Re: still more questions for the candidates
* Stephen Stafford <ssta@clothcat.org> [2004-03-07 23:42]:
> Your platform[3] contains a lot of references to your organisational
> skills and your people skills. I appreciate that last year you
> attended a lot of conferences too. You *do* mention transparency
> and accountability as well, but you're not nearly as focused on it
> as Branden is. Do you see it as less important?
No, I don't see it as less important. I think the main reason I did
not mention transparency as prominently as I could have is that I
implicitly take transparency for granted. It's nothing we should
explicitly have to mention - it should be there, otherwise something's
clearly wrong. If you look at my work, you'll see that I have always
been very transparent. I don't make empty promises saying that "I
*will* be transparent (in the future)" - you can judge my work in the
last years and see that I have always worked this way. Stemming from
this perspective, I might not have mentioned it as prominently because
it is an obvious thing for me. However, I realize it's not "an
obvious thing" for many others, and hence I consider this a very
important point, as described in section "Internal - Core Teams,
Delegates, Communication, Transparency".
> You posted several "Bits from the DPL" over the past year. I found
> them informative and interesting, however they did at times feel
> like just an itinerary of the conferences you were visiting, mostly
> outward looking. Do you plan to continue them? If so would do you
> plan to change the direction of them to make them more inward
> looking?
When I posted an "itinerary of the conferences", it was mostly with
the goal in mind to let people know where I'll be in the near future
so we can meet in person. However, I usually tried to summarize my
conference attendance afterwards and to offer more information than
just an itinerary. For example, I reported after a conference talking
to some hardware companies and getting Opteron systems. This is
clearly of interest to developers (since at that time we didn't have a
single AMD64 system to start a port).
I agree with you that some of the postings should be more inward
directed. For example, while I stayed in contact with various core
teams to see what they need, I should also have communicated more
widely that progress in various areas are being made. I will certainly
keep this in mind. And, as always, please do contact me to make such
suggestions - not only during DPL campaigns, but throughout the year.
> One of the dangers of being a successful conciliator is that you can
> become the first rather than last resort in cases of dispute. Do
> you think that this has been or will become a problem?
It has been a problem when I started as DPL, but it is no longer a
problem. In the beginning, I once made the mistake of acting as
mediator (or better: man in the middle) when it was not necessary
because the two parties were perfectly able to talk to each other. I
certainly learned from this experience, and have not repeated this
mistake. I don't plan to act as Man in the Middle, controlling all
communication going on in the project, but instead to step in when my
help is needed.
> It appears to me that the technical committee has fallen into
> disuse. Do you believe that this is due to there just not being any
> issues for them to look at, because they have lost the faith and
> respect of the developer body. because they are an anachronism and
> no longer useful, or some other reason I've not suggested?
I think the Technical Committee is mostly not working because of its
current members; this is related to the structure of the Technical
Committee and the way members are appointed. As I have argued in
another posting, I don't think that we should make use of the
Technical Committee very often. I think it is much better to discuss
issues out on the mailing lists and to come to a consensus. However,
I think it is very important to have a healthy Technical Committee
just in case it is needed.
> QA maintains an almost staggering number of packages. I am on the
> QA list, and I see almost daily the amount of organisational work
> you put into our QA effort. Do you think that we need to take a
> more proactive stance in removing unmaintained packages from the
> archive?
Packages are being removed when they are no longer maintained and not
adopted within a reasonable time. However, removing a package once
we've released with it is not the ideal solution because users won't
be supported any longer. I think we have to be more proactive, but
not by removing unmaintained packages - rather, we have to make sure
packages don't become unmaintained in the first place. There are
several approaches towards this goal. For one, I clearly see group
maintainership as one very good solution (see also
http://cyrius.com/publications/michlmayr_hill-reliance.pdf); however,
it remains to be seen when group maintainership works and when it
doesn't. Furthermore, I think we should ask *before* uploading a
package whether we really need it in the archive and whether we are
able to support it in the long run (which is unlikely if, for example,
upstream has vanished). At the moment, I don't think this is done
enough.
> What do you see as the greatest weakness of the project?
Uncontrolled growth without a clear vision or focus in scalability
(see what I wrote about core teams in my platform), and disputes about
non-issues.
> What new challenges do you plan to present to the project?
We have to address fundamental issues which have become worse with the
growth of Debian. I outlined some in my platform, but I think the
biggest challenge at the moment is communication. We seem to fight
about everything, instead of forming one community and working towards
a common goal -- to create the best free operating system out there!
> How many and which of the ideas outlined in your platform do you
> expect to be able to implement and/or work towards if you are NOT
> elected?
As always, I'll listen to people and help them find ways to get their
work done. I firmly believe that I don't get credibility just by
being DPL, but by doing the work I have been doing for years. So many
people trust me and would approach me because they know that I might
be able to help them, regardless of whether I am DPL or not. However,
through the DPL election, the project sort of appoints one person
whose task it is to keep the whole picture into account and keep
Debian running smoothly. If I was not elected, most people would
probably approach the DPL rather than me (and perhaps rightly so,
considering the other person is the DPL). So I fear I would not be as
effective achieving the goals I've outlined in my platform simply
because I'd not be fully in the loop. Furthermore, the "DPL" title
significantly helps making contact with 3rd parties, such as companies
(the 4th point in my platform), as I've learned during this year.
In summary, I could certainly implement some ideas, but most goals
outlined in http://www.debian.org/vote/2004/platforms/tbm are rather
tied to being DPL.
> Do you believe that if either Branden or Gergley are elected instead
> of you that you would be able to work with them to achieve the goals
> you outline in your platform?
If Gergley gets elected, he'll probably resign or just hand over
control to me. ;-) We have worked together for years, and it has been
a good time. As to Branden: we raise many similar points in our
platforms, but as he pointed out we have quite a different notion how
to go about them. I would first have to see how he approaches them in
order to find out how I can get involved and help out. However, since
I really want the issues I/we raised in our platforms to be addressed,
I'd try to find a way to help.
--
Martin Michlmayr
tbm@cyrius.com
Reply to: