Re: Git in Outreachy Dec-Mar?

["Ævar Arnfjörð Bjarmason" <avarab@gmail.com>]

On Thu, Aug 30 2018, Jeff King wrote:

> On Thu, Aug 30, 2018 at 02:18:19PM +0200, Ævar Arnfjörð Bjarmason wrote:
>> >   - it naturally limits the candidate pool to under-represented groups
>> >     (which is the whole point of the program, but if you don't
>> >     actually care about that, then it's just a complication)
>>
>> I'm fine with doing selection discrimination of under-represented groups
>> through such a program. Particularly if, as you mention, there's
>> earmarked funding for it which otherwise might not be available, so it's
>> not zero-sum when it comes to a hypothetical alternative of casting a
>> wider net of our own (and as you mention, that would be more work).
>
> Yeah, just for reference, my "you" there was a hypothetical "one might
> or might not care about...", not responding to your particular email.
>
>> I do think it's unfortunate that the selection criteria for the program
>> privileges U.S. citizens and U.S. residents above other people,
>> particularly since they're also accepting worldwide candidates (and
>> we've had at least one non-American participant that I know about), so
>> it's not e.g. for U.S. administrative or tax reasons as one might expect
>> if they only accepted Americans.
>
> I assume you mean this bit from the eligibility rules:
>
>   You must meet one of the following criteria:
>     - You live any where in the world and you identify as a woman (cis
>       or trans), trans man, or genderqueer person (including genderfluid
>       or genderfree).
>     - You live in the United States or you are a U.S. national or
>>       permanent resident living abroad, AND you are a person of any
>       gender who is Black/African American, Hispanic/Latin@, Native
>       American/American Indian, Alaska Native, Native Hawaiian, or
>       Pacific Islander
>
> So there are more categories for the US, but I think that is largely
> because under-representation is somewhat regional. Being black in the US
> is different than being black in Africa. Certainly one could argue that
> Africa as a whole is under-represented in the tech world, but I think
> you'd probably need to draw different boundaries in different places if
> you want to extend opportunities to those who are least likely to
> already have them.
>
> I don't know what those groupings would look like in, say, Europe. If
> you're suggesting that the program would be better off having
> region-specific rules for more regions, I'd certainly agree with that. I
> don't know if it's something the Outreachy folks have considered or
> discussed; it might be worth bringing it up.

[I don't mean to drag this up again, I had a draft here that I hadn't
sent, and thought given that I'm standing for the Git Project Leadership
Committee which presumably has something to say about this it was better
if I clarified].

I don't mean that just doing the equivalent of s/U.S. national//g on the
criteria would improve things, for the reasons you explained that
clearly wouldn't be an improvement or in the spirit of the criteria.

I was imagining that there was some way to phrase this that would
include the current group(s) but be country-neutral. E.g. instead of
talking about some specific minorities in specific countries say that if
you're in a group below such-and-such a percentage.

Although reading this again and consulting Wikipedia they seem to be
using all U.S. census groups below 20% with the exception of one (two if
you count "Other"), so I don't know how that would translate to other
countries, or if that's just an unintentional omission. Perhaps some mix
of group + mean income within that group? I don't know, and I'm not
familiar enough with the U.S. to speculate as to how they came up with
that.

Or, just a third criteria of:

    Projects can opt-in to consider non-U.S. nationals or residents who
    they believe fulfill the spirit of criteria #2 as it would apply to
    another country.

Then we could (if Outreachy approves) opt-in to that, since considering
that on a case-by-case basis is surely less gnarly than trying to come
up with some general rule.

So again, I don't think this particular thing is a big deal, or
something worth spending time worrying about at this point. Just
something to keep an eye out for and potentially gently poke Outreachy
about.

I just think we might stand to get better/more candidates and have more
fair process, and be seen to spend project funds in a less biased way if
the criteria wasn't an OR'd statement whose second half starts off by
outright limiting itself to less than 5% of the world population based
on a specific nationality, before further narrowing things down.