Case Opened on 15:52, 18 August 2007 (UTC)
Case Closed on 03:40, 27 August 2007 (UTC)
Please do not edit this page directly unless you wish to become a participant in this case. (All participants are subject to Arbitration Committee decisions, and the ArbCom will consider each participant's role in the dispute.) Comments are very welcome on the Talk page, and will be read, in full. Evidence, no matter who can provide it, is very welcome at /Evidence. Evidence is more useful than comments.
Arbitrators, the parties, and other editors may suggest proposed principles, findings, and remedies at /Workshop. That page may also be used for general comments on the evidence. Arbitrators will then vote on a final decision in the case at /Proposed decision.
Once the case is closed, editors may add to the #Log of blocks and bans as needed, but closed cases should not be edited otherwise. Please raise any questions at Wikipedia:Requests for arbitration#Requests for clarification.
On 14 August at about 10:00 UTC, Alkivar deleted Wikipedia:Bad Jokes and Other Deleted Nonsense and around two hundred subpages that were part of BJAODN. log. About 8 hours later, Georgewilliamherbert restored the pages without first discussing with Alkivar. log Alkivar's talk page GWH's explanation on Mackensen's talk page
The existence of BJAODN had been discussed numerous times on the Admin noticeboards with no particular consensus to delete. In June, Jeffrey O. Gustafson (talk · contribs · blocks · protections · deletions · page moves · rights · RfA) deleted BJAODN. After 3 days of discussion on the noticeboard, it was restored by Sj (talk · contribs · blocks · protections · deletions · page moves · rights · RfA). Therefore, Alkivar's actions were at least unwise and possibly an error in judgement. By reverting the deletions without consulting Alkivar or going through deletion review, Georgewilliamherbert wheel-warred. This had led to the predictable escalation of wiki-drama:
The deletion and restoration of BJAODN in June arguably represents another instance of wheel warring, although at least here the restoration was a couple of days after the deletion, following discussion on the Admins noticeboard. I suspect the actions in June are too stale for consideration but they are the context in which yesterday's deletion and undeletion ocurred.
There has been a growing disregard by admins for each others' actions, leading to minor wheel wars over various minor issues, and several other admins have commended Georgewilliamherbert for his action. It seems that the community of administrators will not take admonishments to avoid wheel-warring seriously until some admins get spanked. I will not be popular for suggesting this, but I suggest desyopping Georgewilliamherbert for 10 days for wheel-warring. Any temporary desysopping of Alkivar should be of a shorter duration as single instances of bad judgements are generally not punished. Admins need to have respect for each other's actions and consult before unilaterally overturning each other's actions, unless it is an emergency. I am not confident that the community of administrators will take the Committee's warnings about wheel-warring seriously unless there are definite consequences for overturning another admin on a non-emergency basis. Thatcher131 15:06, 15 August 2007 (UTC)
Comment on whether one revert equals wheel-warring
How many administrative reversals does it take to make a wheel war? I am under the impression that ArbCom looks very dimly on admins reversing each other without discussion and consultation. Certainly, I feel this is poor practice as it can lead to (or is perhaps already a symptom of) a loss of respect, deference, and collegiality. If it takes two reversals to make a wheel war, that means that any admin action can be reversed without consequence. The first reverting admin (in this case Georgewilliamherbert) would always "win" because a wheel war would occur only if the original admin (Alkivar here) reinstated his action. The principles on deletion and undeletion here seem relevant as well. I do agree that "Wheel-warring" is an inapt name for a single action. (Although only in a specific narrow sense does the deletion or undeletion of over a hundred pages constitute a "single" action.) I suppose that whether the Arbitrators consider a single administrative reversal without discussion and short-circuiting the proper venue to be actionable will be revealed by their votes. Thatcher131 18:38, 15 August 2007 (UTC)
I am both glad this larger issue is being addressed at higher levels and confused as to the specifics of the case being proposed against me.
Wheel warring requires two or more admins to use admin-specific functions in repeated conflicting actions. There has been no repeated use of admin functions. One operation was performed (delete) on a set of articles, and then one operation was performed (undelete) on a large subset of those. There have been no repeat deletes or repeat undeletes by anyone as of last I looked.
Suggesting that responding to an admittedly highly controversial WP:BOLD admin action with an admin rollback constitutes wheel warring seems to me to be a misinterpretation of longstanding policy. Admins are allowed and encouraged to use their initiative (the longstanding BOLD policy), however that's coupled with a "...until someone pushes back..." caveat. I believed Alkivar went too far (once) and reverted (once).
As I discussed on Mackensen's talk page, there was nothing sterile about my actions. The order they were performed in did start with undeletion and then proceed to notification. Alkivar was as far as I can tell not logged in at the time. Shortly after undeleting BJADON I left a note on his talk page, as my next edit in fact. I then noticed that a DRV had started and once that appeared to be gaining momentum stopped the subpage undeletions I'd started.
I have also promptly responded to all the questions posed to me on my talk page, discussed the situation promptly on WP:AN and on DRV.
Policy strongly encourages us to discuss another admin's actions with them prior to overturning, if we believed they were in error. In this case, Alkivar was not apparently online, and I believed that the magnitude of the controversy of the actions justified taking the more aggressive but reasonably polite undo-first-notify-and-discuss-promptly-afterwards approach.
WP is striving to be neither too legalistically policy bound nor too free-form and ruleless. We have WP:BOLD, but understand that bold actions need to be undoable by admins who strongly disagree, and we have a community consensus and policy mechanism as well. Real wheel warring, particularly sterile wheel warring, is a breakdown of those processes. This incident was not a breakdown. We had one controversial delete (page + subpages), one restore, and the community process kicked off to review the situation about the same time as the restore. That process is proceeding normally so far.
It has been proposed in the DRV that this issue is too controversial for normal community process to come to a proper conclusion on what is to be done with the BJAODN pages, specifically referring to disagreements about what we can or must do about the GFDL issues with credit and edit histories. If so, it may be necessary for Arbcom (or higher authority such as Jimmy, the Board, etc) to review the situation and make a decision on what to do about it.
I believe that Alkivar and his supporters feel that this was too controversial and required admin-level initiative to simply impose a solution. I disagree - this is too controversial to simply impose an admin-level imposed solution. If we cannot come to a workable consensus on DRV or MFD or wherever it ends up, then I support using WP's processes and higher authority to answer the questions.
But trying to simply impose a BOLD solution to something of this magnitude, literally in the dark of the night, was very badly the wrong way to do this. It needed to be undone. It may well be true that either process or higher authority will decide to re-do the deletions, but a legitimate decision on this issue cannot come from any one admin's personal actions.
Georgewilliamherbert 18:16, 15 August 2007 (UTC)
Comment on comment on whether one revert equals wheel-warring
It is neither consistent with prior policy nor common sense to require that an administrator always engage in a two-way conversation with another administrator before reverting any admin action. For one, the first admin may not be active, and it's unreasonable to expect that admins wait an arbitrary length of time given an apparent problem which is serious enough that undoing it seems like the right solution. Alkivar was idle for a long period of time and did not subsequently make another edit for a long time after the undeletes in this particular case.
The proposition that BOLD admin actions are immutable absent two-way discussion or community process seems absurd and inconsistent with the way everyone's operating their admin accounts.
There are clear and evident reasons to prefer discussion first - it reduces the risk of true wheel wars in the multiple-repeated-admin-actions sense, and it's polite. Had Alkivar evidently been active at the time I would have attempted to discuss with him. But he evidently was not, and as has been admitted by the filing party, the original action was highly controversial.
In this specific case, my acting-first-discussing-second did not lead to a true wheel war (only one delete and one undelete per article) and would not have (I would not have re-undeleted if someone else deleted, doubly especially as the DRV process started rolling effectively).
Alkivar is welcome to let me know if he considers my actions rude; I was pretty terse, but I have tried on his talk page and elsewhere to not personally attack him over this. I believe that he made a mistake, but I still respect him and his contributions, and I don't think that desysopping him for the mistake is appropriate at all, even temporarily, though that's Arbcom's call.
If I had to hazard a guess, I bet he's embarrassed at the resultant drama. I am ... not exactly embarrassed, but frustrated at the drama aspects of how this is playing out on DRV; a significant fraction of both the keep deleted and overturn deletion arguments put forth are not helpful. Which was probably predictable given the prior round of discussions a couple of months ago.
Georgewilliamherbert 19:12, 15 August 2007 (UTC)
Comment on GFDL issues
I have just said this in the DRV and am including it here for completeness.
BJAODN is the only section of the encyclopedia where questions about being able to trace the source of content moved from one page to another have ever seriously come up. This does happen on a regular basis, and nobody tracks what is being done well or consistently. If this is in fact a legitimate issue we need to address, the way that large portions of the editors and admins edit and expand the encyclopedia will have to be reviewed and corrected, and large portions of the encyclopedia are vulnerable to history-tracing challenge.
Maybe we legitimately have to do that, but please acknowledge that if you open that can, there are many many worms in it beyond BJAODN, and that the worms are carnivorous.... Georgewilliamherbert 23:02, 16 August 2007 (UTC)
Thank you, — xaosflux Talk 02:43, 17 August 2007 (UTC)
1) As the underlying dispute has been satisfactorily resolved by the community, and as no evidence of bad-faith actions by any party has been presented, this case is closed with no further actions being taken.