this scope, please include supporting statements or background context to explain the relevance to the case. If an extension of evidence lengths is needed, please ask one of the drafting arbitrators."
- Posting an update here: I've asked the clerks to open the case as soon as possible (by Saturday 19 July at the latest). The provisional schedule is one week for evidence (closing weekend of 26-27 July). Workshop page to be open for two weeks (19-27 July). Proposed decision provisionally scheduled for 2nd August. The following notice is to be included on the case pages:
MediaViewer RfC: Arbitrators' opinion on hearing this matter <10/2/1/1>
editVote key: (Accept/decline/recuse/other)
- Recuse. Seraphimblade Talk to me 15:38, 11 July 2014 (UTC)
- This is a complicated issue that I'll mull over for a couple of days, and in the short term, I'd really, really suggest no one attempt to re-implement the fix. Among other things - and correct me if I'm wrong - it appears to prevent anyone from even opting-in, so I'm not sure it did what was intended. But mostly because it would be pouring gasoline on a fire. I really don't know yet what I will suggest in the longer term.
Realpolitik-wise, I suppose we have to recognize that (a) this is not a symmetrical relationship, since WMF owns the servers, but (b) en.wiki volunteers can vote with their feet if they are continually treated with what they consider disrespect. It's kind of like mutually assured destruction, which is kind of stupid since we're theoretically on the same side. One goal I have is to try to convince each side not to push their button. I would have thought that it would be in WMF's best interests to avoid throwing their weight around unnecessarily (i.e. in cases like this where WP:OFFICE doesn't apply), and I was under the impression they were making an attempt to not do that as much recently... --Floquenbeam (talk) 15:53, 11 July 2014 (UTC)
- Peteforsyth has pointed out to me that I'm implying above someone might re-implement his specific edit, which he points out no one is proposing. What I meant was, I think in the short term no one should attempt to implement anything to enact the consensus of the RFC. --Floquenbeam (talk) 20:27, 11 July 2014 (UTC)
- Just being transparent about my evolving thought process (which may be less interesting to people than I think): A lot of this is ultimately political, rather than policy-based, and I'm not not sure ArbCom can make this situation better. It's like the parliament of Ukraine passing a law that Russian annexation of Crimea violates Ukrainian law; that isn't really going to change anything except loudly announce we're impotent. It's also complicated by the fact that this is an imperfect test case: the fix was definitely flawed, and there's a decent possibility that the RFC was insufficient to overcome the fact that 14,000+ users actively chose to opt-in to the media viewer (this is based on @Risker:'s comments; is this really true? I was unaware Media-viewer was ever opt-in. I'd like a pointer to where this data came from, and I'd like to know if anyone on the other side of the argument actively disputes it). If any en.wp admin had reverted this specific change without the threats of a foundation-level block, I would have supported that revert.
Ultimately, the en.wp community should have an expectation that our "reasonable" requests to remove features or make them opt-in will be respected by the WMF; the WMF should have an expectation that their "reasonable" opposition to requests to modify what it considers core features will be respected by en.wp. When people disagree about what is reasonable and what isn't, when there is a long history of dysfunction on our end, and arrogance on their end, it's an intractable problem. I suppose ArbCom would be in a decent position to try to "negotiate" this kind of thing, but I doubt that's really in our remit. If we had a GovCom (which I wish we had), this would be a great job for them.
I'm still undecided, but leaning decline, because really all we have that we can actually do something about are ticky-tack things like "not using an account with (WMF) at the end of the name", and "should have used some different terminology because there's not really anything officially called a WMF action" and "try not to be so rude". Ruling on those type of things is a pretty small potential payoff. --Floquenbeam (talk) 00:34, 12 July 2014 (UTC)
- I suppose I've waited long enough to actually vote that it doesn't matter anymore, but I continue to think that this whole situation requires flexibility, patience, respect, and tact - not strong suits of the en.wp community, the WMF, or (sorry guys) ArbCom cases. I still don't think a case is the best way to handle this, so I'm voting to decline, but I'll hope that I'm wrong and the case proves useful. --Floquenbeam (talk) 02:56, 16 July 2014 (UTC)
- Just being transparent about my evolving thought process (which may be less interesting to people than I think): A lot of this is ultimately political, rather than policy-based, and I'm not not sure ArbCom can make this situation better. It's like the parliament of Ukraine passing a law that Russian annexation of Crimea violates Ukrainian law; that isn't really going to change anything except loudly announce we're impotent. It's also complicated by the fact that this is an imperfect test case: the fix was definitely flawed, and there's a decent possibility that the RFC was insufficient to overcome the fact that 14,000+ users actively chose to opt-in to the media viewer (this is based on @Risker:'s comments; is this really true? I was unaware Media-viewer was ever opt-in. I'd like a pointer to where this data came from, and I'd like to know if anyone on the other side of the argument actively disputes it). If any en.wp admin had reverted this specific change without the threats of a foundation-level block, I would have supported that revert.
