[Linux-aus] About a User Conference (was...)]

[Tim Bowden]

Time for me to come out of lurk mode, as this is an issue that I would
like to see dealt with.


On Tue, 2005-02-22 at 13:24 +1100, Andrew Cowie wrote:
> On Mon, 2005-21-02 at 21:37 +0800, Bret Busby wrote:

<snip>

> > There are a number of possible audiences for Linux conferences, and 
> > catering for all of these, could probably be done at the same 
> > conference, if it was wanted.
> 

There are different linux audiences, and they are large enough and
different enough to warrant keeping any non linux developer conf
separate from lca.

> No. The trend has been (across many organizing teams) to hold the number
> of attendees < 600 people in order to maintain the intimate feel and
> interaction that are hallmarks of the conference.
> 

I think this would be good for any linux conf.  You only need to have
gone to a few of the PC shows in the eastern states in the mid to late
'80's to see what happens to 'professional shows' when they become
mainstream and large.  It turns into nothing more than a glorified
retail experience with show goers looking for bargains.  That is not
what we want to end up with.

> With that kind of a cap (not to mention turning people away as that cap
> is hit) and a absolutely jammed program (of the sort AJ described),
> there isn't scope to add other constituencies to LCA.
> 
> Which is why we're having a discussion about the possible merits of a
> user conference.
> 
> The trouble is, conference economics tend not to cooperate. It goes
> something like this:
> 
> You want to keep the cost down so that users (especially newbies and/or
> non-Linux users-but-interested) can and will come. [It could even free,
> although THAT's a price you can never raise, and is even harder to
> provision for; see below]. The trouble is covering venue costs. LCA is
> inexpensive because it is small, collegial, and typically run quietly on
> a university campus. A user conference could very well end up being
> large and requiring an equally large commercial venue to host it in.
> Those sorts of things are never inexpensive. Neither are logistics.

Do we really need to cater to newbies?  I know they are /very/
important, but the type of user conf that might work better (at least
initially), is one that is aimed at 'professional users', ie, dba's,
small time(?) web/sys admins, serious desktop users (oo.o, gimp...) and
so on, who are not kernel developers and not well catered for at lca.

That type of user profile would not have any problems with a uni type
venue and would be more accommodating to a numbers cap. Surely there
would be vendors who would support such an audience, even if not with
the same $$ as with lca- Novel? HP? Oracle? Does anyone have any
concrete idea of how they would view such an audience?

Just out of interest, can anyone give an of-the-cuff ballpark figure of
what lca2004 cost, with an approximate breakdown of where the money
went?

> 
> So, you probably want sponsors to defray the cost instead of attendees
> footing the entire bill. The trouble is, that most of the people who
> might be interested in sponsoring such a thing (I'm judging by other
> international events' experience here) tend to be vendors trying to sell
> products and services to businesses.

Is it any different for lca?  IBM, Sun, CA etc want to sell you things.
They are building mindshare by sponsoring the conf.  If they see a
professional user audience rather than newbies, surely it would be
considered a target rich environment, even if slightly less 'rich[1]'
than that at lca?
> 
> And there's the rub; a trade show, let alone a conference which appeals
> to a business audience [and therefore would be the kind of audience that
> vendors would be willing to pay to try and reach], is potentially a very
> different atmosphere to a "user" conference (whatever "your" idea of
> that happens to be).

Anything heading towards a trade show type event should be buried 6'
under unless a commercial organiser wants to do it themselves.  IMHO it
should not be lug's or la that drive such a thing.

> 
> An interesting instance of how this can work out is LinuxTag, which is
> half conference (sorta along the lines of LCA, but only sorta) and half
> trade show (sorta along the lines of LinuxWorld or InternetWorld or PC
> Expo). It's mostly free to attendees (one of the days' tracks cost $$)
> but it certainly didn't have any new users walking around - indeed, it
> really wasn't a "users conference"; their tag line ("where .com
> meets .org") is a good one for what they are.
> 
> I'm not so big on LinuxWorld or LinuxTag's style that I'd recommend we
> blindly do either, but that does somewhat bring us back to where we
> started: how do we compose a user conference that attracts the audience
> we're interested in AND the sponsors we're interested in.

I have no experience of LinuxWorld or LinuxTag or what sort of audience
they have so I may well be on the wrong track here, but an audience of
professional users, rather than newbies has to be of interest to
potential sponsors in the same way lca is.  Aim the conf at those who
already know how to install their distro of choice, and have an interest
in using linux in a professional capacity, without targeting kernel
hackers.  Perhaps tracks like prof. dtp, desktop linux for the office,
web services, database tuning & web interfaces (MySQL v PostgreSQL etc)
and so on.

> 
> One final thought, seeing as how we're right off the deep end:
> *personally*, I really like the idea that some of the OSIA people have
> been bandying about of a road-show: a conference that we'd pack up and
> (in the space of a few weeks) do in 4-6 major cities. The aspect which
> was interesting was a notion that 50% of the exhibitors would be local
> SMEs which would of course be different at each location, whereas the
> other 50% would be a backbone of larger national organizations with an
> interest in reaching a nationwide audience.
> 
> I like the idea; it would be awesome for Linux in Australia were we to
> do such a thing successfully. The organizational demands, logistics,
> cost exposure, and risk exposure of such an idea is staggering, and I'm
> not sure I could advocate it unless it was very professionally organized
> (ie not a band of merry volunteers), and that's probably pushing our
> mandate and ability as an organization.

Sounds more than any lug's would be willing to take on. LA's strength in
running lca is using the local lug.  If you want LA to take on something
like that, It would take more resources than LA is used to having.  As
you say, It would be a huge ask.

> 
> [Reigning the flights of fancy back in, I will mention that Mikal is
> considering working on something at the much-smaller end of the scale.
> I'll let him discuss that if and when he sees fit]
> 
> AfC
> Sydney

Tim Bowden

[1] Rich as in good prospective targets for whatever service they are
selling.