Hello Everyone,

I'm sorry I've been quiet lately: we are under a lot of pressure at
work, the good news is that we just heard that we got the contract for
the Archive Data Management system at Silverstone (the car race track
that is).

EuroPython 2009 was a lot of work, and despite the experience we had
with two years of PyCon UK beforehand, we learnt a lot. I've been
thinking a lot about future EPs, and here are some thoughts:

Sponsorship
==========

We put a lot of work into getting sponsorship for 2009, but were let
down by several of the potential sponsors. Some of what we were
promised verbally never materialised, and some of what was confirmed
has still not been paid. Some sponsors, even Europeans, sponsored
PyCon US a lot more than EuroPython. Some sponsors get much much more
out of their sponsorship than their contribution deserves.

I do not think that the time we have spent getting sponsorship has
been worth the effort (largely time paid for by OpenEnd and
Clockwork).

In future I do not think that we should rely on getting sponsorship,
nor should we spend valuable time chasing it, a simple announcement
and basic brochure is quite enough.

We should not rely on sponsorship when budgetting, although if
sponsors do come forward that would be good. For example we could
budget for very basic food and increase the food budget nearer the
event if sponsors come forward.

Sponsorship must be paid before the conference (say two weeks before).
We do not have time to chase payment afterwards.

Numbers of Delegates
==================

In 2009 we had around 440 delegates (I think 446 was the final
number), which was more than we expected. The absolute maximum we can
handle in Birmingham (in facilities we can afford to hire) is 550. If
we go for more we would need to hire a much more expensive venue. In
our case this would be the International Conference Centre, but the
costs would mean that the delegate fees would be a lot higher than
many people could afford for a community conference.

When they moved from Washington DC, PyCon US solved their venue
problem by using hotel conference centres, which meant that they had
to guarantee hotel rooms. As long as the conference expanded this was
OK, but this year the consequences were that PyCon lost, I think,
around $200000, and are liable for $300000 guarantees for 2010.

* We cannot afford to risk this.
* We must not guarantee hotel rooms.
* We should not be afraid to limit the number of delegates to that
which the venue can handle.

Whilst the number of delegates that we might expect at future
EuroPythons may be 800-900, this number is beyond what (I think) can
be handled in a community conference: it is too much work for
volunteers to handle comfortably and rather than go for paid
organisers we should limit numbers and maintain a 'community
atmosphere'. An upper limit of 500-600 may be sensible. Given that
Europe seems to have more 'regional' conferences than the US already,
this may be OK.

Experience
=========

If the PyCon UK Society had not run the 2007 and 2008 PyCon UKs, there
is no way that we could have handled EuroPython 2009. I suggest that
in future any group bidding for EP must have run significant
successful conferences beforehand.

A side effect of this might be that, if we have several groups
bidding, we could move away from holding EP in the same venue for two
years running, as they will not have to go through the first year
learning experience to the same degree, With groups holding EP for one
year only, more would have a chance.

Help
====

We have already said that a condition of a group holding a EuroPython
is that they must have helped run the previous EP. I suspect that this
should be the previous two EPs, as whoever gets the 2011 conference
will not have helped with a previous conference when they are chosen.

It was quite noticeable that it was the same people doing most of the
work for the past several EuroPythons, and not always the locals
either. Several of the UK contingent did help with EP 2008, but this
was unusual.  There was no new blood from outside the UK helping with
EP2009.

Sprints
======

At EP 2009 we did not have sufficient advance notice of what sprints
would be held, nor the numbers of participants. The sprints have a
considerable impact on the budget, as the sprinters do not pay for
them (nor should they), and the costs have to come out of the main
conference budget.

Based on the 2009 experience, for EP 2010 we will not be providing
dedicated sprint facilities before the conference, but will do
afterwards.

Number of Talks
=============

We accepted too many talks for EP 2009, at least for the facilities we
booked. As we accepted more talks we booked more rooms, but these were
not big enough, and we had too many streams.

For 2010 we plan to have an extra day for the conference talks, that
is four days instead of three, which means we can have fewer streams
and bigger rooms.


That's enough from me for now!

Best wishes,

John
--
_______________________________________________
Europython-improve mailing list
[email protected]
http://mail.python.org/mailman/listinfo/europython-improve

Reply via email to