Systemd Discussions--The Good Parts
I've tried to slow down my rate of posting both because I've said what
it was useful for me to say and because after the most recent IETF
meeting I've been taking a vacation in Hawaii, meditating floating in
the ocean and living in the moment laughing with joy as the salt spray
soaks my body.
I'll be flying back to the continental US shortly after the GR results
are announced and it will likely be 36 hours after the announcement that
I'll be done with flights, home, caught on on $day_job and back to
However, I do have one thing it might be useful to say now.
There have been some really amazing moments in this whole systemd
discussion. There have been moments where I've been really proud to be
part of debian and reminded that this is why I love this community; this
is why I'm here. Sadly, there have been other moments too with more
The first was reading
Many of us recognize that bug number, but that particular message was
Ian's message "On Diversity." In that message he talks about how he
wants Debian to be a place where each of us can come to work on our
priorities and have an opportunity to try and succeed at what we believe
as important. He wants Debian to be a community where we can be trying
different approaches and where people have an opportunity to succeed
even when others in the project don't (yet) see the value in your goals.
I think we've all benefited from this. We had something we wanted to
accomplish, and it wasn't clear we'd succeed, or sometimes even whether
it was a good idea. However Debian was a place where we could try and
see if it worked out. If we attracted other like minded folks and the
idea proved good, we succeeded. If not, perhaps the idea floated into
I recently re-read this as I was writing a blog post about this
discussion. I was jumping around my hotel room, filled with the joy of
that vision. i think Ian's vision in that post is similar to what Joey
talks about when he talks about the importance of iterating on decisions
and refining things. I think it's similar to what Russ talks about when
he points to Joey's message and talks about finding what's fun to work
on and doing that.
I think it's wonderful that even when people disagree with the approach
very strongly, there can be so much that they do agree on. Joey's
dislike of process and governance is very different than Ian's formal
attention to making sure he correctly amends and then accepts/rejects
his amendments to work within the formalism of our constitution.
However, I would be unsurprised if they both have valued Debian as a
place where people can work on what's important to them.
Another Yes! moment was reading Russ's mail about the challenges of the
TC decision--the one where he pointed out what it was like to make
decisions you cared about with people you respected when there was a lot
of emotional connection to the decision. You know, the one that got
cited everywhere and that was probably a significant part of why we all
rewarded Russ at debconf. It's great that Debian has folks like that,
and I aspire to meet the level of compassion, empathy and commitment
Russ showed in that post.
When I got to the hack lab at Debconf this year, I met Josh Triplett for
the first time. He was running around talking about the joys of
systemd. It's safe to say that Josh and I have different values
surrounding gentle, phased transitions. Josh seems to value having one
simple way to do things. I was starting to wonder if "O, hey is that
what the annoying systemd folks are like...I see why some people are
frustrated." However, I wasn't sure; Josh was talking to some other
folks who shared similar values. Fortunately later in the conference I
actually got a chance to interact with Josh. We went to dinner, where
Josh was trying to work with the ifupdown2 proponents to understand how
their technology differed from/interacted with networkd. I was pleased
that he cared about understanding others use cases and cared about
helping explain the value of his proposed solution. On the walk back
to the dorms, I talked to him about concerns resulting from some of his
comments about kdbus. He had reasonable answers for all my concerns
including discussions of HURD and KFreeBSD. He cared about the points I
brought up and was willing to revise his approach when he ran into
trouble. It's great to brainstorm with folks like that. You may not
agree but the result will be stronger than either of you could get
Finally, there's the discussion between Josh and Ian in the bug about
the libpam-systemd dependency. It was nice to see Josh and Ian working
together to refine that decision to be accurate, and to be more clear.
I think to move forward we'll need to see more of that cooperation
between people who disagree strongly.
I'm probably missing some other moments in all this where the community
really showed its depth and strength.
I hope for more of that in the future.
We are amazing when we choose that path.