5:52 p.m.
Hi, I'm Ben Pfaff. I've worked on a few GNU projects, but
the best known might be GNU PSPP, which is a free software
clone of a proprietary statistical program. I spend the time
I have available on that.
I founded GNU libavl, which I considered complete a
few years ago but could now use some attention. I will go
back to it someday.
I worked on GNU Checker for a few years, but that was a
long time ago and it is a retired project now.
I have also contributed to Gnulib.
I acknowledge and agree to the social contract and code of
conduct.
I'd appreciate being added to the webpage with appropriate
affiliations. My personal webpage is https://benpfaff.org/.
6:31 p.m.
New subject: Introduction: [...]
Hi -
I acknowledge and agree to the social contract and code of conduct.
(Not singling Ben out at all, but a general comment:)
I am getting the sense that some members of this group are
contemplating actions much more specific than that, proposing or
organizing FSF-disconnected governance structures, possibly forking
GNU projects.
To a fence-sitter or outsider, that doesn't make sense, becuase the
web site's social contract / conduct terminology say nothing about
that. Heck, it seems to me that RMS himself or his fans could all
come here, and with full honesty "acknowledge and agree" to the
general contract/conduct. It is all about aspirational present/future
behaviour, not the past, after all. What then?
Is this group pre-committed to forming new project bodies, setting
rules, taking other actions that change existing maintenance /
affiliation relationships, or excluding people who might not have
lived up to your understanding of the social-contract in the past? If
so, IMHO the web site should just come out and say that. Clarity is
good!
- FChE
6:40 p.m.
New subject: Introduction: [...]
On Fri, Apr 16, 2021 at 11:31 AM Frank Ch. Eigler <fche(a)redhat.com> wrote:
Is this group pre-committed to forming new project bodies, setting
rules, taking other actions that change existing maintenance /
affiliation relationships, or excluding people who might not have
lived up to your understanding of the social-contract in the past? If
so, IMHO the web site should just come out and say that. Clarity is
good!
I don't have any preconceived notions, myself. I don't know what
others are thinking.
6:56 p.m.
New subject: Introduction: [...]
On Fri, Apr 16, 2021 at 2:31 PM Frank Ch. Eigler <fche(a)redhat.com> wrote:
I am getting the sense that some members of this group are
contemplating actions much more specific than that, proposing or
organizing FSF-disconnected governance structures, possibly forking
GNU projects.
Please feel free to just ask what people are thinking.
To a fence-sitter or outsider, that doesn't make sense, becuase
the
web site's social contract / conduct terminology say nothing about
that. Heck, it seems to me that RMS himself or his fans could all
come here, and with full honesty "acknowledge and agree" to the
general contract/conduct. It is all about aspirational present/future
behaviour, not the past, after all. What then?
Could you please expand on what doesn't make sense?
Is this group pre-committed to forming new project bodies, setting
rules, taking other actions that change existing maintenance /
affiliation relationships, or excluding people who might not have
lived up to your understanding of the social-contract in the past? If
so, IMHO the web site should just come out and say that. Clarity is
good!
Could you please clarify your question a bit more?
Let me try to rephrase and tell me if I have it right:
* Would the GNU Assembly accept someone who previously did not live up
to the Social Contract but committed to it going forward?
* Would the GNU Assembly accept RMS if he agreed to abide by the
Social Contract and the CoC?
* What would the GNU Assembly do if someone violates the Social Contract?
* What would the GNU Assembly do if someone violates the CoC?
Cheers,
Carlos.
7:33 p.m.
New subject: Introduction: [...]
Hi -
> I am getting the sense that some members of this group are
> contemplating actions much more specific than that, proposing or
> organizing FSF-disconnected governance structures, possibly forking
> GNU projects.
Please feel free to just ask what people are thinking.
What are people thinking?
> To a fence-sitter or outsider, that doesn't make sense,
becuase the
> web site's social contract / conduct terminology say nothing about
> that. Heck, it seems to me that RMS himself or his fans could all
> come here, and with full honesty "acknowledge and agree" to the
> general contract/conduct. It is all about aspirational present/future
> behaviour, not the past, after all. What then?
Could you please expand on what doesn't make sense?
It does not make sense to conflate nominal support for the
social-contract with support for the other things.
> Is this group pre-committed to forming new project bodies,
setting
> rules, taking other actions that change existing maintenance /
> affiliation relationships, or excluding people who might not have
> lived up to your understanding of the social-contract in the past? If
> so, IMHO the web site should just come out and say that. Clarity is
> good!
Could you please clarify your question a bit more?
Let me try to rephrase and tell me if I have it right:
* Would the GNU Assembly accept someone who previously did not live
up
to the Social Contract but committed to it going forward?
* Would the GNU Assembly accept RMS if he agreed to abide by the
Social Contract and the CoC?
* What would the GNU Assembly do if someone violates the Social Contract?
* What would the GNU Assembly do if someone violates the CoC?
Those are good questions.
Plus "What development community organizational characteristics or
changes are logically Required in order to act in harmony with the
social contract?".
They would help answer the key one: "What is the gnu assembly, really?"
- FChE
10:39 p.m.
New subject: Introduction: [...]
Hi Frank,
> > I am getting the sense that some members of this group are
> > contemplating actions much more specific than that, proposing
> > or
> > organizing FSF-disconnected governance structures, possibly
> > forking
> > GNU projects.
>
> Please feel free to just ask what people are thinking.
What are people thinking?
I have no plans to fork anything — not unprompted anyway.
Proposing governance structures — yes, that is what I think is a
defining feature. There is little point in establishing a shared
vision when there is no collective agency. People don’t agree to
any form of governance by joining the assembly (certainly not at
this point when no such thing exist). They merely step forward to
become part of the discussion that may lead to a governance model
that takes the social contract into account.
(It may also fail entirely.)
It does not make sense to conflate nominal support for the
social-contract with support for the other things.
The social contract is the bare minimum that we agree on without
further discussion. Anything concrete that goes beyond that is to
be determined by the assembly through deliberation, discussion,
and eventual consensus (with consent of those who have not
expressed a different option). There is no formal process for
reaching consensus, but the *necessary conditions* for consensus
have been laid with the social contract (the domain of shared
values) and the code of conduct (the bounds of acceptable
discourse).
> * Would the GNU Assembly accept someone who previously did not
> live up
> to the Social Contract but committed to it going forward?
I don’t think this can be answered in absolute terms — I’m eyeing
the last paragraph in particular. It would likely be difficult to
reconcile, say, “committing to providing a harassment-free
experience for all contributors” with accepting a seemingly
reformed serial harasser into the fold (purely hypothetical
scenario).
This does raise the question of an acceptance process, though. We
don’t have one, in part because there is no actual membership and
no formal decision making processes — membership is inherited from
the existing GNU “project”. Currently, the assembly is open to
any GNU maintainer or contributor endorsing the social contract
and accepting the CoC.
With an eye for the future and on preventing “infiltration”:
perhaps we should agree on an acceptance process.
> * Would the GNU Assembly accept RMS if he agreed to abide by
> the
> Social Contract and the CoC?
This is too hypothetical for me to answer, because I find it very
unlikely for this case to become reality. If rms had such a
fundamental change of character that he would agree to give up his
claim to absolute leadership and be
> * What would the GNU Assembly do if someone violates the Social
> Contract?
The document spells out what it calls the “core committments of
the GNU Project”; when contributors ignore the social contract
(e.g. by changing the license to their project in violation of the
four freedoms) it would become almost impossible for the rest of
the project to cooperate with them, as one of the necessary
conditions for consensus would no longer exist.
> * What would the GNU Assembly do if someone violates the CoC?
A significant portion of the CoC is dedicated to procedure in that
case.
--
Ricardo
2:03 p.m.
New subject: Preventing infiltration
Hi,
I agree with what you wrote, remarkably clearly, as always.
Ricardo Wurmus <rekado(a)elephly.net> skribis:
With an eye for the future and on preventing “infiltration”: perhaps
we should agree on an acceptance process.
Yes. It just occurred to me that infiltration is simple right now:
submit a patch to Guix (picking it as an example because the barrier to
entry is quite low), at which point you can be considered a “GNU
contributor”, and could apply in bad faith, with the intent of stirring
dissent.
I wouldn’t want this organization to be driven by paranoia, but I think
it’s a plausible scenario given what we already observed before.
So yes, I’d be in favor of setting up an acceptance process real soon.
Last week, Andreas explains how the Guix Europe non-profit handles
membership: membership requests must be approved by its “solidary
administrative council”¹. The bylaws also spell out reasons for
membership cancellation (Article 7).
There’s no such council/board at this point in the Assembly. Should we
mimic this or other similar structures, by setting up a board of sorts?
If we do, we must address at list these points:
• guaranteeing turnover of the board;
• defining an approval process for the initial board statutes;
• agreeing on the initial board members;
• ensuring the statutes don’t give the board more power than
membership management.
Thoughts?
What do we do until this structure in this place though?
We could, for example, put membership requests by non-maintainers on
hold until we’ve set this up. It would be a bummer, because we want
equal participation of all contributors, but that’s a simple way to
prevent the obvious infiltration scenario discussed above.
Another way would be to require applicants to have at least (say) two
years of experience contributing to GNU in a visible way.
Lots of questions. :-)
Ludo’.
¹ Article 5 of its bylaws:
<https://git.savannah.gnu.org/cgit/guix/maintenance.git/plain/guix-europe/...;.
2:14 p.m.
New subject: Preventing infiltration
Hi -
On Sat, Apr 17, 2021 at 04:03:46PM +0200, Ludovic Courtès wrote:
> With an eye for the future and on preventing “infiltration”:
perhaps
> we should agree on an acceptance process.
Yes. It just occurred to me that infiltration is simple right now:
submit a patch to Guix (picking it as an example because the barrier to
entry is quite low), at which point you can be considered a “GNU
contributor”, and could apply in bad faith, with the intent of stirring
dissent.
[...]
I'm sorry, what is wrong with dissent? Do you believe the
code-of-conduct is insufficient to limit uncivil conduct? Do you have
examples of propositions that you believe must be unanimous here?
- FChE
2:27 p.m.
New subject: Preventing infiltration
Hi Frank,
"Frank Ch. Eigler" <fche(a)redhat.com> skribis:
On Sat, Apr 17, 2021 at 04:03:46PM +0200, Ludovic Courtès wrote:
> > With an eye for the future and on preventing “infiltration”: perhaps
> > we should agree on an acceptance process.
>
> Yes. It just occurred to me that infiltration is simple right now:
> submit a patch to Guix (picking it as an example because the barrier to
> entry is quite low), at which point you can be considered a “GNU
> contributor”, and could apply in bad faith, with the intent of stirring
> dissent.
> [...]
I'm sorry, what is wrong with dissent?
I don’t think there’s anything wrong with dissent, or I wouldn’t be
here. :-)
What I had in mind above is an infiltration scenario whereby someone
joins, remains within the bounds of acceptable discourse as stated by
the code of conduct, never overtly rejects the values expressed in the
social contract, yet effectively prevents the group from moving forward
by derailing discussions, questioning its foundations, and so forth.
That’s probably more or less the definition of infiltration.
Do you believe the code-of-conduct is insufficient to limit uncivil
conduct? Do you have examples of propositions that you believe must
be unanimous here?
In my view we’ll seek consensus for most decisions, which is not the
same as unanimity—I can consent to a proposal, even if that’s not my
preferred choice.
Ludo’.
2:41 p.m.
New subject: Preventing infiltration
Hi -
What I had in mind above is an infiltration scenario whereby someone
joins, remains within the bounds of acceptable discourse as stated by
the code of conduct, never overtly rejects the values expressed in the
social contract
OK ...
yet effectively prevents the group from moving forward by derailing
discussions, questioning its foundations, and so forth. That’s
probably more or less the definition of infiltration.
If you already have notions of what directions of forward movement, or
foundational interpretations should not be questioned, please post
them for clarity. If no, then how can anyone predict what you or
someone else will label "infiltration" versus legitimate discussion?
- FChE
3:39 p.m.
New subject: Preventing infiltration
On 4/17/21 10:03 AM, Ludovic Courtès wrote:
We could, for example, put membership requests by non-maintainers on
hold until we’ve set this up. It would be a bummer, because we want
equal participation of all contributors, but that’s a simple way to
prevent the obvious infiltration scenario discussed above.
Another way would be to require applicants to have at least (say) two
years of experience contributing to GNU in a visible way.
Lots of questions. :-)
Ludo’.
I am a sort of marginal case as a GNU contributor, and perhaps it might
be helpful to consider my own case as a concrete example.
I have a bunch of experience contributing -- as a writer, manager,
teacher, fundraiser, tester, programmer, etc. -- to free software
projects, but very little in GNU projects. I put some months into GNU
Mailman in early 2015, and you can find my name in commit logs, mailing
list posts, and bug reports from back then. Then I worked on GNU
Autoconf last year, contributing fundraising and project management, but
did not post to Savannah or mailing lists. It's possible to see evidence
of those contributions by looking at blog posts, news articles, and
conference talks, but putting together that portfolio would certainly
take a bit more work than just running "git log" or similar.
I care about the future of software freedom, and the existence and
momentum of GNU Assembly makes me more interested in contributing more
to GNU projects in the future. It might be that I'm not well placed to
be a *member* of the Assembly -- if that's the case, I still want to
keep up with what you are doing, and I'll just keep up via the blog's
syndication feed.
--
Sumana Harihareswara
Changeset Consulting
https://changeset.nyc
8:13 p.m.
New subject: Preventing infiltration
Hi,
"Frank Ch. Eigler" <fche(a)redhat.com> skribis:
> yet effectively prevents the group from moving forward by
derailing
> discussions, questioning its foundations, and so forth. That’s
> probably more or less the definition of infiltration.
If you already have notions of what directions of forward movement, or
foundational interpretations should not be questioned, please post
them for clarity. If no, then how can anyone predict what you or
someone else will label "infiltration" versus legitimate discussion?
