[WikiEN-l] declining numbers of EN wiki admins - The theory that making it easier to get rid of admins is a solution to the decline in their active numbers

[Abd ul-Rahman Lomax]

At 06:43 AM 5/30/2010, David Gerard wrote:
>On 30 May 2010 11:36, WereSpielChequers 
><werespielchequers at googlemail.com> wrote: > As 
>for the idea that we should move to "Hi, I 
>noticed that you > speedy-deleted some files 
>that do not appear to meet the CSD criteria; > 
>your SysOp staus has been removed _while we discuss it_".

By arguing in this way those with elevated status 
have maintained it, thoguh that seems to be 
falling apart. Consider the situation described. 
Obviously, the one writing this is a bureaucrat, 
highly privileged. If we think that there is a 
bureaucrat would would casually *remove* admin 
status over some simple errors, we have a problem 
with that bureaucrat, and, as with anyone else, 
perhaps process should be initiated!

Bureaucrats, though, would only remove status, 
absent emergency, if proper process had been 
followed. Certainly that notice would not be the 
first notice to the admin! Or if it was, and if 
removal was immediately, the admin was massively 
deleting, in a way making undoing it burdensome, 
and the desysop was as an emergency, and would 
normally be temporary until the admin agrees to stop.

By taking proposals for efficient and easy 
desysopping to ridiculous extremes, suggesting 
nightmare scenarios that would be highly unlikely 
to occur, many in the community have been able to 
prevent the system from being improved. It's 
obvious. And it demonstrates that there are 
editors who have a concept of an oligarchical 
core, to which they belong, with the continued 
power of this core, even when it's against true 
consensus, being critical to the future of the project. And that's a problem.

>  I've done > over 4,000 speedy deletions, and 
> very probably there are more mistakes > amongst 
> them that I know about, but if someone thinks 
> I've deleted > something in error I'd expect a 
> first approach along the lines of > "would you 
> mind having another look at [[deleted 
> article]], Â I don't > see how it was an attack page".

That's right and that's quite what happens, and 
the existence of speedy suspension process (much 
better and much less punitive than 'speedy 
desysop') would not change this at all.

>  Â Maybe I've made a mistake, maybe so > much 
> has been oversighted that it no longer looks 
> like an attack page, > maybe there are words 
> involved that have very different meanings to 
> a > Yank and a Brit. But a desysop first and 
> ask questions later strategy > would in my view 
> generate far more drama than would be justified by > the results.

I.e., straw man. The first step in a process 
might be a request to suspend usage of tools in 
some area. It would never be punitive, i.e., "You 
made a mistake, therefore you are no longer a 
sysop." What idiot would propose that? Rather, 
the legitimate concern would always be the 
likelihood of repetition. When it becomes likely 
that an admin will make many errors, such that 
cleanup becomes more work than allowing the sysop 
to continue with tools, *then* removal of tools 
becomes appropriate. I would assume, instead, 
that suspension requests would be handled 
routinely, and normally, a reasonable suspension 
request would be handled with little fuss, it 
would be much more like what David describes as 
what he expects. It is only if the admin contests 
this and insists on personally using tools in the 
area, against maintained opposition by other 
editors, and, then, particularly by editors who 
might be eligible to take part in some formal 
process to suspend (partially, with voluntary 
compliance) or remove tools (i.e., if voluntary 
compliance isn't forthcoming), would there be an 
issue of conflict and actual removal. And then 
the (now former) admin might get that note from a 
bureacrat who reviewed the process and concluded that removal was appropriate.

>  Indeed. The first - and, I would have thought, 
> jawdroppingly obvious - result would be that 
> no-one at all would go near such work in any circumstances.

Of course. It would be even worse if we chopped 
off the hand of any admin who blocks, say, 
another admin or makes any other error, as we 
think. But why in the world would we imagine that 
an efficient and fair removal process would look like this?

Look, if I'm offered the position of volunteer 
custodian at my daughter's school, but I find out 
that some other volunteer made so many mistakes 
that they were asked to stop, would I decline on 
that basis? Losing tools is not a flogging, 
indeed, it's only like a flogging if one resists 
it and believes it's the end of the world if one 
can no longer block editors, delete articles, and the like.

