[WikiEN-l] Reversing admin actions, was: ArbCom Legislation

[Abd ul-Rahman Lomax]

At 01:09 PM 6/18/2008, Carl Beckhorn wrote:
> > A single reversal of an admin action isn't wheel-warring.
>
>I would argue it is, although historically the WP:WHEEL page has been
>written to permit it. I would argue that the definition of wheel warring
>that permits one unilateral reversal of any action is detrimental to the
>project. It short-circuits dispute resolution in favor of the status
>quo; it actively discourages those who disagree with an admin
>action from coming to compromise with the original admin.
>
>I'm not focusing specifically on BLP here, I'm thinking of the
>general principle. Once an admin action is performed to which there are
>objections, the correct course of action is to find a compromise thorugh
>discussion.  This is no different than any other disagreement on
>Wikipedia.

I generally have in mind deliberative traditions. Wikipedia is 
designed for minimum bureaucracy, allowing many decisions to be made 
ad-hoc by individual administrators, who are not required to discuss 
them first. Any administrator, seeing an act of incivility after 
warning, can accuse and block at the same time. Often this will 
simply stand, because, most administrators being reasonable, it is 
efficient to allow this.

However, this was done without discussion. It then becomes legitimate 
for another administrator, seeing what appears to be an improper use 
of the tools, to unblock. Courtesty requires some level of 
communication on this, but it may be enough that the administrator 
reads the warning (or notes the lack of same), reviews the underlying 
incident, reads the block summary and other notes from the blocking 
administrator, and concludes that the block was improper. I'm stating 
that this is essentially the simplest possible review process, and 
that it should be able to conclude, ad hoc, that the block was 
improper and reverse it.

If the original administrator agrees, or at least consents, done. 
Presumably the unblocking administrator will likewise document his or 
her reasons, at least briefly.

This is similar to speedy process in deliberative assemblies. One 
example might be that a chair sees some action that the chair 
considers beneficial, and says, "Without objection, we will A, and is 
about to bang the gavel." If nobody objects, fine, it's done, 
immediately and without any fuss. But if someone object, *no action 
is taken," and if it happens that that hammer came down, the 
objection will still be taken and the action reversed, pending 
discussion. The original action and the discussion don't prejudice the result.

In deliberative process, no formal debate takes place without a 
second. If all we have is two people who disagree, it's a stalemate, 
and neither of these people have any right to impose their desired 
solution; except that there is a default, which in the case of 
editors, is that they aren't blocked.

I don't agree that if an administrator disagrees with a block, that 
they should go to AN/I or the like to get opinion. That's an option, 
but if the matter seems clear, and if the objectionable action was 
only that of a single administrator, unblocking seems most efficient 
and fair and balanced. It establishes an objection, and further 
action should require wider participation, which either administrator 
or anyone witnessing the incident can facilitate. I think of 
"wheel-warring" quite the same as "edit warring." We don't have an 
edit war if someone makes an edit and someone else reverts it. Once. 
It's when it goes back and forth, especially without attempting to 
seek and find consensus, that we start to call it edit warring.

I saw one example of a single unblock being considered wheel-warring, 
and it was a terrible example. Jimbo blocked an editor and 
disappeared, asking that the matter be stayed until his return. Bad 
example, Jimbo, not your finest day! After some discussion, and with 
quite a bit of disagreement in the community, an administrator 
unblocked. And, of course, all hell broke loose. I never did figure 
out why the block was considered so important that it couldn't have 
waited, but .... maybe there was a reason, I should review that case, 
I forget the names involved at the moment. In any case, another 
administrator reblocked. The sky did not fall. There was, indeed, 
wide discussion, and the only reason such a fuss was created was that 
the blocking admin was Jimbo who made a condition that nobody else 
would have made, I think, and the unblock was heresy.

BLP policy, legitimately, I think, makes the exclusion of possibly 
defamatory material the default. And because damage can be done by 
such material with every minute it stands, removal of such material 
could be considered the "status quo." However, what if the material 
is long-standing? The matter is too complex to reduce to simple 
bureaucratic rules. I think that administrators understand the 
principles, and can be trusted to generally get it right. Defining a 
single reversal of an administrative action as wheel-warring is going 
way too far.

It might still be improper. If an admin protects a BLP, and the 
protected article is inoffensive, unprotecting it to insert 
controversial material, without wider discussion, would be dangerous 
and could be administrative abuse. It still isn't wheel-warring, 
because all administrative actions -- it's very important --, in the 
presence of objection, don't stand unless backed up by wider 
consensus (in particular, to the extent that no administrator is 
willing to reverse it. That any administrator can reverse it, once, 
is a huge protection against abuse, just as we allow 1R even in 
articles under probation. If we allow "anyone to edit," we must allow 
"anyone to revert," in general, and only sanction going beyond that 
without discussion and broader consensus, if possible.)