Re: upstart: please update to latest upstream version
]] Ian Jackson
> Josselin Mouette writes ("Re: upstart: please update to latest upstream version"):
> > Or, as has been said countless times otherwise: kFreeBSD should not
> > hinder the improvement of the Linux ports.
>
> It's not just kFreeBSD. Accepting systemd, in the current state,
> means permanently tying ourselves to the Linux kernel.
Not any more than using init scripts ties us to using
sysvinit permanently.
> The Linux kernel project has some serious and ongoing structural
> problems and I think it's very important that as a project we keep our
> options open.
If we assume that Linux falls apart and we want to migrate to a
different kernel, porting .service files is probably going to be the
least of our worries.
> I have no hesitation in utterly rejecting the idea that we should base
> our system on a project whose upstream refuse to consider portability
> as a worthwhile goal.
Do you think we should reject software where upstream doesn't want to
carry portability patches for Windows? For Mac OS 7? If not, why not,
if portability is such a worthwhile goal? Should we reject
module-init-tools because it's Linux-only?
For some pieces of software, portability is important, for others it's
less important. I'm firmly of the opinion that a much more important
aspect of systemd is the file formats. Writing something that consumes
the same file formats and runs on FreeBSD is, I assume, doable if
somebody has the interest.
> There are of course other reasons not to like systemd.
Just like there are reasons to dislike basically any piece of software.
--
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are
Reply to: