Getting Things Done

[Paul W. Frields]

On Thu, 2007-08-02 at 06:07 +0530, Rahul Sundaram wrote:
> Karsten Wade wrote:
> 
> > 
> > I notice it and don't agree with you.
> > 
> > Rahul, don't make me waste time explaining why making an open source
> > project have a good infrastructure is necessary.  One word explains it
> > -- leverage. 
> 
> Steering committee or elections for who would be in the steering 
> committee is overhead. In a large team it might yield some visible 
> benefits but it seems that kind of overhead is unnecessary for a small 
> team. I mean, if everybody who is really part of the documentation team 
> is part of the steering committee who are you steering?

You are implying that we give out FDSCo membership as a prize for people
who do something, which could not be further from the truth.  All the
folks who stood up for elections did so because they wanted to help
encourage contributions and to build ways for people to do it more
easily.  If contributors want to form teams around specific work -- as a
few have done in the past, and which strategy works very well as a rule
-- having sufficient FDSCo resources to track those teams is important.
Karsten wisely decided to err on the side of having too many chiefs in
case some of our recruitment efforts paid off.  It's unfortunate that
didn't happen, but seeing as how no one here is getting extra pay,
there's nothing gained or lost by keeping that particular wheel turning.

> Infrastructure improvements is all well and good and I have suggested 
> several myself including atleast one of what is part of the current 
> Google SoC work now so I do understand the value of that.

The value of infrastructure is not just "well and good" -- if anything,
it's greater than simply writing user docs.  I work on the user docs
because I want to make sure people get at least some minimal assistance
when they start a task.  I work on the tools to make sure that when
someone follows one of the innumerable links to join Fedora, there are
actually ways for them to do that easily.

> > Also, do you think you should dictate what people spend their time on?
> > If someone wants to work on tools instead of content, who cares?  Who
> > are you to try to make people feel bad for working on what they are
> > interested in?
> 
> I am not dictating anything to anyone and I of course have zero control 
> on what people do so that's not the point but it is important to take a 
> step back and look at what we have accomplished that matters directly 
> for end users. I will say the only things that has mattered is
> 
> 1) Installation guide
> 2) Release notes

I disagree that these are all that has mattered, since we have (or have
had) a few contributors working on a number of other documents such as
the Users Guide, a Kernel Building Guide, and others.[1]  With the
progress toward a portable live distro (i.e. boot device + user
storage), the Installation Guide may actually end up being LESS useful
over time rather than more so.  Nevertheless, I would welcome
contributors helping to improve and maintain it.  I've offered countless
times to tutor people (as necessary) in doing so.

> > Another is a very noticeable lack of content from Red Hat when we got
> > started; quite different from the rest of Fedora, who had an entire
> > distro to start with.  Heck, even Infrastructure had more hardware to
> > start than Docs had content.  Every other team, from Art to Engineering,
> > has had a paid person supporting the project, and therefore some kind of
> > budget.  Everything we have done in Docs has been boot-strapped.
> 
> I have been through that flamewars and got that point through to 
> everyone I could possible but you know what? It's high time we moved on.

Karsten's not making excuses, he's pointing out, as a direct
counterargument to you saying we're not getting enough done, that
everything we've done, we've built from scratch.  One more time, since
it was snipped out, let me refer back to your words:

"...[T]here doesn't seem to enough interest or progress being made."

I'll be happy to put our work here up next to any other subproject as
being at least as productive per capita.  Karsten laid out a very brief
list of the progress we've made.

> >   - And often those were the same people you are berating for spending
> > too much time dealing with "meetings, processes and tools".
> 
> That is indeed my point. If we focus more on end user documentation 
> could the same people have accomplished more instead of say dealing with 
> the election? Could the irc meetings be say monthly and more discussions 
> happen on list. Just maybe something to think about.

If your problem is merely that we're having elections, let me assure you
that we spent very little time on that particular issue.  If you measure
the coverage based on FDSCo minutes, list traffic, and IRC logs, you'll
find that to be the case.  I'm sure we took a few minutes of Toshio's
time to enter some data for the elections -- not to minimize at all his
assistance, for which we are, of course, grateful -- but it wasn't much
in the grand (or even daily) scheme, frankly, since the app was already
built.

I would refer you to http://fedoraproject.org/wiki/DefiningProjects
which requires that projects have a governance model.  Although the
specification does not require {s,}elections, the idea of elections was
discussed many times over the last few years and the FDP community (not
just two or three of us) agreed it was the way to go.  Number of members
in FDSCo are not a reflection on, nor a result of, how much work is or
isn't getting done.  I think our committee situation can easily be
compared to FESCo + Packaging, for instance, where there is a very large
total committee membership, and all of the people involved there are up
to their elbows in the real work too.  The fact that the most active
FDP'ers are also interested in running for FDSCo should come as no
surprise, although threads like this make it somewhat more difficult to
remember why.

The one point of agreement I have with you here is to do more on-list.
That's why I've made that point (and not by myself) repeatedly in
meetings and on IRC, and asked people to move discussions here.  I'll
say it again: USE THE LIST!  (Er, on-topic, please.)

> Everyone who is active is doing a good job. No doubt about that.

Golly, thanks, but then why the "there doesn't seem to enough interest
or progress being made" comment?  You can't drop bombs like that, Rahul,
and not expect to ruffle feathers among those who are shouldering a lot
of work.  And since I split my time fairly evenly between tools work and
end-user docs work -- including both the docs you mentioned, the IG and
the RelNotes -- I feel perfectly justified raising my hackles a bit.

>  The question is only on whether we are having the right priorities here. My 
> suggestions are going to be simple

> 1) Dissolve the steering committee and associated elections etc
> 
> 2) Try and reduce the barriers to the team itself so that we can spend 
> more time producing solid documentation for end users instead of dealing 
> with overhead.

With this sort of thinking on the development side, we wouldn't have a
Fedora platform that anyone could custom-build to their liking.
Furthermore, it seems like you have a very skewed idea of what our
"overhead" is.  I haven't heard anyone complain about lost time in a
FDSCo meeting yet, but we have the ability to self-modulate if that
occurs, just as we have in the past.

There is no barrier to team membership beyond what is required for
overall Fedora Project membership -- you introduce yourself, roll up
your sleeves, and get to work.  That's all we ask.  You only have to be
able to write passably in a chosen language to work here.  If that
language isn't English, you may need to scare up a translator so the
monoglots among us can be of more assistance.  We offer to help with
everything else, and can teach many skills, including better writing, if
and as the contributor desires.

Our priorities are to create not just documentation but also a toolchain
that allows any free community using Fedora (or other platforms with
some common tools, for that matter) to build documentation, in the same
way that the developers are providing a platform for people to build
functional code.  Our mission statement says exactly as much.[2]  

You are simply beating the wrong bushes here.  We have dedicated time in
the past to making one-on-one contact with big rosters of community
members who had introduced themselves and then immediately gone quiet,
in the hopes that they would be convinced to help out with small tasks.
The returns from even this personal outreach initiative were quite
meager -- although a few of those who responded have stuck with us, for
which we are grateful and continue to work with them and be responsive
to their contributions.

Instead of trying to reset our priorities, why not spend that energy
finding volunteers?  Technical skill is not as important as writing
skill; we can teach the former much easier than the latter, and in so
doing, find additional ways to improve our Documentation Guide[3] -- oh
wait, did I mention I've been doing quite a bit of work on that too? --
so that other contributors can learn how to participate more easily.

* * *
[1] http://fedoraproject.org/wiki/Docs/Drafts 
[2] http://fedoraproject.org/wiki/DocsProject 
[3] http://docs.fedoraproject.org/documentation-guide/ 

-- 
Paul W. Frields, RHCE                          http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
      Fedora Project:  http://fedoraproject.org/wiki/PaulWFrields
  irc.freenode.net: stickster @ #fedora-docs, #fedora-devel, #fredlug
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://listman.redhat.com/archives/fedora-docs-list/attachments/20070801/76c26e5e/attachment.sig>