5 ms·
> Edit: Yes, there are better and worse ways to go about it. Realising too late and reverting the original list was not the best. Reaching out to / inviting div
by 2muchcoffeeman 9y ago
> Edit: Yes, there are better and worse ways to go about it. Realising too late and reverting the original list was not the best. Reaching out to / inviting diverse groups rather than fixing balance later would be better. Hope they learned and will do it better next year. Organisers which care about it from the beginning somehow manage to get a decent m/f split - both in the audience and in the speakers. (see linux.conf for a good example)
A couple other comments said that they did a blind review of the submitted papers. I can't think of a farer review process.
- boomboomsubban 9y agoIf the selection committee doesn't represent a diverse field, a blind review can still be biased. I write things in a way I find compelling, other people that find them compelling are likely similar to me.
- huffmsa 9y agoThe selection committee should only be "diverse" in so far as it contains a mix of people who understand what makes a good Electron topic and a people who understand what topics make good conference presentations. If in fact, the goal is to out on an enjoyable conference on the topic of Electron.
- viraptor 9y agoNo review process is more fair. But if you end up with all-X selections from a blind-review, you may be advertising to a very homogeneous culture. As a result, you may be limiting your pool of ideas to choose from and reaching only a local maximum. There are usually great non-X engineers, as long as you reach out to them. (they're usually have their own group - guess why?) Mixing multiple varied-X groups usually results in more diverse discussions and ideas, in my experience. On the other hand, not caring about it and accepting that you're just getting dominant group X submissions is not a terrible thing. But we're social creatures - if everybody nearby is similar but different from you, either you'll feel uncomfortable, or have a bard-level charisma. So by allowing your group to become almost entirely X, you're enforcing the limit on your submitted papers pool even further. (applies for many (all?) X, where X influences your life in any way)
- julianmarq 9y ago> (they're usually have their own group - guess why?) Self-segregation, so in-vogue nowadays. That, by the way, is also not the way to do things. > No review process is more fair. But if you end up with all-X selections from a blind-review, you may be advertising to a very homogeneous culture. As a result, you may be limiting your pool of ideas to choose from and reaching only a local maximum. There are usually great non-X engineers, as long as you reach out to them. Mixing multiple varied-X groups usually results in more diverse discussions and ideas, in my experience. Moving the goalposts, as usual for this kind of crowd. What would be your way to blame everyone but the people not submitting papers if you were told that the organizers did reach out as many different groups as possible? I know of several, I'm just curious to read which you'll use.
- viraptor 9y ago> if you were told that the organizers did reach out as many different groups as possible? If we were discussing what-ifs, then I'd say - tough, maybe next year they'll have more success. But that's not the situation we're discussing. An organiser responded with "we can do better, we're postponing", so I don't have really a comment to that.
- julianmarq 9y agoPost hoc ergo propter hoc sure is handy... in this particular situation.
- viraptor 9y agoI honestly don't know where do you see an invalid temporal causality here. I can't unsee the organiser's response :)
- 2muchcoffeeman 9y agoIn this instance the speakers were all men. Can you show that GitHub specifically advertised to only men?
- bsg75 9y agoMany commenters here, and apparently the conference organizers, appear to misunderstand that blind review and random sampling are not the same thing.