On Wednesday 7. December 2016 12.56.06 Luke Kenneth Casson Leighton wrote:
not sure i fully understand what you're saying, but i'm aware that devuan is supporting a huge alternative range of init systems: the only one they *don't* support is... systemd!
now, it may surprise you to learn that i've spoken to them and pointed out to them that if they want to not appear to be hypocritical (i.e. directly at odds with their stated goal of being "inclusive"), they really, *really* need to include systemd as one of the options.
however, because they've gone the "reacting against" polarisation route, which is just as equally bad as the forced-adoption route done by pretty much everyone else, there's still a lot of bad blood that needs to be healed first before anything like that can even *remotely* be considered.
I can understand where the Devuan people are coming from, though. It is a reaction against a form of change that they feel undermines choice and, for many, causes work for work's sake.
Consequently, if they are to resist demands by people to change their code to work with the systemd ecosystem (because it is more convenient for the people making the demands, for instance), and since systemd as an ecosystem itself rather seems to make demands on others but not yield to any itself, then the most effective strategy is to eliminate all parts of that ecosystem from the distribution and not give people the excuse to foist the various systemd technologies on others.
I actually don't have a strong position on systemd, but I do note that it does cause extra work because I recently had to guide someone through a Debian Jessie installation on an embedded system, and without them going through the effort of building a newer kernel (with all the accompanying quality assurance to see if a new kernel works as well as the old one, once they set up cross- compilation toolchains, of course), I had to find out how to switch out systemd because the system will refuse to boot with it enabled, thanks to systemd's additional demands on various kernel features.
Fortunately, Debian does still support sysvinit, and enough time has passed that it is possible to multistrap and just specify sysvinit-core and have everything realise that you want sysvinit and not systemd-initd (or whatever they've called it) as the init system, but I gather that various hacks were previously needed to persuade apt and other things of such intent. In contrast to claims of choice, it really doesn't inspire confidence that warnings appear about using other init systems, however, nor indeed does the apparent delay in getting the tooling up to speed with such choices.
(I'm a bit aghast at the need to have libraries lying around to test for things, especially things that aren't there. I remember bizarre suggestions in ancient discussions about opening graphical programs in a generic fashion on Free Software desktops, where the suggested interface to such capabilities was to dynamically load such shared libraries and then call functions in them, clearly optimised for the random C-only developer who thinks that raw performance must take a back seat every time and that a request to open a Web browser must be done in as few cycles as possible, even though that request is a miniscule portion of the time to actually fulfil such a request. Eventually, common sense prevailed and xdg-open - a *program*, like one would expect in the Unix tradition - was born.)
[...]
y'know... the current hypothesis i'm floating in my head is that the full-time paid-up software libre projects are running at such a faster pace than the volunteer-driven ones that the full-time paid-up developers completely swamp and overwhelm the volunteer-driven ones.
I've had that impression for a while now. Indeed, the phenomenon of the well- resourced organisation dominating or capturing a Free Software project is long understood: you have that kind of thing with stuff like WebKit, for instance. Combine that with permissive licensing advocacy where the only thing encouraging cooperation with the "upstream" of a permissively-licensed project is if upstream is well-resourced, and you have a mechanism for a kind of pilfering of the commons.
And then there's this cultivation of the noble "open source" developer who gives their work away (permissively licensed, of course) to curry favour with their corporate betters. Either that or the software they write is a purely hobbyist endeavour to "scratch their own itch" or whatever. It all adds up to putting the individual in their place in some kind of neoliberal narrative posing as altruism.
Paul
P.S. Where Devuan might be interesting, regardless of the systemd aspect, is the way it has needed to find ways of filtering Debian. Such activities are also done by FSF-compliant distributions and there may be some benefit in those distributions evaluating the Devuan tools for their own purposes.