Re: Discussing issues

On Tue, Apr 15, 2003 at 08:41:25PM -0400, David Dawes wrote:  
> It's absurd to suggest that somone can stop someone else from doing
> exactly that.  If you want to implement a much-needed new library
> API and extension, go implement it.  For some reason, that's not
> enough, because there are those that feel a formalised structure
> is needed to give credibility to their work.

Can you re-read the last reply I sent you earlier today?
I was not trying to say what you are repeating here.

http://www.xfree86.org/pipermail/forum/2003-April/001234.html

The point is about a) coordination, and b) efficiency/effectiveness.
It is not about credibility or formalization.

> The idea that Havoc is suggesting regarding "blessed" APIs seems
> to me to be intended discourage others from deviating from the "one
> true path."  While I can understand how some may find that idea
> alluring in the name of a "unified Linux Desktop" vs Microsoft, I
> think it will stifle innovation in the longer term because you are
> still playing by their rules and not thinking 'out of the box'.

The point is about coordination between those who want to coordinate,
not about discouraging those who don't, or somehow legislating things
a priori. Again, I can only point to
http://www.xfree86.org/pipermail/forum/2003-April/001234.html

Let's look at the freedesktop.org example. GNOME, KDE, ROX, and some
assorted window managers participate in it, Enlightenment and others
don't. Great. The people that want to coordinate do so, the people
that don't don't. What is the problem? I don't get what's wrong with
that.

I understand you to be saying that you're opposed to having multiple
projects sharing resources such as web hosting etc., and that you're
opposed to projects coordinating between each other in a particular
area or in a well-advertised way. That doesn't seem logical.

The thing is that what I'm suggesting is something that already
happens every day, in projects from GNOME/KDE to Apache to Debian.  If
I was making up a utopia, I would not be able to point to empirical
examples of coordinating a big-picture development direction.

> The XFree86 vs X Consortium example shows this clearly.

You're missing an important point. You say that XFree86 wouldn't have
existed if you hadn't forked from the X consortium.

However, the reason you had to fork, and thus exist, is that the X
consortium would not adapt to meet a slightly broader interest than it
was at the time.

So the actions of XFree86 made sense in that situation, but the
actions of the X consortium *didn't* make sense, they were a mistake.

In this case, I understand you're saying you're in the role of the X
consortium.  Why are you doing what they did? Why does that make
sense?

That's what I don't understand, and why I don't think I understand
your full point of view.

In any case, could you answer the questions I asked in this mail:
http://www.xfree86.org/pipermail/forum/2003-April/001234.html

I didn't intend them to be rhetorical. I want to understand what
you're suggesting as the proper course of action, to avoid future
misunderstandings.

Havoc