- Peteforsyth has pointed out to me that I'm implying above someone might re-implement his specific edit, which he points out no one is proposing. What I meant was, I think in the short term no one should attempt to implement anything to enact the consensus of the RFC. --Floquenbeam (talk) 20:27, 11 July 2014 (UTC)
- As a person with a legal background, I find this case rather interesting, because it demonstrates the difference between power and authority. In this case, the Foundation do have the power to override the community, but not the authority to do so, unless WP:OFFICE applies or a decision has been made by the Board. And neither one of those conditions is met.
On the matter of desysopping, as I affirmed earlier this year, during another case involving a WMF employee, [...] it's always a good idea to restate that, barring cases of emergency, the power to restrict and sanction administrators rests squarely with the Arbitration Committee, even in the event of a WP:OFFICE violation. To understand why that's a good idea, one just has to take a look at [the employee in question]'s impulsive and precipitate reaction (for which, it's true, he has duly apologised).
That said, even assuming we accepted this case, I don't know what our powers would be, considering, that the Foundation do indeed have the de facto power to do as they please (Wikipedia is run on their servers, after all). So, in short, I'm still on the fence and will welcome all input to help me make up my mind. Salvio Let's talk about it! 16:10, 11 July 2014 (UTC)
- I think we always have jurisdiction to determine our own jurisdiction, indeed we do (cf. Kompetenz-kompetenz). Salvio Let's talk about it! 21:15, 11 July 2014 (UTC)
- I believe ArbCom should, as a body, answer Tim's questions. Accept. Salvio Let's talk about it! 10:27, 12 July 2014 (UTC)
- I think we always have jurisdiction to determine our own jurisdiction, indeed we do (cf. Kompetenz-kompetenz). Salvio Let's talk about it! 21:15, 11 July 2014 (UTC)
- I'm not sure if we have a case here or not. As has been observed, there are several issues. At least one of those issues, "can the WMF sumarrily desysyop one of our admins without consensus" is not something we can make any sort of binding decision on. I'm not at all sure the rest of it is something that can't be resolved by lesser means, but I will wait till I've had more time to review the situation and everyone involved has had an opportunity to comment here before considering accepting or declining this as a case. Beeblebrox (talk) 19:19, 11 July 2014 (UTC)
- Still on the fence. While I am not comfortable with doing nothing here, I'm also not sure it requires a full case. Frankly, I find it unlikely the WMF would still be paying attention when we rendered a decision a month or more from now, and the question of the appropriateness of the RFC close seems like something that can be sorted out without a case. I'm thinking of drafting some motions to deal with this more expeditiously. Beeblebrox (talk) 17:34, 12 July 2014 (UTC)
- I have just posted a motion [2] to deal with at least one aspect of this. I think we can deal with this bit by motion regardless of whether we take on a full case here. It's only a request, because we can't write policy ourselves and we can't bind the WMF, but I would hope if the committee passed it the Foundation would respect it. Beeblebrox (talk) 18:47, 12 July 2014 (UTC)
- Please note that I've moved the proposed motion to directly under this voting section, just for continuity and easy-to-find-ness. This is consistent with how motions arising from case requests (as opposed to ones arising independently) have been formatted in the past. My moving the section is not a substantive action. Newyorkbrad (talk) 20:06, 12 July 2014 (UTC)
- I am sufficiently convinced that there are other issues above and beyond the one I have attempted to deal with by motion below, and therefore vote to accept this case regardless of the fate of the motion.Beeblebrox (talk) 22:37, 13 July 2014 (UTC)
- Please note that I've moved the proposed motion to directly under this voting section, just for continuity and easy-to-find-ness. This is consistent with how motions arising from case requests (as opposed to ones arising independently) have been formatted in the past. My moving the section is not a substantive action. Newyorkbrad (talk) 20:06, 12 July 2014 (UTC)
- I have just posted a motion [2] to deal with at least one aspect of this. I think we can deal with this bit by motion regardless of whether we take on a full case here. It's only a request, because we can't write policy ourselves and we can't bind the WMF, but I would hope if the committee passed it the Foundation would respect it. Beeblebrox (talk) 18:47, 12 July 2014 (UTC)
