Am 06.02.2016 um 02:50 schrieb martin f krafft:
I think closing that early is unrealistic as we won't have the speaker lineup, workshops/BoFs by then. So we'd cut ourselves short of Debianites that are interested mostly in the "conference" piece or people attracted to a specific workshop-topic of the DebConf.also sprach Daniel Lange <dl.ml1@usrlocal.de> [2016-02-06 03:50 +1100]:1.1) breakdown of these: # people sleeping at UCT {DebCamp | Debconf} -> needs to be detailed later into individual daysWe'll be telling UCT exact numbers for each night, right? So knowing how many people plan to come isn't going to be helpful, unless we are willing to guess the individual nights from a bell curve overlay or so. I think that would be the right thing to do unless we plan to *close* registrations by early April to meet the 45 day deadline,
We need the peak occupancy for DebCamp and DebConf because this is what we buy (and pre-pay) now. This will be a f(pre-registered people). The bell curve is an attractive cost-containment measure we need to calculate in April. Because then we can still cancel beds we don't need on arrival and departure days. But as marga and you know who did the capacity estimations and allocations for DC15, it is a hell of work, needs somewhat sane data and there will still always be an error margin.
How likely is it that the venue will fill up? Given the current situation, I am inclined to reserve a smaller number and hope to book additional beds as we require them.Yes, that was the plan. The team meeting on 13.01.16 decided to reduce the room allocation for DebConf 300->250 exactly for this reason. And with 250 we still need allocations for both of Smuts and Fuller Halls. Which increases the probability we can up the allocation still once we have better numbers. But this way we have to pre-pay less and have less of an issue with the (potential) refund (=cash in ZAR).