[eclipse.org-eclipsecon-program-committee] Re: EclipseCon 2010 Conferenc

[Oisin Hurley <oisin.hurley@xxxxxxxxx>]

Notes from this call.

Attendance:
Oisin/Dave/David/Donald/Doug S/Chris/Ed/Ketan/Wayne

Resolved - structure of conference to include talk on
Monday (thanks, Don).

Panels -
 * Dave and Bjorn to put heads together on Software Craftsmanship
 * Ketan and Dave getting things together on Testing
 * Oisin got 2x volunteers for OSS SOA Panel
 * Ed cogitating about Modeling related Panel
 * Chris hooking in Tom to help with Runtime Panel
 * Doug S got some names in mind for Enterprise Eclipse Panel
 * Doug S also got some ideas re Mobile Panel - collusion with Doug G
   imminent
 * Wayne and David 'gathering community' on Processes Panel

Any more panel ideas, push them out onto the list - the 'making
money from OSS' Panel idea hasn't gone away, BTW, hopefully
an update on that over the next week on the list.

Talks -

Lots of good points here from all around, so I'm going to condense
here.

1. Lightning (10 or 12.5 minute) talks are useful for some things:
   * Mini-demos
   * Small project N+N and updates
   * Early stage project combos

All of these are valid starting points for conversations, which is what
a talk should be. But we think that there should not be a excessive
amount of these talks.

2. Talks (20 or 25 minute) are useful for pretty much everything
   * Gives space for more people to get a chance to speak on
     different topics
   * Is below the boredom threshold for slide-only talks (hopefully)

Consensus was that the majority of the talks should be of this
length.

3. Extended Talks (50 minute) are useful really just for one thing
   *  It's a 20 minute talk + 30 minute demo

The interleaving of actual demo work and slideware can extend
the mean-time-to-mego. But we think that there should not be too
many of these kinds of talks.

So there was a fair consensus on these timings from what I
heard. Considering the issue settled, there's two follow-on issues

* What are the proportions to be?
* What's the coupon strategy?

These are entangled in some complex manner. In terms of
proportions, I got the impression that the group was thinking
of

  'extended' < 'lightning' << 'talk'

I mentioned perhaps using a normal distribution to decide, but
I think I was just caught up in the moment, I didn't enjoy stats
at college at all :)

The more important point of coupons, passes, that kind of thing
came up. Everyone is aware of the tensions -

* Speaker can't get travel authorized unless it's a free pass,
* we can't give out free passes for all
* we want to reward the good stuff, regardless of talk length

What we talked about was that it is good to have some kind of
an algorithm to set the expectation of the submitter, but we
the PC must also have a pool of free passes that we can
dispense for what we perceive as quality. There was a general
agreement on that.

Of course, a PC-judgement-based system, while it makes
sense - we are, after all, professionals - can appear to be
whimsical and arbitrary to submitters. There was a suggestion
that perhaps submitters be made aware of the free pass pool
and that they might want to help the PC decide by submitting
either an extended abstract (abstract + paper), or indeed,
mirabile dictu, produce their slides early, i.e. compete.

Sundries
------------

I'm currently planning on not having a call next week, due
to ESE, but some kind of email synchronization will be
necessary - let the list know how things are getting on with
the panels, and if you have any other ideas, concerns or
inventions.

There was some issues with calendar entries being
screwed up, my apologies, and I'll look at sorting that
out.

Ok, let me know what I missed or misrepresented. Thanks
for the call today, nice to have a crowd on and get into
some details :)

 cheers
   --oh