[Debconf-team] Rethinking the way travel sponsorship works
Hi,
After today's meeting, we idled a bit in the IRC channel exchanging
some ideas. Tiago talked on what he perceived to be the main
problematic that other DebConf attendees complained about, and he got
some people to answer, me among them.
So, I'll basically paste the IRC log — excuse me for being lazy. I
know it is not the easiest format to read, understand and answer to,
but it is more natural and will get me to send this mail easier and
with more certainty ;-)
Oh — and I am not convinced of what I'm proposing, although it does
sound somewhat sensible... What I want is to get some discussion going
and to sketch how this year's decision will work. I'm editing a little
bit for format, and to filter out extraneous topics.
tiago_: so it's time to implement a bit of what has
been discussed during/after dc12
tiago_: bremner, agreed
bremner: I tried, it was too much work, I gave up.
tiago_: i see, and i couldn' help much :(
tiago_: I think a simple idea of calling volunteers out
of the team for rating would avoid issues we had in dc12
h01ger: tiago_, hi. can you explain that idea more,
please?!
tiago_: h01ger, we from team were kind of accused of not being clear
about money allocation for travel sponsorship last year, so
I've been thinking that the team should take care of
procedures, but rating should be done by (semi)random people,
in short.
gwolf: tiago_: interesting idea... Well, I would not say "random
people", but maybe yes do a wider call instead of rehashing
last years' Herb team
It's anyway hard to be clear being Herb (and having to work
privately, as personal issues are weighed).
When I've been in Herb, we are often accused (rightly) of
applying different criteria to different people... but it's
very hard to be "linear"
h01ger: tiago_, i'm happy gwolf picked up the conversation as i
believe its super important to have and b.) cause i want+need
to stay out of it
tiago_: gwolf, that's why we need strong and clear procedures, and let
people rate based on that, under our 'instructions' and
support
most important is having more people rating to reduce
individual 'mistakes' on rating
gwolf: tiago_: ...or spreading the statistical load on more people :)
lets go to the absolute opposite... Crowdsourcing - lets play
with the idea and see if it looks better
tiago_: but having a team with ~30 people discussing procedures wont
help at all
gwolf: Every DebConf13 registered person has 5 "votes"
(paid and sponsored, each has 5), and they allocate their votes
to 5 different people requesting sponsorship.
I still see a problem with this: People don't want their
requests to be public. But... It could work? :)
nattie: it could so easily turn into a popularity contest, though
gwolf: nattie: sure, I completely fear that.
that's why I said I'm proposing to play with this
tiago_: gwolf, yes, but then i'd limit the 'vote' power for DC attendees.
gwolf: Maybe we could get people to vote on the anonymous pairs of
(reason,amount)?
hiding who submitted each request
tiago_: i'd like to have something similar, but involving non-DC attendees
gwolf: some requests will be obvious (i.e. if I say "I have to do
DebConf orga and keyring stuff", it's obvious to any involved
person it's me)
tiago_: opening the vote to DDs?
tiago_: I don't know if popularity is the only merit for getting
sponsorship. There'll be valuable non-popular contributors who
have good reasons to ask,
gwolf, anyway, people would vote based on the requester
comments/reasons?
gwolf: of course - that's why I suggest anonymizing the requests
on the reasons and (possibly) on the requested amount
although that would need to disclose origin
tiago_: it may work...
gwolf: or at least it could be weighed. Say, 0.6×crowdsource + 0.4×herb
tiago_: i like the idea
Tiago had to leave at this point, and we decided the conversation
should be moved over to the list.
So, what would you feel about such a scheme?
Reply to: