[EuroPython] Lack of diversity within selected talks

[Radomir Dopieralski]

On 04/15/2014 02:14 PM, Radomir Dopieralski wrote:
> On 04/15/2014 02:11 PM, Simon Cross wrote:
>> Greetings
>>
>> Is it possible to get some demographics on authors who submitted talks
>> vs authors of accepted talks? It would be good to know whether the
>> blind selection process functioned as intended (i.e. the demographics
>> of the two are statistically similar) or not.
>>
>> The answer probably won't help us fix things for this year, but it
>> will tell us where more effort is needed next year.

I'm sorry for the previous, empty reply, I hit "send" too soon.

I'm all for checking the statistics, especially to check if, by any
chance, all the more diverse speakers are not in the "disputed" group of
talks, which still have a chance of being accepted. It wouldn't be
surprising at all, since they are likely to stand out and therefore
attract votes both in plus and in minus.

There is also the fact, that blind review doesn't really work so well,
if the talk submitters know some of the reviewers, and can ask them for
voting on them (or even without asking, when the reviewers know the
topics of the talks submitted by their friends). The effect is opposite
to the intended -- the club becomes even more closed.

-- 
Radomir Dopieralski