On 22.02.2014 21:21, Stroller wrote:

On Sat, 22 February 2014, at 10:38 am, Yuri K. Shatroff
<yks-...@yandex.ru> wrote:

On 22.02.2014 11:40, Mark David Dumlao wrote:
[ ... ] Even as the complex beast it has become systemd is still
simpler than the alternative of having abominations of unreliable
shell scripts checking to see which version of grep and sed is
used to split the command line, or whether the system uses
tempfile or mktemp, or depending on perl.

Well, simpler yeah, supporting only one kernel of specific versions
is always simpler than trying to support everything from SunOS to
NetBSD. This way, if the kernel supported only e.g. Intel
IvyBridge+ with one chipset family, one graphics (VESA) and so on,
it would have been incredibly simpler.

I’m doing a (free) operating system (just a hobby, won’t be big and
professional like gnu) for 386(486) AT clones. … PS. Yes – it’s free
of any minix code, and it has a multi-threaded fs.  It is NOT
protable (uses 386 task switching etc), and it probably never will
support anything other than AT-harddisks, as that’s all I have :-(.

Good luck!

Linux did indeed once support only one CPU family and one or two
hard-drives, and the reason that it now supports more is that people
dug into the code, submitted patches and got it fixed.

Had all the original Linux developers spent their time on the
comp.os.minix list, complaining "oh, those splitters, they're trying
to fragment the Minix community" and "this Linus guy should be
putting his effort into improving the Minix kernel", where would we
be today?

Actually I don't get what you are arguing [against].

It's almost hilarious the volume of traffic expended here on this
subject, especially that by the naysayers. When I first learned of
systemd I did not feel favourably towards it, but those ranting
against it have only given Canek a platform to convince me.

I partially agree. In an emotional discussion the most probable winner (as seen from outside) is the calm one.
But being calm doesn't refute all technical and `political` stuff.
I personally was going to try systemd about a week ago when the discussion started. Now I'm quite convinced not to do this in the near future. No calm arguments of systemd's supporters, such as the complexity of shell scripts, the simplicity of systemd compared to the Kernel, the ease of use of journald tools, the shitload of troubles of configuring syslog, the replacement for all network setup tools, the good intents of Red Hat, etc etc, didn't convince me. Emotions pass, results remain.

And whilst I'm still of two minds on which init system I'd ideally
prefer, I am not under any delusions that I can influence the
developers of the Gentoo distro or those of the Linux kernel (who
AIUI are adding kdbus to support systemd), either by ranting about it
here or otherwise.

No delusions, there will always be an alternative.
Nobody actually has disagreed yet with my words that in a couple of years systemd is going to dominate "90%" (meaning the majority of) linux distros. But "10%" hopefully will remain without it. Anyway since systemd is not intending to support any other kernels, we'll probably see other OS or stuff like Debian/kFreeBSD develop more intensively. Yet, of course, these alternatives will necessarily be poorer supported and one will have to take effort to migrate - to either the distro he used, but the version with systemd, or a different distro/OS.

The amount of energy spent on this, you could have established a fork
and written code by now - if y'all really want to prove your point,
that's the way to do it.

What point?
I personally am terribly satisfied with the SysV init and shell scripts so what am I to fork and write? What a fork to establish? A fork of debian, to maintain it w/o systemd? Let that be done by debianners maybe, if they so desire.

As for `ranting`, I do see a point in such talks (until these get personal), as I learn many new things (both from the posts and while trying to prove/refute the points) and I always try to ask a concrete question and answer a concrete question. Note: I do repeat *I* here because you answered *my* post.

In any case, no offense, your reply is a rant, too.

--
Best wishes,
Yuri K. Shatroff

Reply via email to