- Still on the fence. While I am not comfortable with doing nothing here, I'm also not sure it requires a full case. Frankly, I find it unlikely the WMF would still be paying attention when we rendered a decision a month or more from now, and the question of the appropriateness of the RFC close seems like something that can be sorted out without a case. I'm thinking of drafting some motions to deal with this more expeditiously. Beeblebrox (talk) 17:34, 12 July 2014 (UTC)
@Eloquence: I just had a look at your user rights, and so far as I can see you do not actually have the technical ability to desysop a admin. It is part of the 'crat toolset. I am left wondering how, if you had been reverted again and had felt obligated to live up to your assertion that you would desysop, you would have actually gone about doing so? Beeblebrox (talk) 18:42, 14 July 2014 (UTC)
- As Floquenbeam and others above state, the actions threatened are troubling, and do not appear to have a basis in any local or global policy. However, the committee's jurisdiction excludes actions of Wikimedia Foundation staff taken while in a staff role—what appears to be the major issue in this case. I'll have to investigate further to see if there are in-scope matters that merit arbitration. LFaraone 19:45, 11 July 2014 (UTC)
- Accept. Posting somewhat later than I had anticipated—outside commitments this week prevailed. It is apparent that there is a need for clarification on a number of issues here. I think we can navigate this case in a manner that stays in-scope while addressing the concerns raised by community members above. LFaraone 20:04, 16 July 2014 (UTC)
- I generally agree with Seraphimblade's discussion in his statement. I think we always have jurisdiction to determine our own jurisdiction, including whether Eloquence's actions fall within one of the exceptions to our jurisdiction, and whether we may nonetheless take notice of it in light of the third paragraph of Wikipedia:Arbitration/Policy#Jurisdiction. Thinking about this some more. T. Canens (talk) 20:46, 11 July 2014 (UTC)
- Accept, to consider, at the very least:
- Whether the freshly minted interpretation of WP:CONEXCEPT to apply to MediaWiki namespace pages editable by the enwiki community is correct, especially in light of the VE precedent;
- Whether Eloquence's edits fall within one of the exceptions to our jurisdiction;
- If so, whether the committee should nonetheless take notice of them under the third paragraph of Wikipedia:Arbitration/Policy#Jurisdiction. T. Canens (talk) 01:14, 12 July 2014 (UTC)
- Accept to examine Eloquence's conduct, which it appears may have fallen below the required standard. On the jurisdiction question, I am of the mind that no user involved in this dispute was entitled to act as a deus ex machina. AGK [•] 23:41, 11 July 2014 (UTC)
- Of the arbitrator comments thus far, Floquenbeam's thoughts come closest to mine. Allowing a little more time for input on the request before voting. (If we do ultimately accept this as a case, let's please adopt a reasonably expeditious schedule, and stick with it?) More later. Newyorkbrad (talk) 15:10, 12 July 2014 (UTC)
- I've asked Armbrust and Fabrice if they plan to submit statements. Newyorkbrad (talk) 22:53, 12 July 2014 (UTC)
- Leaning decline largely per Floquenbeam (above) and David Fuchs (below) as well as the lack of clear answers in my mind to NativeForeigner's question (also below). I am not clear, at this stage, what helpful outcome a full case might have, though I remain open to persuasion. I am considering a motion addressing some of the issues raised, although that would be more for the point of emphasizing some points of agreement than sanctioning anyone. Thoughts:
- Going forward, I believe the WMF leadership has stated that it is a priority to improve liaison with editors on software/interface changes; obviously the community agrees, and it would be good to focus, via an RfC or otherwise, on how this can be done. Creating a group of editors, at varying experience levels, to work as some sort of focus group is one possibility that should receive attention.
- Second, there is clear agreement that WMF official actions should be clearly marked as such, preferably through a separate "(WMF)" designated account—this has largely been happening, but we can emphasize the desire for it where practicable.
- Third, although I don't enjoy wiki-legalism for its own sake (I am unlikely to discuss "jurisdiction to determine jurisdiction" in any document I'm not being paid to write, and bizarre overemphasis on process belongs Offwiki), it may be time for a community discussion on how our most important RfCs should work, particularly how they should be publicized and to whom (I hesitate to suggest an "RfC on RfCs" only from fear of infinite regress).
- Fourth, while we don't have the ability to tell the Office when they are authorized to desysop or threaten to desysop an administrator, common sense suggests that this should be reserved for extreme circumstances and be a last resort, particularly on a project like this one with an existing desysopping mechanism. (I understand the temptation they feel, of course: I've sometimes suggested myself when someone banned here is trolling as an admin on another project, "Why don't they just ban this clown as an Office action?" But that's different, he said, while not explaining why just now.) The Office's communications with our admins will not always ring with Eloquence, but neither should they be crass.
- With all that said, I'm not sure what would be left for a case. Newyorkbrad (talk) 13:49, 14 July 2014 (UTC)
- The majority having voted to accept, I defer to their judgment, and hope that this will be a productive exercise with a positive outcome for all concerned. Accept dubitante. Newyorkbrad (talk) 15:41, 16 July 2014 (UTC)
- Accept. I think that Tim's questions should be resolved. For example, although the arbitration policy is clear that the Arbitration Committee has no jurisdiction over the "official actions of the Wikimedia Foundation or its staff," it does not clarify if this is limited to OFFICE actions, or if it extends to everything that is claimed to be on behalf of the WMF. GorillaWarfare (talk) 02:12, 13 July 2014 (UTC)
- This is very clearly problematic, but I'll hold off on a stance on whether we should take this or not until I can adequately read the background and gain a good understanding of the issues at play. To some degree, I am sympathetic to those who say we should take the case, this whole mess is incredibly fuzzy, and if we wished to we could draw this out in terms of de jure and de facto powers (in the context of WP:CONEXCEPT). Though this isn't exact, it seems we have been asked to Arbitrate between the foundation and the community, if we can call it that. I'm not sure what we gain out of this. Yet I would disagree with Risker that T Canens questions are inherently answered. We can answer them via current practice, but my perception is the community, if only on this case page, has expressed issue with the status quo. However, as a counterpoint to that, it really should be the community causing these changes, but as of the present, they are being thrust upon us. (Additionally, the WMF still holds the ability to more or less define WP:CONEXCEPT.) WP:CONEXCEPT certainly defines scope of powers, but certainly not best practice for the foundation, and here I think that the recent series of scuffles certainly shows that a clarification of best practice on technical issues ought to be formalized. But again, my hesitation lies with the fact that I don't have a good idea of how this case can be well-resolved, and what action would be taken. (Although it is possible in this case, there is no good solution for us, even though we should take it.) Broadly speaking, what outcomes (concrete actions) would those asking for a case like to see? NativeForeigner Talk 07:41, 13 July 2014 (UTC)
- Accept with some trepidation. I'm somewhat reluctant to dive into such a gov'comish issue, but the community has requested a clarification, and one should be delivered. NativeForeigner Talk 01:40, 16 July 2014 (UTC)
- Decline Not that this is likely to matter, but after reading through everyone's responses, I don't see any avenue a case can be productive. I don't, as many have opined, see our role in performing judicial review on an RfC's outcome. No one is disputing that the case in question fell under the WP:OFFICE remit--it didn't. All that remains is whether you think this is a case of WP:CONEXCEPT, and I think it is. The WMF has the right to block such changes, but it should also have the common sense to acknowledge significant user feedback from its editors and work collaboratively towards a solution instead of threatening them with blocks, especially when the basis of the block would be invalid. But this doesn't seem to be within our scope. Perhaps a better question is to ask people what ultimate proposed decision do they see coming from this case, and what "evidence" can be provided besides multiple interpretations of a few lines of policy. Der Wohltemperierte Fuchs(talk) 13:24, 14 July 2014 (UTC)
- I've spent a lot of time thinking about this and when it comes down to it, I don't see that there's anything that Arbcom can do here. My thoughts mirror Floquenbeam's almost exactly, though many of the arbitrators make good points. We're not a "GovCom" and accepting this case would be yet another step to being one. I'm leaning decline at the moment. WormTT(talk) 10:34, 15 July 2014 (UTC)
- After reading through the above and the immediate background material, I think the concerns raised over this dispute are serious enough to warrant accepting a case of limited scope to help clarify what happened here and to reiterate some of the important principles at stake here. Carcharoth (talk) 01:09, 16 July 2014 (UTC)
- I've been thinking this through over the last couple of days and also believe that there is a limited role for ArbCom here. Accept, Roger Davies talk 06:41, 16 July 2014 (UTC)
This motion was proposed on 12 July 2014 but did not receive the required level of support.