It's not even an important part of most editor's 
work, but, unfortunately, it does become an 
important part of some admin's work. Some have 
suggested that admins should be required to 
maintain good article work. I disagree, because 
some people might be *better* as admins than as 
article aditors. But "better" doesn't mean that 
they control the articles, and, indeed, it should 
mean the opposite. It would mean that they 
encourage cooperation among editors, defuse 
disputes, using blocks judiciously and without 
inflaming and expanding disputes with them. We 
allow, in the U.S., police to wear guns. But any 
police officer who is firing the gun, or even 
just pulling it out of its holster and pointing 
it at someone, frequently, is liable to be 
dismissed or worse as dangerous. Administrators 
are supposed to have no special privileges as to 
deletion of articles, personally, as to their own 
vision of what the project should be. But some 
admins do, in fact, use their tools to further 
their own agenda and POV, and I took that one to 
ArbComm and prevailed, and it was useless in the 
end. The admin was admonished, and then, not 
being desysopped, retired. And then returned and 
requested return of tools. Because they were not 
removed "under a cloud," technically, he was able 
to get his tools back. I've seen no similar 
violations from him, though, but having admin 
status has allowed him to have influence in the 
community that has been, on occasion, just as 
damaging. Pursuing the same POV as before.

Administrators are, in fact, specially privileged 
over content and behavior, and adminstrators 
frequently engage in behavior that would get 
another editor immediately blocked. That's part 
of the problem. Jimbo, even, tried to address it, 
and a huge fuss was raised, by admins who don't 
want any restraint on their power, and by those who support those admins.

>  The problem with RFA has long been arbitrarily 
> increased standards, and in recent years the abusive nature of the gauntlet.

That's part of the problem. But it is because it 
is so difficult to remove the tools that the 
"gauntlet" became so abusive and the standards so 
apparently increased. It was pretty stupid, 
because there is no way to anticipate how an 
ordinary editor will behave with the tools, or, 
at least, it's extraordinarily difficult. There 
is an obvious solution that, however, will be 
opposed by those who have gained admission to the 
privileged group, because it will dilute their 
power. It's natural and instinctive as a 
response, I don't necessarily blame them. We can 
see this in the votes on the community desysop 
proposal. (Which was, by the way, a lousy 
proposal in my view, far too reliant on our 
heavily dysfunctional discussion process. DGG has 
it right.) It looked like the proposal was being 
massively rejected, but when administrator !votes 
were set aside, it was about fifty-fifty. My 
guess is that a better proposal might even pass.

And the solution is to make removal much easier, 
so that when it's approved in the first place, 
that approval can be undone *by those who 
approved it.* Under Robert's Rules, it's called 
Reconsideration. And a motion to reconsider must 
be made by someone who approved the motion in the 
first place. That's designed to avoid frivolous 
requests for reconsideration....

I'd suggest something like this: a standard 
"admin recall" agreement is worked out. This 
could be *very* efficient and at the same time 
very unlikely to be abused; having those who 
support an RfA become some kind of recall 
committee is one idea. If that approving number 
is smaller because it becomes easier to pass RfA, 
I'd only be worried about it becoming a factional 
committee planning on using the admin to further 
factional goals, but this would not be the only 
way for an admin to lose tools, in the first 
place, but also there would be ways to avoid 
that, and it's possible that a closing bureaucrat 
would, for example, appoint a committee from 
among those who approved and who were willing to 
"monitor" the situation with the admin, at least 
for a while. I won't go into more detail, but 
will note that I can anticipate piles of 
objections, and the problem won't be fixed until 
we realize that *any proposal can generate 
objections,* but some of the objections might 
easily be met with features, and some are merely 
imagination as with the idea that someone would 
just remove tools, as an individual, as described 
above, without there being some safer process. 
(But, of course, any bureaucrat or other highly 
privileged user can already do this, and 
sometimes they do, on an emergency opinion.)

Then, perhaps a consensus develops that not only 
new admins but also all admins should agree to 
this process. Nobody would be punished, per se, 
by refusing, but refusal would then call 
attention to the admin, and the admin's actions 
might be reviewed.... and I could imagine some 
case filed at ArbComm requiesting the removal of 
tools en masse from administrators who had not 
agreed to a community consensus on recall 
process. Exceptions could then, obviously be 
made, but if "removal" was merely a default 
suspension, overcome by agreeing to the "pledge," 
I fail to see how it would actually be harmful. 
There would be no denial of the already-existing 
and valuable contributions of the administrator, 
only a realization by the community that 
different standards may be appropriate for the 
future. There might not even be an actual 
removal, but an admin might be treated as if the 
pledge were in effect, i.e., that process might 
be followed anyway, and it would be up to a 
bureaucrat whether or not to respect it, with 
appeal being possible to ArbComm. The same ad hoc 
process that often works with articles could work with this as well.

Expect many existing administrators to make sure 
to vote against any such proposals. Part of the 
problem is that the active core is top-heavy with 
administrators and wannabe administrators.... 
However, many admins are realizing how impossible 
the status quo is, so it's always a possibility 
that sanity will appear and prevail. 
Unfortunately, most of the admins who wake up and 
realize how bad the situation has become instead 
retire, they may have burned out before realizing 
the problems. Others simply become abusive in frustration....