There are 2 types of projects, those run by an expert/s and those run by
The first type is useful for disruptive innovation. The second for the
sharing of use cases/experiences and thus improvements.
In both cases, a COC is good if the participants enforce it or just
In the first type, in the case of the expert or dictator, the COC is there
to say one thing. If you want to work with me, I want a certain standard of
In the second case, the community enforces it . It is a set of rules that
it finds important. After all, the code doesn't matter as much as the
We should't sacrifice our well being for a slightly better code. Always
prefer the people over an obnoxious good programmer.
That said, we should not supervise/check superficial manners and we should
get to the essence of things. I want to be able to call a black person a
nigger when I do not use it as an insult.
It is also up to the community to not use that power to shut some ideas
down. Quite the opposite.
Personally, I have been the target of such a misuse of power. So the fact
that I still agree in a CoC says a lot.
2016-01-06 5:48 GMT+02:00 crocket <crockabiscuit@xxxxxxxxx>:
I'm ok with your benevolent dictatorship, and I wouldn't propose
anything else for the community since it already works. Keep what
works. Experiment elsewhere.
I think the CoC you suggested will definitely work as long as bad guys
can't become maintainers. But, I don't know whether CoC is going to
help down the road.
Thus, I defer to Garret D'Amore's judgments.
By the way, I'd like to nitpick on some things for the reputation of
anarchy and polycentric law. Polycentric law is a variant of anarchy
and reduces contentions around concensus in the form of forking or
some other means. Any open source project that claims to mix anarchy
and concensus is probably neither polycentric nor anarchy but a
democracy that enforces a single set of rules upon groups that do not
agree with the rules.
Democracy left bad tastes in my mouth in the past.
As a private researcher of new political systems, I recently have been
designing a new system that lets polycentric legal system permeate
into projects, but it wouldn't be wise for anyone to try an early
theoretical effort for real projects. It might turn out to be rubbish
On Wed, Jan 6, 2016 at 11:45 AM, Garrett D'Amore <garrett@xxxxxxxxxx>
Ok, let me be clear. I’m *not* interested in changing to some otherform of
governance. Anarchy and consensus don’t work with open source projectsthat
share a code repo. I’ve seen this fail hard first hand, and I’ve beenbest
burned too badly by it to want to ever walk down that path again.
I’m the benevolent dictator — or so I perceive myself — of nanomsg. I
didn’t *ask* for this job, but its mine for now and I’m going to do the
I can at it. I don’t think shared leadership works, so I’m rejectingthat
notion. If however Martin decides he wants the job back, I’ll be happyto
give it back to him, and then he can delegate others to govern or do sosaid, if
himself. I kind of figuring I’m only standing-in for him anyway.
As you rightly point out, if you don’t like this situation, you can fork.
If Martin doesn’t like it, he can take the name back (although as I
he wants it, I’ll return the mantle of leadership back to him with mysuccessful,
compliments and best wishes). A coup attempt would probably be
mostly because if the community doesn’t want my benevolent dictatorship,more
then I’m prepared to just walk way. Believe me, nanomsg has sucked far
of my life away than I’d ever desired.document
So, I *can* institute a CoC if I so choose. In fact, let me be clear —
there is an effective one already — its pretty much the same as the
I pointed to, just implicit instead of explicit. I don’t need thedocument
to exercise power to eject someone who I feel is destructive to theproject.
I will do so if I feel that’s appropriate. I hope it never comes about,is
because although I’m a “benevolent dictator”, I do believe a light hand
best, and I’m not looking for excuses to punish or find cause forargument.
I believe most people here are adults and can operate withoutsupervision.
Supervision / moderation only becomes necessary in the rare event thatit
someone either doesn’t realize they are being an ass, or is intentionally
and consistently doing so. In the former case a private email pointing
out is usually sufficient; the latter few are toxic persons wherebanning is
appropriate, and generally doing so is noncontroversial because*everyone*
recognizes the asshat for what they are.will
I guess I’m more like nanomsg’s caring step-father than a dictator. I
do what I can, but I can always walk away if I feel that’s best foreither
myself or the community, but in the meantime its my house, my rules, andI’m
trying to do the best I can for common good.wrote:
On Tue, Jan 5, 2016 at 5:41 PM, crocket <crockabiscuit@xxxxxxxxx> wrote:
My current guess is that mixing a polycentric legal system similar to
the xeer(from somalia) and C4.1 in a project would be better at
protecting victims than a CoC imposed upon the rest by the central
authority of the official repository.
If the central authority is not corrupt, then CoC may still work well
enough, but the central authority would not be incentivized to behave
as much as now unless there are threats of forking.
The problem of the central authority occurs when a victim perceives
personal damage but the central authority doesn't consider it as a
damage. This is purely theoretical at this point, but it is a probable
On Wed, Jan 6, 2016 at 10:23 AM, Michael Powell <mwpowellhtx@xxxxxxxxx>
On Tue, Jan 5, 2016 at 8:17 PM, crocket <crockabiscuit@xxxxxxxxx>
generatingFor example, you could implement polycentric legal order by
I’mmultiple nanomsg repositories with its own judges and its own rules.
How do we merge commits from multiple repositories into one? That's
problematic. It may not work, but it was a fun thought experiment.
I use Boost as an illustration of the perceived, at any rate, molasses
that has happened as a result. Moves to Git / Github were supposed to
help with that flow, but I haven't seen it; vetted patches and such
are left sitting in their repositories without seeing the light of
day. Probably the "problem" (or opportunity) was always there, just
more pronounced now. This is just one man's opinion outside looking
in, of course.
On Wed, Jan 6, 2016 at 7:15 AM, Garrett D'Amore <garrett@xxxxxxxxxx>
Normally I try to avoid getting mired into political debates, but
thecurrently thinking of adding a standard CoC to the nanomsg source
Basically my proposal is to add this:
to the nanomsg repo. (Updated with email addresses of course).
I do have my doubts about the efficacy of CoCs, and I don’t think we
problem — but if we did we might not know it — I suspect that all
somethingcurrent contributors are male.
The concept of setting a stake in the ground for the future, and
our “community” (are we even that big yet?) as welcoming and
setting a standard tone for acceptable behavior, seems like
ought to embrace.
The cost to the project is low, at least at this point, since I
we have any of the toxicity that has plagued other open source
But the potential benefit of attracting additional contributors
it to me.
Additionally, doing this *now* is something that can be done
non-controversially (I hope). If we later have a problem and don’t
CoC, the consequences for the project may be worse (in several
That said, Martin has the trademark for nanomsg still. If he
objects I’ll shelve the proposal. If anyone else has strenuous
to this, please let me know privately, with specific rational
To be clear, there’s been no past need for any kind of enforcement
I hope such will never occur, and any kind of corrective response to
future misbehavior is something I’d like to limit except in the face
And I hope the community will hold me to the same standard of