[Foundation-l] Re: CheckUser questions

Wed Nov 9 10:24:05 UTC 2005

Aphaia wrote:

> Hello,
> Danny's mail brought rrelevant questions rather than comments to me
> which I've held since last night after a communication on JA WP Pump.

I can clarify some of those.

> On 11/9/05, daniwo59 at aol.com <daniwo59 at aol.com> wrote:
> 
>>Although I have not been following this discussion, I was asked this  evening
>>to grant 5 arbitrators on the English Wikipedia CheckUser powers.
> 
> 
> Although I don't know not all those five closely, personally I have no
> opposition to any of them, however,  frankly I was a bit surprised by
> so many people requested (I found it first on meta), because I
> understood only two people could be granted it on a local project.
> 
> As for the number of checkuser priviledge holders, I understood as the
> below: if a project is large but lacks its own arbcom, two users (and
> not one, nor more than two) can given the checkuser right. Yesterday
> an editor said "more than two: would be given it Perhaps I was wrong.
> Or not. If any explanation is given, this guy (or girl) will be very
> appreciated.
> 
> Let me continue a talk on assumption; if more than two may granted on
> a project, it isn't worthy for a surprise that five people request for
> checkuser priviledge. Otherwise, a question arises; a big project with
> arbcom can have more than two checkuser granted editors. a rather
> small but still big project without arbcom (like Ja or De) can be with
> two such users. It seems to me somehow strange.
> 
> For your information, few months ago JA WP had a discussion about
> checkuser request. Then we argued how many checkuser group editors
> would be needed; a certain opinion said three would be the best mainly
> due to availability and frequency of such investigation.
> So "two' or "more than two" is not a petty thing to that community.

The "two" is a minimal number and is there to ensure that there are at 
least 2 editors who can watch each other activity. This was proposed to 
reduce the impact of lack of public log.

There is no maximal limit.

As a reference, which might be interesting, David makes checks every 
day. Yesterday (a bad day I guess), he made over a hundred checks. At 
this rate, clearly, more than 1 and actually more than 2 must have the 
tool in the hand.

As a second interesting reference, the second biggest project (dewiki) 
has it seems never made a request (even to a steward). The third 
community (frwiki) had two "valid" requests in a month or so. The fourth 
(jawiki) never asked either. Seeing Oscar and Waerth log activity, the 
nlwiki is ready to use the tool as well.

Might be interesting from a cultural perspective though...

>>While the request was made, I think it is sage for us to let it sit there
>>for some time to see if there are any valid protests to giving it to these
>>individuals. Usually, steward requests can remain for several days until someone
>>acts upon them, and I don't see why this should be any different.
> 
> 
> One interesting point to me, the request submitted Raul mentioned to
> Jimbo's participation to the discussion about request. I have no
> argument about it; it is simply nice for such participation and
> approval. On the other hand, such participation couldn't be expected
> on most projects couldn't. It means in most cases the Board and
> Foundation has to rely on  their credibilitygiven by a community -
> another question arises here to me, wheter "70%-80% approval with more
> than 20-30 supports" is truly acceptable (Not personally, just a
> question).
> 
> A relevant thing: a sysop election on a big project can gather
> sometimes over 50 or close to 100 support vote (like vote on elian
> last year) - I am not sure if the current criteria is enough for some
> meta-project with 10,000+ registered users. If everyone is happy, no
> problem.

The numbers were proposed to limit the access for small projects with 
only 2-5 editors. Two reasons for this
* first, the needs are less important, and can be handled by steward 
ponctually.
* second, on small wikis, sysophood is just given with no vote right 
now. And we have several times observed that the new sysop aquire a sort 
of special status, with special rights over the others. On some 
projects, the sysops are the only ones authorized to vote over 
decisions... on some projects, sysops are free to block a page on which 
they are in edit war with someone... on some projects, the bureaucrats 
decide who shall be sysops :-) This shall pass as the community grow. I 
would just prefer that tiny projects with a handful of editors have the 
time to gather a real community before having access to the tool. So the 
criteria is just here for those tiny community.

Any community with more than 20 editors active enough to vote is a big 
and diverse enough community.

>>I also wonder whether they should be given the power on a permanent or a
>>need-to-know basis. That is, should they always have this ability, or should
>>they only be given it when a case requires invesigation, and then have it
>>removed? Personally, I favor the latter option. I do not think anybody needs to  be
>>able to do this in all instances.

> Sounds wise. One concern is time expirinng and steward availability.
> In some circumstances investigation should be done within a day or
> hours (like emergent request for desysoping). If stewards don't care
> to be bothered with frequent on-time basis requests, it would be
> prudent. On the other hand, if they are so trusted, it would be
> unlikly problematic to give them permenent right.

Well... on enwiki, the checks are done everyday. It would be unworkable 
to have to restore status everyday...
Now, I suppose that if some requests are done on a temp basis by editors 
we generally trust very much, we could just give them a temp status ?

> --
> Aphaea@*.wikipedia.org
> email: Aphaia @ gmail (dot) com