[aklug] Re: systemd - was Re: multiple distros coordinate to establish /run directory

From: Arthur Corliss <acorliss@nevaeh-linux.org>
Date: Wed Apr 06 2011 - 17:58:50 AKDT

On Tue, 5 Apr 2011, barsalou wrote:

> Nothing to disagree with here either. Maybe that is what the silence you are
> getting is about....they all agree with you! :)

:-) I seriously doubt it.

> Some of the commentary isn't well thought out and many of the people make
> conjecture and guesses that could be right on or completely off.

I haven't read much commentary on the matter. I read an admission from the
author about separate /usr disclaimers, looked at the web site and the
source tarball, and came to my own conclusions.

> I propose that instead of arguing, or trying to logic our way into a
> particular position, we actually work to make it fail with simple specific
> purposeful problems. It should be easy to do if it is 'overly complex and
> buggy blob'.
>
> Something demonstrable (<-note the sorta play on words) is much more
> effective than arguing one point or another with logic.
>
> The other thing that needs to be brought to light is that the entirety of
> that conversation was many internet years ago.
>
> My main point was to ensure that I show proper appreciate to Jim for making
> my life just a little easier by pointing me to what might be considered close
> to the beginning of implementing systemd distribution wide.
>
> Who among you are willing to actively work toward breaking systemd to prove
> its inabilities?

I don't think you need to go there, Mike. The fact that they're designing
the software from a perspective that willingly discards UNIX design
philosophies is all you need to know. The question isn't "is it biting us
in the arse *now*", it's will that design methodology make it virtually
inevitable that it will? As I said, UNIX principles exist for a reason.
Those who ignore history will be doomed to repeat it.

And let's not forget that they're rejecting not one UNIX principle, but two:
the fact that they aren't explicitly designing software that bootstraps your
system to work reliably and robustly (by their own admission) *without* /usr
on / is reprehensible. They could have also had the "init" process launch
their pseudo-xinetd, service/mount monitoring, etc., separately would have
been an acceptable choice. Make systemd a *collection* of UNIX-like
separate components, rather than a monolithic blob.

But, they didn't. And I'm not about to subject any of *my* systems to crap
like that.

         --Arthur Corliss
           Live Free or Die
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
Received on Wed Apr 6 17:59:08 2011

This archive was generated by hypermail 2.1.8 : Wed Apr 06 2011 - 17:59:08 AKDT