[WikiEN-l] Turn off AfD

Sat Dec 10 12:51:53 UTC 2005

On 12/10/05, John Lee <johnleemk at gawab.com> wrote:
> Anthony DiPierro wrote:
>
> >How do you deal with crap everyone agrees should be deleted, but is
> >not actually speediable?  Here's a thought - make it speediable!
> >
> >Anthony
> >
> >
> Are you kidding? A year ago, the paranoid community rejected two
> different measures for something *in between* speedy deletion and AFD
> (both involved admin discretion but not full reliance on it). What makes
> you think anything will have changed this year?

I *don't* think anything will have changed.  I said was talking about
what we *should* do, not what's going to happen.  If everyone agrees
something should be deleted, we should delete it, period.  Running it
through some AFD process first is not smart.

That said, if we turned off AFD, I think the number of speediable
categories would quickly grow.

Plus, in my opinion, Jimbo has set us into a new era of
experimentation on Wikipedia.  We no longer need consensus before we
can at least try something out.  Don't you think the paranoid
community would have rejected a proposal to disallow users that
weren't logged in from creating new articles?  I certainly do, in fact
if you want consensus I don't even think the community would accept
such a proposal now (maybe it'd get a majority, I really don't know).

> If people won't trust
> admins to semi-speedy something, why would they trust them to speedy
> something?

I don't understand that question at all.  Why should the votes of a
few non-admins and a few admins override the consensus of all admins? 
If all admins insist on keeping something deleted, then there's
nothing that can be done about it.

> If anything, considering how we've grown, I wouldn't be
> surprised if people would vote down proposals like Preliminary Deletion
> even harder. People are scared of the potential for admin abuse. And
> frankly, I don't blame them.

As long as the ability to delete remains possible, the potential for
admin abuse is very small.  What is the potential abuse here?  GNAA
gets deleted?  I actually doubt it would.

You know what, forget the AFD experiment.  Let's give admins free
reign over deletion for one week.  Surely they won't vandalise
Wikipedia to the point where we can't fix things after the week is up.
 After one week, we look at what got deleted, and we see if there was
any admin abuse, and if so how severe it was.

AFD is still there, but it's there for actual disputes.  Preferably
it's there mainly for discussion about the underlying issues of what
should and shouldn't be deleted, and doesn't get bogged down in
minutiae of every single article (in some especially heated cases
maybe there will be a vote over a single article, but not hundreds a
day).  AFD doesn't get it right perfectly 100% of the time.  There's
no reason admins need to either.  As long as there aren't gross
violations of the standards which we reach *through consensus*, it
really doesn't matter.

We'll probably wind up with somewhat more deletions this way, and
really I see that as a bad thing.  But as AFD has pretty much
completely abandoned the notion of consensus anyway, it's probably
inevitable.

> Scandals like the bitter edit war over
> licence vs license in a Mediawiki template have proven that, sadly, our
> admins can and do get into trouble. Even the best of us, like Everyking,
> get too heated and involved in articles we edit that we cause pointless
> edit/revert wars.
>
This is actually more a problem with the enormous reluctance to take
adminship away from people.  In my opinion (and this is really a
completely separate topic), the bar to adminship needs to be lowered,
and the bar to deadminship needs to be lowered as well.  The real
question is how can this be achieved.  I don't think it can be
achieved through voting, at least not voting among a large group of
people, because adminship and deadminship should not be based on
popularity, and large groups of people invariably factor popularity
into a vote (because there is essentially no *accountability* in
voting).

> An objective measure of an article's "deletability" would be one nobody
> could disagree with.

If you want perfection...  Of course, if you just want "better than
AFD", it'd only need to be a measure that more than 2/3 of people
would agree with.

> For instance, you can't disagree that, say,
> fuddlemark got >90% support on his recent RfA, because that's an
> objective measure of support. (At least, you can't disagree without
> appearing insane.) However, many articles on AfD that get deleted by
> unanimous support have their "deletability" measured by subjective
> methods. To some people, schools are notable. To others, they are not.
> How do you resolve this?

Well, I'd resolve it through compromise.  Keep the articles somewhere
outside the encyclopedia, somewhere which isn't controlled by a
for-profit corporation and somewhere that respects the GFDL.  That
leaves two possibilities, really.  We put it in a sister project, or,
if the board won't let us start the sister project, we start a new
organization and put it there.  Now that we've moved the information,
we change the links to that article from red links into light blue
links, that is, external links to the other source.

Sure, not *everyone* will be happy with this, but I think most people
would find it acceptable.  And sure, we wouldn't wind up with exactly
the same breakdown as we'd get through AFD.  But that second point I
really think is irrelevant.  AFD is not perfect.  Even those that
really like AFD only feel that it gets things approximately right.  If
a few borderline cases get decided differently from AFD, it's not the
end of the world, especially not if the information is still preserved
somewhere else, and maintained under the GFDL so that we can retrieve
it back, most likely in better condition, if we change our minds.

> How do you objectively measure a school's
> "deletability"?

Personally I don't feel that any schools are deletable.  I think
enough people agree with me that one couldn't legitimately say that
there is *consensus* to delete an article on any school.  That some
get deleted through AFD is simply an unpatched flaw in the deletion
system.  In any case, we've strayed off the original topic.  This
particular subthread got started when someone claimed that there were
articles which *everyone* felt should be deleted but wouldn't be
because we couldn't come up with an objective measure.  Articles on
schools don't fall under this category.

> Is "deletability" even an objective value in itself?
> Truth be told, it's a miracle our CSD have grown so big, because even
> they aren't objective. After all, how many of them received unanimous
> support? Clearly some people disagreed with them, and probably with good
> reason. This would inherently make them subjective.
>
> John Lee
> ([[User:Johnleemk]])

True, though CSD deletes quite a few articles which not everyone feels
should be deleted.

Anthony