For disclosure: I've been a member of this year herb team even though in
the end I didn't manage to rank anyone nor to attend the final decision
meeting (due to a lack of time, shame on me!). To cover up, I've managed
to find a substitute who kindly agreed to take my place. I've also
asked for sponsoring of about 80% of my travel expenses --- explaining
why I couldn't afford to pay them myself, as well as what I do in
Debian. I've been granted sponsoring for the amount I requested.
Furthermore, before asking here, Richard has asked me if Debian (via
SPI) could act as an intermediary for reimbursing Clint (receiving
donations to that end and then approving his reimbursement). I've
replied no, without knowing of the potential tax problem, on the basis
that I don't feel good with Debian doing that for one person alone.
It might happen that sponsoring decisions get something wrong; in fact I
believe it will *always* happen, no matter the process we choose --- in
that respect I completely agree with Phil's mail earlier on in this
thread. When that happens one can say "that's life" or try to
compensate. But in compensating I think we should make an effort to find
all the people affected by a specific problem [1] and deal with all of
them. Up to now, I've seen no evidence of that an effort to identify
others in the same situation of Clint has been made.
Therefore I disagree in using official Debian structures to fix the
problem only for him. Of course that is not meant to block other,
non-Debian backed ways, of solving the problem. To be honest, I don't
see where the alleged "sketchy-ness" of doing the special sponsoring not
relying on Debian is: just have one person collect the money and
transfer the money to Clint. With my friends, I routinely do money
transfers even for restaurant bills when it happens that someone
anticipates for others at the restaurant and it's not particularly
cumbersome.
OTOH, if the proposal of doing it more officially also serves the
purpose of having a public discussion on how to improve the process over
time, then by all means let's do that.
[1] in this case the problem can be described as "respected and active
people in the Debian Project who, according to the sponsoring team,
didn't enter enough info and hence get rated down"
On Thu, Jul 21, 2011 at 08:49:17AM +0200, Joerg Jaspert wrote:
> I will be so happy to see something. This years team and process (done
> like past ones) came up when there was noone else doing it. If we
> have something nice and shiny next year already prepared, all the
> better.
Indeed. An obvious place for potential improvement I could see is in
how the members of the herb team get selected. I understand thus far
they have been chosen by "snowballing", i.e. who took the work on their
shoulders choose an initial set of people and ask them to propose other
people. There are flaws in this approach of course; for one thing, while
I don't personally see any malice into that, it can easily be accuse of
not being a transparent process.
What I've failed to see thus far in this thread are alternative
proposals. How could we choose the members of the herb team in a
transparent manner? Would a call for volunteers on a public mailing list
be enough? Has it been tried in the past? With which results?
> And then we had, in the past, the rule to not rate yourself. Which in
> the end turned out to punish people for doing the work (no rate, lower
> score, WAY down in the sponsorship, and that just because you wanted to
> help DebConf). Which changed the policy to "Rate yourself. The rest of
> the team WILL rate you down if your request is insane". Which did
> happen, this year too.
I've heard this argument before (I believe it was last year in New York,
talking with members of the DebConf10 herb time). It sounded convincing
back then, but a bit less so now. In particular: if instead of taking
the sum of scores we take the average a large deal of the problem should
go away. You will still have the problem that the average for herb team
members is taken on a smaller sample than for non-herb team members, but
if the team is large enough (which I believe was a very good feature of
this year team) the difference should be meaningless. Has that option
been considered in the past?
Cheers.
--
Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Quando anche i santi ti voltano le spalle, | . |. I've fans everywhere
ti resta John Fante -- V. Capossela .......| ..: |.......... -- C. Adams
Attachment:
signature.asc
Description: Digital signature