[EuroPython] Lack of diversity within selected talks

[Martijn Faassen]

On 04/15/2014 11:06 PM, Armin Rigo wrote:

 > In this case, woman
 > participation is going slowly up year after year.  I certainly think
 > (and hope!) that it's not just because of favorable discrimination;
 > instead, it is most probably just a slow process of natural regulation
 > that occurs inside a historically strongly biased subculture.  This
 > process can be encouraged, e.g. I'm fine if some grants are reserved
 > to women; but I think that judging technical merits on a different
 > scale is not a good way to do that.

There are a lot of things that can be done instead of quotas.

I think one function of a Python conference is to help foster the Python 
community. If we agree that we would like to have more women speakers 
and participants, or just plain broaden the nature of our conference in 
general, then you can actively work towards in a whole range of ways:

* Looking for high-profile female invited speakers.

* Broadening the scope of topics. The conference should still be Python 
themed, but the occasional talk about, say, morality or astronomy or 
game development or business can be fit in. I remember such talks from 
previous EuroPythons. Keynotes tend to do this already, but there's no 
reason to restrict this to keynotes. I myself find that such variety 
improves the conference and makes it more inspirational for me.

* Considering whether we want a self-selected democracy for anonymously 
selecting talks based on individual merits, or whether we want to 
involve other methods too. Say a smaller group of people that looks at 
the overall balance of things.

* Judging talk proposals on other things than technical merit only. 
Originality, presentation, humor, all of these count. Armin is a good 
example actually: your talks wouldn't be half as much fun for people 
without your presentation style. This may be written down somewhere 
already for all I know in the talk selection guidelines actually, but if 
not, that may make sense.

* Having women visibly be present at the conference. PyCon DE last year 
was a good example; there were a lot of women involved with its 
organization. You can also make this visible explicitly, like at PyCon 
DE: everybody involved was called onto the stage in the end. I 
understand many of them are involved in the organization of EuroPython 
this year. I would certainly recommend getting folks on the stage again 
at some point (though I would be bold enough to ask whether you could 
speed up that procedure compared to PyCon DE).

* As was proposed, simply increase variety of speakers by having each 
speaker only have one talk.

* Active outreach to PyLadies and such. It's my understanding that this 
exactly that was done.

* Some conferences let sponsors give some talks. That's a tricky thing 
to get right. But here's a less controversial idea: for a community 
organized conference I think it's fair if active organizers get a good 
chance at getting *their* talk submissions approved. And then if 
PyLadies is involved...

* Grants, as you mention.

Some of these ideas *do* influence the talk selection process, but not 
in the form of quotas. The talk selection process is influenced by many 
factors already, and we shouldn't pretend that the current way is only 
fair way to do things.

PyCon is the obvious place to go look for more/better ideas.

Regards,

Martijn