[OSGeo-Conf] Conf for 2008

Tim Bowden tim.bowden at westnet.com.au
Tue Mar 27 01:48:17 EDT 2007


Well, I've taken Paul's advice and I'm here to agitate ;-)

After looking over the list archives and wiki material, I've a few
thoughts to offer.  Before I go on though, a disclaimer:  My open source
conf experience is limited to being on the periphery or sidelines of a
number of lca's- linux.conf.au and my observations are coloured
accordingly (and yes, that is how you spell coloured!).

Background: lca is one of the worlds best technical linux/open source
conferences that rotates around aust each year.  It's entirely volunteer
run, has a strong developer focus and has been growing at an amazing
rate every year since 1999 when it was first held.  This year it capped
numbers at 800 with a budget of abt AU$1m.  That's around US$800,000.
The speaker travel budget this year was equivalent to the entire budget
for the first conf.  Don't get me wrong, I'm not suggesting we adopt the
model Linux Aust is using; There are some significant differences
between the two communities, but I do think there are some lessons
learnt at lca that we can take advantage of.

Ghosts of lca's past:  Being a rotating conf with new people running
things every year, the wheel was being re-invented too many times.  LA
put in place a process to pass on learnt knowledge.  It has proved very
valuable.  Key organisers from the previous year are flown to the next
location for a meeting with the next organisers and LA committee to pass
on their experiences, help solve problems and raise issues that may not
have been considered.  Given that in this instance all the travel is
domestic (apart from when Linux Australia held the conf in NZ last year)
the costs are reasonable. They would be much greater for OSGeo to do the
same but perhaps OSGeo could find some way of formalising the transfer
of knowledge from one team to the next.

Equipment:  LA is starting to build up a cache of equipment that can be
rotated around conferences.  AV recording sets, wireless access points
etc.  In the long run, it saves money and is one less problem for local
teams to deal with.  Processes need to be put in place to keep track of
it all and so on, but it looks like it is working so far.  It's taken a
few years to get to this point though.

Cost:  LA strives to keep the conf as cost friendly as possible.  All
work is done by volunteers (though this does result in a high burnout
rate- It takes some years of recovery before anyone is willing to have
another go at it).  The venue is always a local uni that offers
sponsorship in the form of cheap facilities and has a keen community of
CS students that get involved.  This year cost was $99 for students,
$300 for hobbyists and $690 for professional delegates (if your boss was
paying or you were 'on the FOSS gravy train', you were considered a
professional delegate).  The rest of the costs were covered by
sponsorship.

Now that the conf is well established, this is not too hard to do.
Sponsors line up for the chance; IBM, HP, Google etc.  As one local
luminary said in the early days, The sponsors need lca more than lca
needs the sponsors.  Maybe that's got something to do with the calibre
of the delegates.  Look at it this way, if the OSGeo conf is a developer
rich environment, with the cream of the open source GeoFOSS community
there, do you really thing companies like google are not interested in
being associated?  Do you really think they are uninterested in paying
to be involved in an event attracting some of the best open source
geospatial developers in the world?  Have we approached google about
sponsorship?

Potential sponsors know that many of the hardware and software system
purchasing decisions are made or strongly influenced by the tech
community at the conf.  It's also a great way to pick up key staff.  In
the case of LA and linux, it's a cool tech to be associated with.  To
what extent does OSGeo and GeoFOSS in general have this type of buzz
about it?  Maybe not quite as much as linux, but I don't think it should
be discounted.  This 'reputation buzz' is growing and is a valuable
asset for the long term.  Lets not sell ourselves too cheaply to
sponsors.  I suspect if sponsors always got their way, everything would
end up a trade show and we would all be the worse off for it.

As far as the balance between suits/developers/hobbyists/students goes,
I think we are best served by keeping it as 'community' as possible; To
me that means having a focus that is developer friendly.  If the
developers aren't interested in coming, then either the rest will fade
away or it will become just another trade show.  If the developers are
there, the rest will want to tag along anyway.  Keep the event
accessible to students; in a few years they will be the professional
delegates we want actively involved.  To me this is an opportunity to
serve /our/ needs, rather than a sales pitch to the rest of the world
(though that will be an incidental benefit).

Lead time:  Experience with lca has shown that it takes a local user
group about two years to get to the point where they can make a
successful bid to hold the conf.  In part this might be because every
conf gets better, but it takes a lot of volunteer manpower and learning
be in a position to make a go of it.  The lead time once the successful
bidding team has been chosen has been extended to 18 months from 12
months.  There is just too much to do for a conf of that size to be
organised by volunteers in 12 months.  LA has looked at the option of
using a professional conf organiser, but has rejected it several times
because it would drive up costs and change the character of the conf
away from being a grassroots event.  I don't know if that's so, but it's
the decision that has been made several times.

Website:  Could conf.osgeo.org be a permanent pointer to the current or
immediate past conf?  It provides a permanent point of reference for the
conf.

We don't have the same manpower available as the linux community and our
needs are not always the same; we're a lot smaller in numbers so we
certainly can't do things the same way.  If we rely on volunteer efforts
only, we may be limiting ourselves too much, but I think we should still
be trying to maximise that avenue.  The more we can contain costs, the
more community involvement we will have.  At the end of the day, if we
loose community, we loose everything, because community is the lifeblood
of open source.

FWIW, that's my take on things.  It's probably different to what many
others are thinking, but that's what makes an interesting community.
Now who is interested in hosting foss4g2008?  ;-)

Regards,
Tim Bowden

PS, I just got this feedback from Michael Davies of LA, who has been
involved in the LA ghosts program.

On 27/03/07, Tim Bowden <tim.bowden at westnet.com.au> wrote:
> Hi,
>
> I'm looking at giving some input to the conf committee of the Open
> Source Geospatial Foundation, and I'd like to get some feedback on the
> lca ghosts program.  Specifically, how beneficial has it been and what
> sort of costs (ballpark) has it involved?

Hi Tim,

The "linux.conf.au Ghosts of Conference Past" program has been in
place for a while now, and it serves a number of purposes:

a) provides a debrief opportunity for last conference's team to tell
LA and next year's conference organisers things they need to be aware
of, do both a physical and virtual handover, including sponsorship
arrangements etc etc etc;

b) provides an opportunity for the next conference to bounce ideas off
Linux Australia and previous conference organisers, to validate their
thinking;

c) provides continuity for LCA by allowing input into the planning of
the next conference by those who have been there and have sweated
blood before (to maintain the flavour of the conference);

d) provides an avenue for Linux Australia to audit where the next
conference's planning is at, and what intervention (if any) it needs
to make to ensure the next conference rocks (especially financially
since LA "underwrites" the conference);

e) allows support structures to be put in place to support the
organising committee of the next conference.

As someone who has benefited from "ghosts" and has also been involved
in supporting LCA in this capacity over the past 4 years, I think it's
very beneficial to a roaming conference like LCA with a different
organising committee every year.

Cost wise, we're looking at domestic airfares and accommodation for
approx 6 people over a weekend.  I think it's a small investment
considering the size of the budget for LCA these days.

Hope this helps,
-- 
Michael Davies           "Do what you think is interesting, do somthing
that
michael at the-davies.net    you think is fun and worthwhile, because
otherwise
http://michaeldavies.org  you won't do it well anyway." -- Brian
Kernighan




More information about the Conference_dev mailing list