|
---|
}}
MotioneditThe Wikimedia Foundation is requested to do away with "grandfather rights" regarding personal and official accounts of staff members, and to instead require all staff members to use accounts with a personally identifiable name with "WMF" appended when acting on-wiki in their capacity as staff. We ask them to extend this courtesy in order to reduce confusion regarding when our users are dealing with a fellow member of the community or a representative of the Foundation. Staff accounts, their actions, and their user rights or permissions shall be under the sole jurisdiction of the Foundation. Personal accounts of staff members shall be considered members of the community and shall be treated as such, including access to user rights or advanced permissions. While the English Wikipedia cannot make a binding decision on this matter we believe it is evident that making this change will ease communication and improve general relations between our project and the Foundation and we further believe the Foundation can institute this minor change with little to no disruption of their activities, and we further ask that they consider making this a requirement not just here but at all WMF projects. For this motion there are 14 active arbitrators, not counting 1 recused. With 0 arbitrators abstaining, 8 support or oppose votes are a majority. Support
Oppose
Abstain Comments
Community commentsedit
|
Case suspended (II)
editPassed on 00:03, 1 September 2014 (UTC)
This case was accepted to consider longstanding issues affecting the English Wikipedia community and the Wikimedia Foundation (WMF), which came to a head during the implementation of the Media Viewer extension.
Since then, the following has occurred:
1. The WMF has introduced a new staff user account policy, prohibiting the use of the same account for both work and non-work purposes. With effect from 15 September 2014, staff are required to segregate their work and non-work activities into separate work and non-work accounts, with the work accounts containing the identifier '(WMF)' in the account name.
2. Eloquence (talk · contribs) has resigned as an administrator on the English Wikipedia. While this does not prevent him holding staff administrative rights on a designated work account, it does mean that as he resigned the tools while an arbitration case was pending, he may only regain administrative rights on his personal non-work account via a successful request for adminship.
3. The WMF has announced a number of initiatives aimed at improving working practices. This includes a new software implementation protocol which provides for incremental roll-outs of upgrades and new features.
In the light of the foregoing, proceedings in this case are suspended for sixty days and then closed; in the intervening period, the case may be re-activated either by volition of the committee or if fresh issues arise following a successful request at ARCA.
- Passed 12 to 0, 00:03, 1 September 2014 (UTC)
Final decision
editAll tallies are based the votes at /Proposed decision, where comments and discussion from the voting phase is also available.
Principles
editFindings of fact
editRemedies
editAll remedies that refer to a period of time (for example, a ban of X months or a revert parole of Y months) are to run concurrently unless otherwise stated.
Enforcement
editEnforcement of restrictions
0) Should any user subject to a restriction in this case violate that restriction, that user may be blocked, initially for up to one month, and then with blocks increasing in duration to a maximum of one year.
- In accordance with the procedure for the standard enforcement provision adopted 3 May 2014, this provision did not require a vote.
Appeals and modifications
0) Appeals and modifications
|
---|
This procedure applies to appeals related to, and modifications of, actions taken by administrators to enforce the Committee's remedies. It does not apply to appeals related to the remedies directly enacted by the Committee.
Appeals may be made only by the editor under sanction and only for a currently active sanction. Requests for modification of page restrictions may be made by any editor. The process has three possible stages (see "Important notes" below). The editor may:
No administrator may modify or remove a sanction placed by another administrator without:
Administrators modifying sanctions out of process may at the discretion of the committee be desysopped. Nothing in this section prevents an administrator from replacing an existing sanction issued by another administrator with a new sanction if fresh misconduct has taken place after the existing sanction was applied. Administrators are free to modify sanctions placed by former administrators – that is, editors who do not have the administrator permission enabled (due to a temporary or permanent relinquishment or desysop) – without regard to the requirements of this section. If an administrator modifies a sanction placed by a former administrator, the administrator who made the modification becomes the "enforcing administrator". If a former administrator regains the tools, the provisions of this section again apply to their unmodified enforcement actions. Important notes:
|
- In accordance with the procedure for the standard appeals and modifications provision adopted 3 May 2014, this provision did not require a vote.
Enforcement log
editAny block, restriction, ban, or sanction performed under the authorisation of a remedy for this case must be logged at Wikipedia:Arbitration enforcement log, not here.