Re: Contribution of Sun documentation team to the GDP.



Hi Patrick,
Here're my 2c on what I think you're missing.

Today at 14:55, Patrick Costello wrote:

>
> This is completely spurious. I thought I'd dealt with the "closed doors"
> issue. I am not suggesting "putting back" "complete work" that has been
> developed behind "closed doors". I am suggesting "putting back" works in
> progress at sensible review stages. These review stages are well-defined
> and regular. See the GDSG. Anything else will lead to confusion. And
> spurious bug reports.

[snip]

>
> I have run out of steam on this "closed doors" topic. I really have. Does
> anyone else want to say something, one way or the other? 

You seem not to buy in the free software development model.  The idea
behind having *current* (however flakey may it be) and *latest*
documentation available to testers, is to reach potential
contributors.  If I see a FIXME, I might offer to help there and
write a chapter or two.  If, OTOH, I'm not aware of what is currently
in the works, what might have changed, and what should I work on, or
what I might work on and will later be lost because you decided to
work on that yourself.  This is the core idea behind success of free
and open source software (besides political issues).

The collaboration model in free software world is an open one.
Whoever feels competent to step in is allowed to step in, and we let
them make most of it but giving them access to the latest and greatest
stuff, so their contributions don't get lost.

So, I think your idea of "authors" is what conflicts the idea of
"author" in free software: I'd prefer to use the term "maintainer"
instead, which implies the duty one holds when working on a
document.  Maintainers' responsibility is to maintain progress of the
document, to assign duties, and to help others help him.


And while there always are spurious bug reports, community testing
from CVS will likely know not to bug about FIXMEs or such.  And if
you actually *use* CVS to handle work (like many free software
developers and contributors do), then everyone else knows what FIXMEs
are done, and which are not, so the risk of doing duplicate work is
minimal.

CVS is *work area*, it's the place where unfinished stuff *will* show
up.  Gnome currently makes only one guarantee about stuff from CVS:
CVS source will build (well, that fails sometimes, but that's what
we're trying to do).  This means that people are used to some FIXMEs
and unfinished features, as long as you don't break the build,
everything is fine during development phase.

It's just that you don't seem to have enough trust in the free
software development most of us are otherwise used to (and even prefer
it over any other model).  If it were so earlier, perhaps there would
be a natural person to inherit Desktop Guide maintainer duties.  Just
like maintainership moves from one person to the other with software:
the second best contributor is the natural choice, but with all
persons being from Sun, we have no such choice.

Cheers,
Danilo



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