AW: [Members] Memberbot

Peter Saint-Andre stpeter at jabber.org
Tue Sep 12 12:20:48 CDT 2006


Alexander Gnauck wrote:
>> I think the most important thing is getting a system that's fair.
>> At the moment the memberbot method is horribly horribly biased
>> towards the first five candidates: despite being one of the very
>> strongest candidates, I'd be very suprised if Remko (the last
>> candidate,
> 
> I don't think so. When you pinged memberbot it gave me all the
> instructions and a link to the Wiki Pages. It tells you that you have
> to vote for 5 persons out of 8. I don't see where this is biased.

If someone wants to write a different bot that randomizes the order,
that'd be fine by me. But IMHO it is too late to change the voting
process for this voting period.

> I also wondered when memberbot told me: "vote on 5 out of 8", because
> i can't remember when we had more than 5 council candidates in the
> last years. And at the very beginning of the JSF there wasn't the
> restriction to 5 seats in the council. The 1st and 2nd council had 9
> members. So i started a quick chat with stpeter to clarify.

The 1st and 2nd Councils had 9 members (2001-2002 and 2002-2003). This
was modelled on Apache but it turned out to be not very efficient (hard
to get everyone in a chatroom at the same time, difficult to collect all
votes, etc.). In 2003 we decided to change the number of Council members
to 5, after some discussion among the members of the 2nd Council, the
JSF membership, and the Board of Directors. This change was never
reflected in the Bylaws because (1) the Bylaws have never fixed the
number of Council members and (2) at the time (August 2003) we had a lot
of move pressing issues to resolve, namely whether the JSF would survive
and whether the Jabber community would split apart over the Jabber vs.
XMPP issue. (Remember those fun days?) However, I'm all in favor of
making a change to the Bylaws so the the 5-person limit is explicit. But
it's too late to make that Bylaws change in the next meeting (Sept. 20).
We could make that change in a special meeting or in the October meeting
(i.e., when we elect new members, who will start applying October 1).
The current wording in the Bylaws makes it sound as if any person who
runs for Council and who receives 51% of the votes will be elected to
the Council, which I think is clearly wrong -- what if all 57 JSF
members ran for Council and they all received 51% of the vote? Would we
then have 57 Council members? No. Since 2003 the Council has been
limited to 5 members, which has been more efficient. No one objected to
this in 2003 and no one has objected to this since then, so I don't
quite understand where the objection is coming from now. But maybe I'm
missing something.

> If somebody made a mistake in his votes please contact stpeter to
> reset your votes and vote again.

That's right. And *no one* has contacted me yet to change their votes,
so I think this is a tempest in a teapot.

> If you are with Kevin and say the memberbot voting is biased please
> let us know. Then we should find another way and vote again. We have
> enough web programmers here (including myself) to hack a voting
> script in some minutes. Better to start over now before ist too late.

We've used memberbot for years. No one has ever raised an issue of bias
before. I see no reason to change the voting in the middle of the stream
here. Especially because this voting period has already been delayed (we
should have voted about a month ago, since the JSF was founded on August
20, 2001 and we attempt to hold the Board+Council elections each year
around that time, but couldn't this year because of delays with the Q3
membership voting). Also, it's not quite as simple as hacking up a
web-based voting script in a few minutes. What about security? How will
we handle authentication? We don't have a database of JSF members to
auth against, we don't have a way to auth against Jabber accounts on
various servers from www.jabber.org, etc. Working on that infrastructure
is a good idea, but it's a bit late to do that in time for the Sept. 20
meeting at this point!

Peter

-- 
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.shtml

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 7358 bytes
Desc: S/MIME Cryptographic Signature
Url : http://mail.jabber.org/pipermail/members/attachments/20060912/0baea23d/smime.bin


More information about the Members mailing list