Re: [HEADS-UP] systemd for F14 - the next steps

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Jul 22, 2010 at 9:46 AM, Adam Williamson <awilliam@xxxxxxxxxx> wrote:
> Can we please stick to the technical issues here? That is, how we should
> implement systemd to make the transition from upstart/sysv as painless
> as possible, and perhaps some semantic improvements to the parameters
> and command names Lennart has chosen for systemd. These really don't
> need to be emotive issues.

Mailings lists are really are good at twisting personality quirks into
weird conflicts. Its always amazing to me how differently I react to
some people when I'm in person with them versus talking to them on a
mailing list or email about the same issue...especially when there is
some conflict...doubly so when the other person isn't a native English
speaker.  It's really easy to read malice and intention to offend into
statements in this medium.  I have a sneaky suspicion that
statistically, the amount of time you interact with a person
face-to-face greatly impacts the tone of  conflict resolution is
handled in mediums when aggregated across the interactions of many
many people. Not that knowing that helps any of us be better at
avoiding it from happening to us.

I'd wager that if we were having this discussion in a room with the
very same people, I think the emotional reactions over the areas of
conflict would be much reduced and personality quirk mismatches
wouldn't cause so much static.

We also seem to get into this sort of cycle when parties have inverted
priority lists.  They may actually care about the very same things..
but the priority values don't match up. When priorities are inverted
its actually more heated than when one party doesn't care at all about
something that the other party cares about.  Areas of compromise come
in the middle ground of a priority list...so when things are
inverted.. its harder to get to that compromise even though we
actually all _care_,

In this case, we know that systemd developers care about backwards
compatibility to some degree. It does work with traditional sysinit
files out of the box. In fact is clearly a pretty high priority to
them..they didn't break the world.  We have a workable fallback
position from native systemd unit files because of that.

One of the questions _we_ as a distribution should consider is how do
we expose that fallback position to sysadmins. How easy do _we_ make
it for local sysadmins to transition their heterogeneous management
infrastructure to support transitioning to system native files..or if
need be falling back to legacy sysinit style files to better integrate
F14 into their environments.

Personally, speaking as a person _and_ a sysadmin, it would be
worthwhile to have a big freakin button somewhere that allowed me to
disable all native systemd config files and let me run sysinit style
files when the situation demands... ie crap like puppeted
configurations. It doesn't have to be the default but something that
lets me get Fedora 14 into my existing local managed environment and
start the transition process to native systemd on the timescale that
makes sense in context of that local management infrastructure.

-jef
-- 
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/devel


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]
  Powered by Linux