Kicking off the GNU Assembly

Posted Apr 16, 2021 13:15 UTC (Fri) by Zack (guest, #37335)
In reply to: Kicking off the GNU Assembly by dvdeug
Parent article: Kicking off the GNU Assembly

Nix' Swaggart-like performance not withstanding, doing the work doesn't give you governance [1] for the simple but counter-intuitive reason that GNU maintainers need not be Free Software people.

This is how we can have an Emacs maintainer that openly uses and develops on OSX. It's not ideal, but if someone is only interested in GNU for the technical challenges it provides, that's okay. It does mean there needs to be an ultimate arbiter when it comes to applying software freedom.

Now if this "ultimate arbiter" role would have been challenged in good faith, there might have been a constructive discussion, because rms is not immortal, but it became clear that this movement was mainly about removing rms from GNU without demonstrating how he had failed in his role as "ultimate arbiter" or even how to set up a replacing governance.

So no, "doing the work" is not sufficient grounds for wanting to remove rms, because that would mean that developers that are full-time employed because their interest lies in commercially viable projects would be over represented and, except for the minimum, GNU maintainers are not beholden to the idea of Free Software.

[1] for GNU as a whole project. Every maintainer has full autonomy with regards to their own project, including installing a CoC.

Kicking off the GNU Assembly

Posted Apr 16, 2021 15:43 UTC (Fri) by farnz (subscriber, #17727) [Link] (14 responses)

But if all the projects leave GNU as a whole, what is left for GNU to govern?

This reads to me like people getting progressively more frustrated with the governance of GNU as a whole, for reasons separate to free software, unable to get traction within GNU for fixing it, and instead walking away. Would you be happier if they forked their projects, the way egcs did back in the day, and left GNU to govern the left-overs that didn't fork?

Kicking off the GNU Assembly

Posted Apr 16, 2021 16:20 UTC (Fri) by mhw (guest, #13931) [Link] (13 responses)

But if all the projects leave GNU as a whole, what is left for GNU to govern?

For the record, only about 6% of GNU maintainers have ever supported these attempts to radically change GNU governance. There are currently 386 GNU maintainers, and 23 of them support this.

Kicking off the GNU Assembly

Posted Apr 17, 2021 0:10 UTC (Sat) by dvdeug (guest, #10998) [Link] (1 responses)

You mention only current statistics, so I don't know why you say "have ever supported". If 23 maintainers have added their names to this, that doesn't mean that only 23 support this; it means that only 23 have come out in support. Nor does it mean that the other 94% support you; I suspect a full poll would find some more full support for the GNU assembly, and a wide range of opinions from mild support for the GNU assembly, to apathy, to mild support for the status quo, to full support for the status quo. I don't know what the exact breakdown would be, but it would be a lot more complex than your 6% implies.

It strikes that I can't recall a single new GNU project since GNOME and GNU Classpath. Looking it up, I see GNU Poke, nice, but hardly major. Existing projects continue to be important, but new projects don't seem to be joining the GNU project.

Kicking off the GNU Assembly

Posted Apr 17, 2021 1:00 UTC (Sat) by mhw (guest, #13931) [Link]

You mention only current statistics, so I don't know why you say "have ever supported".

Back in 2019 and 2020, I computed similar statistics for the signers of their Joint statement on the GNU Project (2019) and GNU Social Contract (2020).

They had plenty of time to collect signatures for those two documents, and even took the questionable step of scraping emails from GNU's internal list of maintainers in order to send email directly to every GNU maintainer, soliciting feedback and signatures for their "GNU Social Contract", and arguably giving the false impression that this was an official action by the GNU project.

Ultimately, they were not able to persuade more than 6% of GNU maintainers to sign either of those documents. More precisely, 20 current GNU maintainers signed the first document, and 21 signed the second.

Kicking off the GNU Assembly

Posted Apr 17, 2021 0:50 UTC (Sat) by NYKevin (subscriber, #129325) [Link] (10 responses)

Please excuse my ignorance, but I would like to better understand this 6% figure.

What exactly is a GNU maintainer, and what do they do? Is that just another word for "developer," or do they have some additional authority?
If it is not a synonym of "developer":
- How are they chosen?
- Why are there 386 of them, instead of 387 or 385, or indeed 3,860 or 3?
- Why shouldn't the opinions of regular developers count? Do we know those opinions, or have any way of measuring them?
Do maintainers all do exactly the same amount of work?
Are they all currently active in development efforts, or are there maintainers emeritus?
Has there been a formal survey to establish the views of the other 363 maintainers?
Realistically, if the 23 maintainers involved walked away from GNU and forked any projects with which they were involved, what would be the likely outcome?

Kicking off the GNU Assembly

Posted Apr 17, 2021 1:24 UTC (Sat) by mhw (guest, #13931) [Link] (9 responses)

What exactly is a GNU maintainer, and what do they do?

This, and most of your other questions, are answered here.

Do maintainers all do exactly the same amount of work?

Of course not. I guess this was a rhetorical question.

Has there been a formal survey to establish the views of the other 363 maintainers?

Ludovic Courtès sent email directly to every GNU maintainer in early 2020, soliciting feedback and signatures for the "GNU Social Contract". Only about 6% of them signed the document.

Realistically, if the 23 maintainers involved walked away from GNU and forked any projects with which they were involved, what would be the likely outcome?

I'm not sure, but they certainly have the right to do that (thanks to Richard Stallman, I might add). What they do not have the right to do is to seize the name "GNU" for their own purposes.

Second request.

Posted Apr 17, 2021 1:53 UTC (Sat) by corbet (editor, #1) [Link] (8 responses)

There are now at least nine comments from you on this article mentioning "6%".

We get your point

I will ask you one more time: stop now. Repetition does not help your case. It is time to end this interminable cycle.

Second request.

Posted Apr 17, 2021 14:50 UTC (Sat) by hummassa (subscriber, #307) [Link] (5 responses)

May I respectfully disagree? It is my opinion that the whole thing smells like a coup, as I already stated. The numerical fact stated just lends more plausibility to that hypothesis.

Second request.

Posted Apr 17, 2021 15:24 UTC (Sat) by corbet (editor, #1) [Link] (4 responses)

For the record, I was not objecting to the posting of the 6% number¹. My objection was the broken-record experience² of posting it over and over and over and over again. There is a line of reasoning that says whoever posts last in a thread "wins" the point, but it's not true and creates a severely tiresome experience for everybody else involved.

¹OTOH, that number shouldn't necessarily be accepted without question. As I understand it, the GNU project has a lot of listed maintainers who have not been active in years, for example.

²For the young folks out there who have not spent years enduring broken records, substitute "infinite loop with noise".

Second request.

Posted Apr 17, 2021 16:20 UTC (Sat) by hummassa (subscriber, #307) [Link]

Oh, I don't object to the "tiresome experience" bit, at all. As a "merely (sporadically) commenting" person, I think I understand your point, as you have a bigger stake in the quality of the discussion. There is the problem that, as any discussion grows in size, newcomers tend to answer similarly to different nodes of the discussion tree (I know I often comment something just to see it in such-and-such wording some branches ahead). This happens because, well, that person there is *obviously wrong* at that point in the discussion, and one *must answer*.

And, yes, RMS *does* elicit emotional (and sometimes irrational) responses. The calm and thought you asked before any response is a reasonable request.

Second request.

Posted Apr 17, 2021 20:13 UTC (Sat) by mhw (guest, #13931) [Link] (2 responses)

Hi Jonathan,

For the record, I was not objecting to the posting of the 6% number. My objection was the broken-record experience of posting it over and over and over and over again.

It's possible that I've repeated myself too many times on a few of my points. It's hard for me to judge that objectively; I'm not sure that anyone with an opinion on this issue can do so. There's been a lot of repetition of misinformation here as well, and I feel compelled to counter it, but I also acknowledge that at some point, someone has to be willing to let the other side have the last word.

However, I believe that my repeated use of "6%" is quite different. I've recently started using it as my preferred label for this minority faction, in the same way that the Occupy movement used the label "1%". I personally don't see what's wrong with that. I disagree that using that label is tantamount to repeating the same point over and over again.

Second request.

Posted Apr 18, 2021 15:32 UTC (Sun) by rsidd (subscriber, #2582) [Link] (1 responses)

You should probably update it to "the 7.7%" then: 30 are currently listed, out of, you say, 386 maintainers.

That's without getting into the question of how many of the 30, and how many of the 386, are active maintainers.

Second request.

Posted Apr 18, 2021 21:22 UTC (Sun) by mhw (guest, #13931) [Link]

Only 24 of them are GNU maintainers, so it's still 6%, or more precisely 6.2%.

Second request.

Posted Apr 17, 2021 17:46 UTC (Sat) by NYKevin (subscriber, #129325) [Link]

It would appear that different commenters subscribe to different norms about the level of discourse that is acceptable on LWN. In other parts of this thread, I have seen a fair amount of back-and-forth over this, some of which struck me as quite rude.

Might it be a good idea to formally write up the level of discourse that you expect of commenters, and post it somewhere prominent? I checked the FAQ but found nothing on this point.

Second request.

Posted Apr 17, 2021 21:37 UTC (Sat) by rodgerd (guest, #58896) [Link]

I know that it's unlikely to be a worthwhile use of your time and energy, since I doubt it would change any minds, but it would be fascinating to see something similar to the breakdown that you do of kernel dev contribution stats for gcc and the other important projects under the GNU umbrella.