Hey Marga,
thank you for your response. Before I try to address the (valid)
points you raise, I'd just like to clarify two things:
- I probably didn't communicate this clearly enough, but the
numbers in current budget.ods are nothing other than numbers
I used to create the model. They are not even trying to be
a proposal, let alone have any relation to reality. They *all*
need to be tweaked, which is why I reached out to dc-team.
- The way I approached this budgeting question is derived from the
way I've budgeted conferences in the past, as well as other
projects. For me a budget is good when it has two qualities:
realistic, conservative assumptions, and an implicit
understanding that it won't be final until the event is over.
That said, I am keen to learn of other approaches and if my
approach is not the way dc-team wants to do budgeting, then I am
open to any suggestions on what to change. I just don't want to
lose so much time with this, so let's agree to agree quickly on
a way forward, okay?
> The main issue I have with this is that you are already including
> decisions that have not been taken, like exactly what format
> DebCamp will have:
First, the proportion of DebCamp in the costs is low (6 vs. 8 days,
approx ¼ of the attendants…). But we can and must change these
numbers. If there'll be a decision such as "from Tuesday onwards",
then it'll be possible to reflect this in the numbers (create
a scenario for it).
> Also the amount of sponsored people each-night is very closely
> tied to the amount of people we are willing to sponsor which
> should not be "guess work" but rather a set number that we (the
> whole team, not DC15 nor budget team) decide upon.
How do you suggest we arrive at this number?
Unless you know the number of attendees each night, it would be
real "guess work" if you just picked any number and went with it.
So my approach was to estimate the number of attendants we get each
night, which are the important data because I feel like we need
those to even start thinking about what ratio we want to sponsor.
The number of sponsored attendants people filled in are only one
vector of input we can use in making the decision.
Anyway, whatever we come up with can be reflected in the numbers. If
we decide that there ought to be 150 sponsored attendees, then we
just create a scenario that reflects this.
I repeat: The current set of numbers in there is not intended to be
any proposal. I am reaching out to "the whole team, not DC15 nor
budget team" *because* we want and need your input on this. Since
the Delphi technique didn't really seem to work out, I called
a meeting.
Meanwhile, you noted on IRC that the expenses I list (they are all
just examples, this is not meant to be a proposal) are not
prioritised. Your idea here seems to be — please correct me if I am
wrong — that when we don't meet the income we needed, we know
exactly which expenses to slash. So while conference rooms would be
a requirement, a conference dinner assumes varying degrees of
optionality.
Obviously, it is correct that some expenses can more easily be
slashed when the budget would otherwise not be met. However, I don't
think we can or should do this at this point in time. Also, it's not
always a black-white process. If we don't meet the budget, maybe we
can meet it by simplifying the food offerings at the conference
dinner and skipping the karaoke rental. My point being: let's cross
this bridge when we get to it. If the budget is realistic,
conservative, and up-to-date by this point, we can freely experiment
and evaluate all the options we have to make ends meet.
Budgeting is an iterative process IMHO. We start with assumptions
and then work on that basis until we decide to amend the assumptions
and give ourselves a new basis to work on. This should obviously not
happen too often, but it's really useful to do so when new, salient
information becomes available. Nothing is set in stone.
Finally, you proposed on IRC that we should separate travel
sponsorship into "mandatory" and "optional" segments, with the first
being backed by Debian in case we don't raise enough money. I think
this is a good suggestion and I'll happily try to work this into the
model.
What else is there, though? Debian can back a certain volume of
travel sponsorship, and Debian will probably also back a certain
volume of diversity outreach programme. There might be more such
pairs, especially when sponsors start buying specific perks. Then
it'll become necessary to earmark certain income for specific
expenses. I'll think about how to do this.
There is also more to-do in the current budget model, e.g.
parametrise other expenses to allow for easy evaluation of options
when we receive less (or more) money than expected.
I hope this addresses your points. I am available now on IRC if you
want to discuss. We can also, obviously, pick this up in tomorrow's
meeting.
Busserl,
--
.''`. martin f. krafft <madduck@debconf.org> @martinkrafft
: :' : DebConf orga team
`. `'`
`- DebConf15: Heidelberg, Germany: http://debconf15.debconf.org
DebConf16 in your country? https://wiki.debconf.org/wiki/DebConf16
Attachment:
digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)