Re: State of the GNUnion 2020

[Alexandre François Garreau]

Le dimanche 23 février 2020, 00:02:27 CET Samuel Thibault a écrit :
> Alexandre François Garreau, le sam. 22 févr. 2020 23:32:13 +0100, a 
ecrit:
> > giving a link to GNU coding standards (actually even packaged into
> > debian), for instance, would be pretty reasonable mentoring.
> 
> Sure (but also telling which piece is questioned in the contribution,
> e.g. "there are missing spaces in function calls, see the corresponding
> part of the GNU Coding Standards").

Sure, if there’s enough time (otherwise it is anyway good to read it from 
the start to the end anyway, isn’t it?)

> Providing a link may not even be
> necessary, the contributor can easily find it with a webcrawler.

GNU existed before they did.  And I dislike them.  We actually can easily 
do without, when they’re not needed.  And now they’re not needed.

> > > So the new generation will have to learn by itself? Do not be
> > > surprised
> > > if it doesn't wish to pick up the software that was produced by the
> > > previous generation, and will just rewrite everything with non-free
> > > tools etc.
> > 
> > If they do non-free, it’s not because “they are not teached”
> > appropriatedly, that’s the fantasy of the “homo homine lupus
> > est”.  If they ever do, it’s actually because they were *teached*
> > into non-free software, or tricked into it, either by bad teachers, or
> > by bad laws.
> 
> No: in many juridictions it's simply the default if you don't explicitly
> make it free.

That’s precisely why I said “tricked” and “bad laws”.

> I see my students not think that much when they put software on github,
> if I don't discuss with them. When you create a repository on github,
> it proposes to set a licence, and it happens to list essentially free
> licences (it may not be so long-term wise on github). But if you
> don't explicitly make a choice, no license is set, and thus in many
> juridictions the software is not free.

This is bad of github, but this is a bad copyright law trick (that all 
States favoring copyright (that is, imho, all of them) do), and people 
should be warned against.

So we need education… but only because work is done *against us* in the 
other direction!  Wouldn’t the State instantiate such tricky laws, in a 
world where everybody publish stuff and don’t expect it to have special —
and technically obvious to circumvent— restrictions, we wouldn’t need to 
do that.

We have to be ambitious, and look forward for a world where there’s no any 
special work to do for everything to be free.

We shouldn’t expect that a society where it is to be said to everybody 
that murder is bad and torture is bad is the end goal of improving 
society.

> > We shouldn’t be defending free software because it is good but because
> > it gives freedom, so we shouldn’t try to make it good *just* for the
> > fantasy of it to be considered as such so then people agree on free
> > software. People need to be *politically convinced*.
> 
> Sure. But if your software is unknown, it will not attract new
> contributors, and you will not have the opportunity to discuss with them
> about the politics.

It’s not with contributors that you should discuss about the politics but 
with the users.

And by “user” I nowaday actually mean the physical people you find out 
there, AFK, not the few one who will think to come to you, if they use 
your software.  It’s more like “potential user” or “people needing to do 
things with computers”.

> But anyway the matter I was discussing was not about the software being
> "good", but about welcoming contributions. Some software might be very
> good, if it is not welcoming new contributors they will just rewrite it,
> even if that'd result with a much less good software.

Okay, but that can’t be accused of being the cause of proprietary 
software.

> > So people will want to uphold freedom anyway.  Maybe fewer, and that
> > would be sad, but then we’d need to give them *political*
> > hand-holding, not technical one.
> 
> But they'll most often come from a technical door.

Personally I’ve more seen the opposite.  As most people aren’t technically 
skilled, or programmers anyway.

If we (like, actually, more: States) started to teach *everybody* about 
programming, so that most of population would try to contribute, *then* 
that question would be really urging… but it is not the case, and I’m 
really doubtful that the current discussions about welcoming or dying 
would occur if that happen.  Given a such wave of new contributors, there 
would be necessarily some that would be content with current state of 
things (so our software would grow as well anyway), or who would fork and 
either stay wrong, or technically prove they were right so to be merged 
again (but then it’s not as bad as our software would be slow and lacking 
in comparison, the effort wouldn’t be duplicated, but only, at first, 
happening somewhere else).

> We often see this kind of situation in the community-driven ISPs of
> FFDN: people often come with technical questions to help some people
> with Internet access, and they come home not only with technical
> answers, but also political aspects of why e.g. network neutrality is
> important etc.

Do people join *before* knowing about political aspects, in areas where it 
is already possible and even common to use commercial ISPs (just as it is 
already possible to use proprietary software most of the time as it stays 
the most used software)?

That’d be different if people started by knowing and being able to use only 
free software… I could see how it’d happen if the only computer available 
runs architectures on which no proprietary software run, or have computer 
so old and bad that only very old or light free software would run… but 
the “github community” you are willing to attract is very different from 
that.

Anyway, as I’ve told before, these local ISPs are way different, and a 
really good example of the opposite: they are local, thus a limited number 
of people can participate, and we can physically check their number and 
existence, so we can vote, we can avoid company-sponsored entrism, etc. so 
in the end it *can* be democratic, and as it is targetting users and not 
developers, it *should* be (I need to post a message about that).

> If we were not nice/welcoming on the technical questions, they would
> just not listen to whatever politics we'd like to talk them about in
> addition to the technical parts. Because they have no idea that these
> questions are important. We have a similar situation with free software.

My current ISP is part of FFDN.  And like others, it is exclusively set up 
in areas where internet is already available, commonly with commercial 
ISPs (but who offer fiber).  It has even a more democratic constitution than 
yours (it doesn’t have a all-powerful “collège”, but everybody have equal 
vote power on all decisions, who are meant to be transparent).

However it is ran in a quite undemocratic way by people who care only 
about internet neutrality and IPv6 availability, who don’t at all about 
software freedom, who are against concepts such as copyleft (and emacs, 
but it’s a joke).  So politically it’s pretty bad, and I’ve learnt that 
it’s people who make democracy, not texts.  That words never have any 
meaning nor even exist at all without interpreters.  That it is all about 
people, people disagreement, people ability to disagree and keeping 
disagreement and yet be able to arbitrate between them.

And then, actually because of that, I wouldn’t be able to get any 
technical help, not only because (especially sometimes when I’m tired) I’m 
too verbose, and bad at explaining, and sometimes at understanding (not 
even technical things, but at some point I become unable to understand 
unambiguously to what refer each pronoun >< and the more pronouns get 
used, the less I understand), and it creates endless misunderstandings 
such as I end with my technical problem being still there, and no more 
help than “google it, it is obvious”, but also because as we use two 
completely separate sets of software (they use only OpenBSD stuff), I would 
be unable to set up network-manager or a multihoming setup on GNU/Linux 
because they use different stuff and mine would be considered as a “newbie” 
thing and wouldn’t help someone they would, after so much useless effort, 
only consider as a troll :) so in the end you can always decide that 
someone you don’t understand or don’t agree with is a troll and whatever 
the current status, if you have enough power it’ll just look justified.

It has been proposed to do stuff with Debian, do do stuff with free 
software, but it has been refused as they “need to provide appropriate 
uptime” with things a few would master as them such as getting expensive 
ring--1 hypervisors servers (got “in the street” “from a friend”) that 
couldn’t run free-software, with things such as RAID-
i_forgot_the_number_but_it_involved_4_disks, and several hundreds of VMs 
(who bring half the money of the charity… the other half being… their own, 
used to rent a dedicated server for personal use) to all too often help 
their users set up “financially-critic applications” (proprietary or at 
least centralized SaaSS), and they don’t trust anybody to be able to 
provide such “uptime” as theirs (yet actually the uptime is bad, but then 
they bark about “0 downtime” and “almost being there”, and excusing all 
along about downtime while nobody ever complained, because it’s a charity 
anyway ><), and they don’t want anyone to touch “their baby” that they 
carefully setup… except they setted up it for others, actually, in the 
first place! but too late…

But we can’t vote to change that, or to change config, or to changing 
pricing (actually we’ve got 1 or 2 *gratis* *transit* links, but nobody 
except them knows why, apparently it’s a “favor” but we ought “not to 
abuse of it”, so we shall not lower the prices…), or responsability 
charges (they don’t want anybody to being able to tweak anything without 
giving juridical responsibility (because “downtime” and “financiary 
expectations” and “professionalism”), which, from what I’ve been 
explained, is the opposite of FDN (who aimed to let beginners tweak, and 
sometimes break, everything, but then to learn how to fix it, and FDN would 
take responsability for this process)) because otherwise since they got 
their signature on the last statuses that were given to the mayor, they 
would just go and tell nothing has changed, that they’re always in charge, 
and if we try to do would say we lied, or at least they say we couldn’t 
get it accepted…

So here, in the end, you could think it is like GNU, a bunch of expert old 
hackers who control everything because they were here from the beginning…

…except actually it’s not, pretty much the opposite:

— these weren’t there from the beginning, they just happened to arrive in 
the middle, and were enough “skilled” and “meritant” to set up a 
completely informal meritocracy (like what was argued for since several 
months), by doing the right things in the right time;
— statutorily it is democratic, so it is a proof constitutions proves 
nothing (it just takes from people who gave their name to the states, who 
reserved domain names or who are root on the right machines to do whatever 
they want if we let them, and them justify themselves);
— constitutionally, it is upholding free software, as FFDN statuses 
require to, so it shows how people can be hypocritical about free software 
(they just happen to deem GPL “proprietary” and “against freedom”, while 
setting up proprietary software and hardware that couldn’t run free 
software, even for others’ usage, is their “personal choice/freedom”);
— they justify themselves with “professionalism”, so that to block what 
they don’t like or don’t find common enough (recently jokes (usually 
defended by the fact GNU (ie. rms) is against “professionalism”) from (not 
only the abort one) were deemed as “inappropriate” by maintainers (who 
happened to be… professionals!), so the abort one was removed as an 
“extreme” exemple of it).