Hi, On Montag, 16. März 2015, Wouter Verhelst wrote: > Note to self: if you're going to suggest a meeting, better f*cking > actually be there, yourself. Grmbl. #meetbot pingall is also really useful. > Having said that, though, I do believe the urgency is gone now that > -team has come up with a budget; we "just" need to come up with a way to > do our stuff within that budget (which, I think, is perfectly possible). > So rather than wasting everyone's time yet again, I suggest we do things > via mailinglist instead. > > What needs to happen: > - We need to get four cameras, four tripods, six laptops, and a whole > bunch of audio gear and cables from Paris to Heidelberg and back > again. Richard seems to think we can do that at €500, which I think is > too low; I think a guesstimate of 2k is more realistic, which Richard > thinks is ridiculously high. I dont think we should aim to pay money for that, rather find nice french people to drive it for us. > I can't come up with a better answer to the question at hand from > where I am (typing this in a restaurant on my way home), but I'll take > this as an action point for later this week: figure out what a > reasonable guesstimate would be. If someone has an answer already, be > my guest ;-) look at videoteam expenses from last years? switzerland should be pretty comparative. > - We need to buy some consumables: > - batteries for the wireless mics. > - duct tape (or is this done by network team?) > - media to put the recordings on (disks)? Not sure if we need this. > How is this usually done? > - Anything else? > - We need to organize a sprint. Ideally, people from non-debconf video > teams would be present at this sprint, too. We should talk to those > people, see if they're interested in joining, see how much time > they're willing to spend, and come up with a date. It should be: > - Long enough before debconf that we can use whatever we figure out at > this sprint at debconf itself > - long enough in the future that it's still possible to organize it > properly. end of may? > I think one month before debconf is the absolute minimum to pass on > the first restriction, and that 6 weeks to two months is probably what > it would take for the second one; that leaves (late) may, all of june, > and early july for the sprint. > Open questions: > - how long should such a sprint be? We could make it a weekend one, so > that people don't need to take vacation, which is probably a good > idea. Alternatively, we could make it a week sprint, which would > give us more time, but would require people to take time off from > work. Maybe do something like wednesday through sunday, and people > who can only join for the weekend arrive friday night? i'd prefer thursday to sunday. > - Where are we going to do this? I don't personally have any space > that I could use for this kind of stuff. Anyone have an idea? linuxhotel.de in essen would be an option i'm sure, if they still have free weekends. else maybe some place in heidelberg? or a hackspace in hamburg or berlin? or brussels? > - We need a plan for recording stuff at conference itself. I think for > this year, going with a dvswitch-compiled-on-wheezy is an acceptable > plan that will work (the issues with dvswitch are that the API changed > in recent libav versions, which happened after the release of wheezy; > when compiled against the libav libraries from wheezy, even if the > result runs on jessie, everything works fine). However, I would feel > more comfortable with switching to some other option. At the same > time, I don't feel comfortable with such a switch until after the > sprint. agreed. also i still think dvswicth on wheezy is totally fine today. > - We need a plan for reviewing. In past debconfs (that I was present > at), we used a penta extension, which seemed to work fine. However, > since debconf has moved away from penta now, that is no longer an > option. I wasn't at debconf 2014, so I can't evaluate how well last > year's solution (veyepar) worked; but I can note that for FOSDEM 2014, > it wasn't a complete success. Then again, FOSDEM and DebConf are two > completely different beasts; what is a bad idea for one conference > isn't necessarily so for the other. > Can someone who actually was at DebConf14 comment on how well veyepar > worked for DebConf? i'll pass here. (just note that all the video extension needs is a penta install (which we still have) with updated datam, wchih we could extract and feed into penta... the ccc also has a video reviewing system. cheers, Holger
Description: This is a digitally signed message part.