A concrete example from IRC: for a day or so, most of the discussion was
about the code of conduct. We have two foundational documents so far:
the code of conduct and the social contract. If one joins and starts
questioning these, we won’t be making progress towards the vision
spelled out by the social contract.
I hope this clarifies what I meant.
Thanks,
Ludo’.
8:20 p.m.
New subject: Preventing infiltration
Hi,
Sumana Harihareswara <sh(a)changeset.nyc> skribis:
On 4/17/21 10:03 AM, Ludovic Courtès wrote:
> We could, for example, put membership requests by non-maintainers on
> hold until we’ve set this up. It would be a bummer, because we want
> equal participation of all contributors, but that’s a simple way to
> prevent the obvious infiltration scenario discussed above.
> Another way would be to require applicants to have at least (say)
> two
> years of experience contributing to GNU in a visible way.
> Lots of questions. :-)
> Ludo’.
I am a sort of marginal case as a GNU contributor, and perhaps it
might be helpful to consider my own case as a concrete example.
I have a bunch of experience contributing -- as a writer, manager,
teacher, fundraiser, tester, programmer, etc. -- to free software
projects, but very little in GNU projects. I put some months into GNU
Mailman in early 2015, and you can find my name in commit logs,
mailing list posts, and bug reports from back then. Then I worked on
GNU Autoconf last year, contributing fundraising and project
management, but did not post to Savannah or mailing lists. It's
possible to see evidence of those contributions by looking at blog
posts, news articles, and conference talks, but putting together that
portfolio would certainly take a bit more work than just running "git
log" or similar.
It’s this diversity of contributions that I had in mind. Clearly,
commits are just one kind of contribution and commit count cannot be a
useful metric.
I care about the future of software freedom, and the existence and
momentum of GNU Assembly makes me more interested in contributing more
to GNU projects in the future. It might be that I'm not well placed to
be a *member* of the Assembly -- if that's the case, I still want to
keep up with what you are doing, and I'll just keep up via the blog's
syndication feed.
We have to figure out membership but I think someone with a track record
like yours should have a say or “members” risks being a small clique.
Ludo’.
8:27 p.m.
Hi Ben,
welcome to the assembly!
Hi, I'm Ben Pfaff. I've worked on a few GNU projects, but
the best known might be GNU PSPP, which is a free software
clone of a proprietary statistical program. I spend the time
I have available on that.
[…]
I'd appreciate being added to the webpage with appropriate
affiliations. My personal webpage is https://benpfaff.org/.
I have added you to the /people page, and recorded your
affiliation with PSPP on the /software page.
(I couldn’t find libavl on the list of GNU packages, so I haven’t
added it. Perhaps that’s a mistake, but it’s easy to add it if I
was too cautious.)
I have also contributed to Gnulib.
One worry we had when publishing the /software page was that the
impression might arise that the official GNU maintainers of the
package endorse the social contract when actually the package was
listed because of a contributor having endorsed it. Since none of
the Gnulib maintainers have as yet endorsed the social contract I
decided against recording the affiliation at this point lest we
contribute to confusion and avoidable drama.
I hope that’s all right.
--
Ricardo
5:49 p.m.
On Sat, Apr 17, 2021 at 1:27 PM Ricardo Wurmus <rekado(a)elephly.net> wrote:
> Hi, I'm Ben Pfaff. I've worked on a few GNU projects,
but
> the best known might be GNU PSPP, which is a free software
> clone of a proprietary statistical program. I spend the time
> I have available on that.
[…]
> I'd appreciate being added to the webpage with appropriate
> affiliations. My personal webpage is https://benpfaff.org/.
(I couldn’t find libavl on the list of GNU packages, so I haven’t
added it. Perhaps that’s a mistake, but it’s easy to add it if I
was too cautious.)
Well, sometimes it's under just "avl". It's on Savannah at
https://sv.gnu.org/p/avl, downloadable from https://ftp.gnu.org/gnu/avl/,
and www.gnu.org/software/avl redirects to a webpage that I
maintain.
> I have also contributed to Gnulib.
One worry we had when publishing the /software page was that the
impression might arise that the official GNU maintainers of the
package endorse the social contract when actually the package was
listed because of a contributor having endorsed it. Since none of
the Gnulib maintainers have as yet endorsed the social contract I
decided against recording the affiliation at this point lest we
contribute to confusion and avoidable drama.
I hope that’s all right.
Makes sense to me. I wouldn't want to be taken as speaking for
the lead maintainers of Gnulib.
1104
days inactive
1106
days old
14 comments
6 participants
participants (6)
-
Ben Pfaff -
Carlos O'Donell -
Frank Ch. Eigler -
Ludovic Courtès -
Ricardo Wurmus -
Sumana Harihareswara