Re: Perfect Jessie is something like this...
Scott Ferguson wrote:
On 05/11/14 12:28, Miles Fidelman wrote:
Didier 'OdyX' Raboud wrote:
Le mardi, 4 novembre 2014, 17.13:05 Miles Fidelman a écrit :
Personally, the more this drags on, the more I'm convinced that those
of us who deploy and manage servers would really benefit from a fork
that retains the "flavor" and philosophy of pre-systemd (and maybe
pre-udev) Debian.
As JS posted - it's likely that the majority of us sit somewhere in
between the two extremes (Conservative Fundamentalists and Cutting Edge
Advocates).
Just.
Do.
It.
It might even be successful! We'll be in a better world with a new
distro out there satisfying needs not satisfied anymore by Debian!
I'd have to agree with another poster - it's logistically more likely to
succeed[*1] (and be less divisive) if it was a Debian derivative.
Isn't that a rather fine distinction? How is a fork all that different
than a derivative?
[*1] I haven't noticed anyone volunteering to do the work - as opposed
to "organizing" others.
Well, that's the point of the discussion - to suss out who's willing and
able to do the work.
By all means, transform this energy into action and create the free
software environment that you need. It will be more work than whining on
lists, sure, but that's the only way to make the change fitting your
hopes eventually happen.
Before embarking on a major project,
i.e. Your project - to remove systemd from Debian?
Ill-defined as yet, and may not be "my project" - might be someone
else's project - initial goal is to see who's interested, and who's
initiating what.
General outline from my point of view of an ideal distro:
- Debian like/derived (Wheezy as baseline, probably; possibly Squeeze as
jumping off point)
- Apt packaging
- focused on server deployments (normal operation intended to be headless)
- particular focus on cluster/cloud support
- support multiple init systems, sysvinit as default and lowest common
denominator
- support maximum portability of code across Linux, BSD (including Mac),
illumos, unikernals
Alternative model that is somewhat intriguing:
- build up from LFS
- Nix or GUIX based packaging
And still evaluating whether an existing distribution/platform is
sufficient:
- who knows, maybe Debian will come to its senses and provide true
multi-init system support, from install onwards, with serious testing
for more than systemd
- maybe the GNU folks will really get to a real release; seems like the
systemd situation has lit some fire under the project
- SUSE seems to be moving down the systemd path, but with less
aggregation (e.g., systemd w/o journald)
- Gentoo/Funtoo
- Slackware
- BSDs
- illumos (particularly SmartOS and Dilos)
I figure I have a year, maybe two, before having to migrate off Wheezy -
so no real hurry.
Good advice deserves the same in return:-
If you want to play general, you'd do better recruiting volunteer
soldiers if you didn't insist on staying behind the backlines.
<snipped>
So, in other words, bugger off.
Please try and be pleasant - no one has said that, you clearly have the
wrong end of the offence.
Aimed specifically at the latest round of systemd fan-boys who can't
tolerate anything but 100% acceptance of the program. Getting a little
tired of it - this was the polite version of what I kept from typing.
Sadly, it's your constant haranguing, unsubstantiated claims, and
character attacks that are the offence.
What... pointing to bug reports and published writings by Poettering are
now unsubstantiated claims and character attacks? Since when?
Miles Fidelman
--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra
Reply to: