Re: [Debconf-team] Report from the talks team
Anthony Towns dijo [Fri, Sep 19, 2014 at 09:37:15PM +1000]:
> >> * We must find a way to make submitters to make better talks
> >> descriptions. Bad or incomplete talks description made to waste
> >> a lot of time to both the talks team and attendees.
> > Yeah, I can see this very well. We should make sure that people put
> > at least as much time into making a submission as it takes us to
> > evaluate it.
>
> Would some sort of wiki-ish approach to talk proposals be possible?
> ie, let people propose talk ideas publically, with the ability for
> other people to help improve the description, add suggestions or
> correct typos before the talk review happens? Could let attendees
> provide an indication of interest in a topic in advance too?
Humh... In one other conference I helped run at some point, as the
talks team we tried to group people that wanted to present similar
topics — Even suggesting unrelated speakers of too-similar topics to
get in touch with each other and present a single talk proposal.
I'll be gentle, and will limit myself to saying... The idea didn't
fly.
If I'm proposing a talk, I might be interested in some input on how it
is viewed. But I usually don't want others to _tell me_ what is my
talk going to be about. Many of the talk submitters ended up just
dropping their proposals.
Of course, different conferences are made by different people,
but... I don't think this would be a very good idea.
> An alternative approach: just reject any talks with poor descriptions.
> Try to tell submitters early if their description isn't good enough --
> maybe give them a short extension after the deadline to resubmit a
> better description even, but otherwise leave it up to the submitter.
> Worst case, they get rejected and can organise an ad-hoc session,
> can't they?
We have tried to do this, with decent results. This last year I have
been too busy with all kinds of stuff to really have my head sitting
properly on top of my shoulders, so I might have only partial memories
of the work we did, but in past years we _did_ ask prospective
speakers to fill up the descriptions - Or accept we would vote them
down. A talk title is NOT acceptable as a talk proposal.
> > - Ask participants to provide links to previous events or videos,
> > allowing us to evaluate the quality of the speaker. Note that
> > I am not talking about witty audience magnets only, and I have
> > seen fantastic(ally prepared) speakers who presented in their
> > !first language and didn't have perfect slides.
>
> Does/can debconf offer any help to poor speakers with great ideas?
> Like, maybe hooking up a new speaker with an experience speaker to
> help draft/review slides, or something like that? Could have some
> volunteers available to help folks write good descriptions for their
> proposal too, maybe?
I am not a fan of this idea. We should IMO strive to judge talks based
on their content, interest, relevance, etc. Of course, the talk
presenter will largely shape the interaction, but anyway — Shunning
somebody because they don't have a great on-stage personality does not
seem fair to me.
Over the years, I have refered to and recommended Meike Reichle's
great presentation from eight years ago:
http://ftp.acc.umu.se/pub/debian-meetings/2006/debconf6/theora-unscaled/2006-05-19/tower/Cheap_thrills_Instant_inspiration_for_the_masses-Meike_Reichle.ogg
And yes, your suggestion is good, although it would probably not be
something doable by the Talks Team itself but by just about everybody:
Helping speakers prepare their material.
In earlier DebConfs, we used to print proceedings. Then, proceedings
became digital only. And starting several years ago, people don't
prepare a written document before presenting the talk, because after
all, it won't be read. And, of course, every day we have legions of
people missing talks because they are busy preparing their own talk
for the following day.
If we could in some way recover the practice to prepare a small paper
for a talk presentation, I think the aspects we are discussing would
surely get better. But I don't know how we can require people to
prepare a paper.
Reply to: