Open Stack

On Thu, May 19, 2016 at 4:15 PM, Tristan Goode <tristan at aptira.com> wrote:
> I wasn’t serious to begin with, but faced with the two options of either:
>
> The Foundation dropping the voting and using the (same old) "cheer squad" to
> select talks..
> Or
> Your proposal of only letting speakers pick talks..
>
> It increasingly looks as valid selection method as any. Both the above
> methods, by dropping the audience's ability to have their say in any way at
> all say fuck the audience.
>
> Maybe voting for the track chairs is a way to give the audience some say.
> The current method of the foundation picking the cheer squad denies any
> critique.

Problem with that, just like we're now having a popularity contest
with talk voting, we would then have a popularity contest with talk
chair voting.

> I was once a track chair, but haven't been for a couple of years because I
> got dropped from the cheer squad possibly for not agreeing that everything
> the foundation and the TC do is beyond question. But at least I get to vote
> for talks and take great pleasure in voting against the usual blowhards that
> the cheer squad nominate _every_single_summit_.

Quite frankly, I don't have a good idea for naming track chairs, and
your grudge highlights exactly what the problem is with track chair
appointment.

I also don't have a good idea for talk voting that involves the entire
audience and doesn't make it a popularity contest. That's why I
proposed one where the group of reviewers is larger than the current
hand-picked track chair group, and where the votes are more meaningful
than with what we have now.

But here's another idea that does away with track chairs, with voting
talks into the Summit, and with favoritism altogether:

- Immediately after the CfP ends, all abstracts go public and are in a
review period for, say, two weeks.
- Anyone can comment on any abstract providing input and questions.
- The only possible audience *vote* that exists is pass/fail. Meaning,
would this be suitable for the Summit or not (the idea is that the
default is "pass", and that you would not light-heartedly give a fail,
particularly if your judgment is public).
- Once the review period is over, based on the number of fails given
overall, you can work out what's the cutoff to be eliminated from the
race.
- From all talks not thus eliminated, there's a randomized draw.

What does this mean?
- Everyone gets a chance at voicing their opinions.
- Speakers can get audience feedback before they even start their
talk, which gives us better talks.
- Everyone has the same fair chance at getting a talk slot.
- We get speaker churn; new speakers are more likely to get a shot
than in the current system which is biased in favor of repeat
performers.
- No-one needs to be disappointed because their talk doesn't get
accepted. It's just the roll of the dice.
- The only way to mathematically increase your chances of getting a
slot is to submit more talks, which is easily solved by capping the
total number of talks you can submit. Which we already do.

Thoughts?

Cheers,
Florian