Re: Guix (and Guile's) promise, and how to (hopefully) get there

[Ekaitz Zarraga <ekaitz@elenq.tech>]

On 2024-10-27 19:12, paul wrote:
> Hi Ekaitz,
> 
> On 10/27/24 11:47, Ekaitz Zarraga wrote:
>> Hi,
>>
>> On 2024-10-27 11:00, indieterminacy wrote:
>> > I think a useful measurement of any community is the diversity 
>> within it.
>>
>> I do think we are a pretty diverse group.
>>
>> But it all comes to what you consider diverse. Some people brought 
>> this conversation during the Guix Days, probably ignoring the fact 
>> that we had people coming from all over the world. I don't know what 
>> was their definition of diversity at that moment.
> I can't talk for the people you mention, but usually by diversity one 
> means social diversity, it meansco-existence of different social groups 
> within a given setting.

But what is a social group?

>> On the other hand, we have something in common, so it's really hard to 
>> be diverse in the broad sense. Should we include people that like 
>> proprietary software, too? Those are also people, and I'm sure they 
>> would feel uncomfortable between us.
> For sure they would, but "proprietary software user" is not a social 
> group, so they would not be meaningful in a measure of  the Guix project 
> social diversity.
If you go to the wikipedia definition of it, it says social groups come 
in myriads of sizes and varieties, and its definition is pretty loose:

   In the social sciences, a social group is defined as two or more 
people who interact with one another, share similar characteristics, and 
collectively have a sense of unity.

Proprietary software user is not a social group, but two proprietary 
software users that are Apple fanboys is.

So, what you propose that is very well defined I don't think it's 
totally true, at least not to the extent that is going to let us take 
accurate and predictable decisions.

>> This is not to say we shouldn't try to make things better and more 
>> welcoming, of course we should. But I don't think "diversity" actually 
>> means that much as a measure because I don't think it's an absolute 
>> concept and I think it's very easy to misunderstand.
> 
> I am not an expert but there are many social scientist that are, for 
> sure for them social diversity is a pretty clear concept since they 
> sometimes have to measure it for their research work.


As I know, few social concepts are really measurable in terms an 
engineer would feel confortable with. I'm open to be educated in the 
subject, but if someone has a proper measure for diversity in a 
community that is not affect by bias and is applicable to us and can do 
a good prediction on how investing everything on it can really change 
the quality of the software we produce and the quality of life of the 
ones who produce it, please let me know.

Until we have it, we are just talking about things we don't know, and 
I'm not interested on that kind of conversation.

 >
 > While for sure all your points about care, overwork and burnout are
 > valid, I believe the fundamental problem to be of governance.

We can agree with this. Most of our discussion goes in this direction.
But just that is not enough to change things.

 >  To make
 > Guix better software (there is scientific consensus that diverse
 > communities have better governance, it is easy to find), we should find
 > out more about how to measure social diversity and take concrete actions
 > towards making the Guix community a more diverse one.

What is "better" governance?
That's what we are trying to define here I think. A governance that fits us.

With the social diversity here we can't really do much, as I said. Every 
day I work in Guix with people from many different backgrounds and 
ideas, but still we are always going to be people that likes computers 
and share some cultural aspects like talking in English, for example.

Again, this is a futile conversation because there's not much we can do 
and those who push the debate on this direction don't really have 
specific proposals: just throw more diversity to the thing.

That's not the solution to everything.

Mostly, because as we didn't measure (and it's not clear yet that is 
measurable), we might be trying to solve a problem we don't have, that 
we cannot fix, or that is inefficient to tackle due to other reasons.

And also, there's only so much diversity we can throw to this, and it's 
not clear to me that we didn't peak on it already.

 >
 > At last, image is fundamental otherwise big corporations wouldn't spend
 > so much to try to do whatever-washing, getting distance between us and
 > GNU/FSF would concretely help very much with diversity.

Big corporations also spend a lot of money lying to you, selling you 
things you don't need, lobbying to governments and so on, and I wouldn't 
say any of those are a good thing.

They need the image to do those things. We are not a corporation. We 
don't sell things. We don't lie, we convince.

I don't think we can compare to that.

If we consider the image is important, as you suggest, we should also 
remember the GNU and the FSF have also a reputation. We could decide to 
reject that with the promise of more diversity, but we didn't make sure 
yet diversity is the only thing we need.

Let's remember the FSF still pays most of our bills. Regardless this 
might be a good argument or not, I don't feel comfortable risking all 
our financial stability for a promise that we would be more diverse 
(while still being a super-niche project written in a weird programming 
language almost nobody uses), even more when nobody can prove that 
diversity is our only issue.

Many people on this project have tried to change GNU from the inside and 
are very critical with the FSF (see the https://gnu.tools/). I think 
that's also a good way to do things, changing them from the inside. 
Fixing them for all our friends. Honestly, the argument of getting 
distance with GNU and the FSF is too simplistic to be taken seriously.

Again, this discussion as interesting as it might be for the very long 
term, is uninteresting to me because it won't produce any practical output.

 >
 > Thank you all in the Guix community for your awesome work so far,

Thank you for sharing your thoughts. I might sound a little bit harsh, 
but it's not on purpose.

Ekaitz