[Debconf-discuss] Travel sponsorship BoF: minutes
Hi,
We had a well-attended (and well-behaved :-) BoF chaired by Zack (I'm
cc: directly, but I remembered you saying you'll joing debconf-team,
so let us know and we can drop the explicit CC:).
Thanks to everybody that participated and contributed ideas and solutions!
I scribed the following notes during the meeting, while projecting
them on the big screen (my apologies for not sending them earlier, I
was moving). We couldn't use gobby as there were no internet in the
room.
The question is of course how to follow on from here. Ideas? (maybe
discussing this in -project?)
Note: segments with (...) were not transcribed due to some DrDub's
lapse-of-memories during the BoF.
If you remember, please add them. Also, as with any transcription,
this is imperfect so feel free to reply with edits, clarifications,
etc.
----------------------
(drdub) Following a discussion with Ana, the project might save quite
a bit of money if buys the tickets for key contributors very early on
the year. Following this idea, I propose we split the people we fund
into two categories: "Debian award" attendees, which are selected by
the project in, say, November/December and the project decides how
much money they'll need to travel to the location. The process for
selection is similar to the current herb process. The recipient can
opt for not receiving the award and it will then go to the next person
in the list. The project can use $10,000 to fund this, for example (or
separate DC(N-1) money specifically for that, pending the result of
Friday's BoF). If the money assigned to the person is considered
insufficient, the Debian contributor can then ask for extra money in
the second round, which is similar to how we have conducted it this
year (i.e., contributions / money requested? / need? weighted).
* people should be nominated (even self nominated)
(drdub) I feel this year we discounted too much for people the herb
team felt were asking too much money. If we go for a similar process,
people shouldn't be discounted, their amounts should be adjusted.
* ana: done other years
(drdub) I think we can easily fundraise for specific people if they
are willing to go on the record in a suitable Web page. We discussed
it before but people felt it was too close to the conference for it to
make sense. So we can talk about it for next year...
* won't work, too dunc-tanc-ish.
(moray) try to consider the location for distributing the money.
* Sometimes it can be that getting money in one year reduces your
changes in next years?
* allocation of money based on maximazing the number of people in the
conference?
(ana) how did we do in previous years?
- schmultc: last two years we have a ranking interface and we ranked
people based in a number of issues: speaker, known Debian contributor,
relative expense of their ticket. (-2 to 2 rankings.)
- micah: agreed, prices were considered too much for the metrics.
- ana: the metric is OK, but we're using it wrong. Initially the
idea was to find who gets sponsorship and who doesn't, and then
everybody should be considered (every contributor).
- faw: problems, the herb team is never the same and the focus keep
changing. The dichotomy of "prize" vs. "need". Some people can put
more money but try to game the system.
(from here on, the discussion wasn't focused, here's a list of
comments, person by person)
- zack: we don't need talks 6mo. in advance if sponsorship is the
only reason for it (moray: it was not, but it is now). Zack: I did
the DebConf newbies (...).
- krose: don't get too build up about the conference...the most important
thing is to have a good way to fix mistakes rather than avoiding them.
Also, all tricks to fundraise should be always welcomed.
- phil: it is important to make the metrics very clear
at the time of the request.
- moray: people don't interpret "need" in the same way. We should
define the expectations better.
- ralf: the 6mo. submission of papers help to get attendees payed by
their employers / PhD-supervisors. (moray: that's more the
original reason.)
- clint: do not consider talk proposals as part of the criteria and
problem solved.
- gwolf: (...)
- ana: free food and accommodation should be the 'reward' for
working on Debian.
- molly: consider the food + accommodation + ticket as a full
"travel grant" continuum.
- drdub: "why I'm asking for money?" Ans: "because it is offered" as
valid as "I choose to be poor".
- gaudenz: (...)
- micah: there're a lot of people who submit talks because they
think they will have more chances of getting the travel
sponsorship. (drdub: seriously!? audience: yeah!). Is that good or
bad? It can be seen both ways. Jonah (a local volunteer) had said
a couple of times that he is so amazed to see a conference where
the organization committee takes so much care of their attendees
(signs, food, accommodations). That's part of our money woes. By
offering money we are changing the dynamics.
- ana: we're making it too easy to ask for money. (Just tick it in
penta and that's it.) Many people mark that up to see what
happens. (audience: that's very true). In other conferences it is
needed to fill in much more data.
- gwolf: we're not communicating reality. People expect us to have
money because we're offering it. If we say it from the very
beginning "you should request for travel sponsorship if you really
need it". He expects people to act on good faith in that
situation. We should encourage people not to ask. (faw: the
mandatory written paper was implemented to make sure people had a
paper to send.)
- jeremyb: (to gwolf) people submit just to get travel sponsorship?
- faidon: we shifted from how to be fair to sponsor less. Debian used
to have a lot of money.
- drdub: we no longer have so much money. And we are focusing too
much on talks.
- faw: we need people to stay put for years in the core team for
this to work. (...)
- moray: in DebConf Oslo the sponsored accommodation were two empty
rooms to sleep in the floor. And next year there were
airline-style foods. But over the years the things have
improved so much.
- gaudenz: "deserve" vs. "need", can be solved with very clearly
written guidelines.
- blars: the travel sponsorship gets approved very late. Do I get
the airline ticket in my credit card and hope it will get approved
or continue to wait and the ticket gets so much more
expensive. And then there's the problem of "what will the
conversion rate will be".
- zack: there's much less money in Debian than it used to have. And
meetings are great ways to spend the money, but it shouldn't all
go to DebConf. And DebConf should be an amortized zero budget
- ana: people submitting talks to get food and accommodation. Last
year ana was in the team and it worked very well on her
opinion. This year they said 'need to be more fair and bring new
people' but then they went with the same behavior as previous
years.
- schmultc: Debian has less money than it does in the past, because
we have less hardware sponsorship.
- luciano: same problem as blars with the ticket but he decided to
pay for it earlier. Maybe the Debian cash can be used to *finance*
the tickets.
- molly: "deserve" vs. "need", split into two categories.
- faidon: we keep giving the impression to people that we have
money.
- zack: Debian can guarantee an amount of money for early year
tickets. Filter $tourist and then pay for tickets.
- moray: we should be asking people more information about what they
did in DebConf with that sponsorship, and public. That can help to
decide to give money again next year and to communicate success
stories to DebConf sponsors.
- micah: DebCamp work-plan... it was a farce to scare people away. It
was terrible for various reasons (e.g., he would have come to
DebCamp for previous years). The problem is that lots of people
wanted to tourist^Wattend DebCamp for no particular reasons with
sponsored food and accommodations. That created quite a bit of
financial strain.
- gwolf: we need to get feedback on the perceived impact of the
funding only a percentage of their costs. He doesn't think there
was a big impact of tourists on DebCamp.
Reply to: