[WikiEN-l] Major dysfunction in RfA Culture

[Earle Martin]

On 15/04/07, geni <geniice at gmail.com> wrote:
> > What has this to do with quiet admins being harmful to the project?
>
> If you had a large body of fairly active admins such issues would be
> less likely to arise.

You have yet to convince me that these "issues" are in fact real
issues. People learn. The project goes on.

> > Are you trying to imply that factions would develop in the admin
> > community based on how much work people do? Sorry, but I think that's
> > ridiculous.
>
> So you are suggesting that there will be no split between those with
> actual practical experience and those who do not? That goes against
> almost the entire history of human activities.

You're answering a question with a question, and one that distorts
what I actually asked you in the first place. I hate having to do this
in discussions, but it seems to be time to drag out the dictionary:

faction, n.:
1. a group or clique within a larger group, party, government,
organization, or the like: a faction in favor of big business.
2. party strife and intrigue; dissension: an era of faction and treason.

or

A group of persons forming a cohesive, usually contentious minority
within a larger group.

(http://dictionary.reference.com/browse/faction)

My question to you was whether you are saying that factions will
develop, not whether there will be "a split", which only implies a
difference. It certainly appears that this is what you are saying.
This suggestion, as I see it, is absolute hogwash.

> > > 5)security  risk (admin accounts getting hacked) without the gain.
> >
> > This is conjecture and scaremongering.
>
> No basic maths.
>
> [math nusked]
>
> So the chance of an admin password being acquired = mean x*number of admins.

The chance of an admin password being acquired is the chance that of
someone simply downloading the list of admins right /now/ and managing
to crack one of those users' password. Even if no more admins were
ever created, this chance increases over time as more and more
crackers will try their hand at it. What you are doing is trying to
recast an existing failure of our code's security model as somehow
being the fault of admins for... being admins.

In fact, now I think of it, you are implying that we should have
/less/ admins, because that reduces our chance of being violated. Well
done for losing your adminship, then; you have increased the security
of our project.

I also fail to see why you attach such significance to admin access
being obtained; Wikipedia has a spectacular backup system at multiple
levels of granularity, and the damage that one compromised admin
account could do is insignificant compared to the methods available to
fix it after the account is killed. (Don't bother telling us now that
you know how you could bring the site to its knees; seriously, we've
heard it before, and we know how much you love dropping hints about
it.)

If an account is compromised, the techies will implement additional
security to make it that much harder next time (https logins, anyone?)
and life will go on.

In a reply to someone else, you wrote:
> Fails to provide protection against key logging and in transit interception.

It also fails to provide protection against criminals who kidnap you
and put a gun into your mouth to make you give them your password. So
what? It doesn't matter, for the reason I outlined above.

> We can't promote at a high enough rate to deal with bumps in need for
> admin actions. the only way to meet these is for admins to become more
> active. fairly active admins would appear to be the best candidates
> for increase action rate.

Or a big increase in the number of admins. As others have commented,
you seem to be expecting that the current crop of keen admins become
robots. This is not a sustainable concept, and you shouldn't expect it
to happen.

What will happen is that RfA will change, and there will be more and
more admins, of all levels of activity. If you don't believe me, wait
and see.


-- 
Earle Martin
            http://downlode.org/
http://purl.org/net/earlemartin/