Re: Discussing issues

On Wed, Apr 16, 2003 at 09:03:29PM -0400, David Dawes wrote: 
> So what is changing?  If the way we operate isn't changing, what is?
> 
> I thought the point was improved communication.  Is that wrong?
> I'd like to understand what you are really aiming for.  

What I'm aiming for is: an X platform/spec and implementation that is
moving forward with a good effort at the speed and quality that it
could be, that is living up to potential in terms of the overall
benefit it could provide to "humanity", "people", "users", "free
software", or whatever general good you want to choose.

Improved communication is part of that, absolutely.  The word I've
been using to describe the overall aim though is "scalable" - ability
to harness the efforts of hundreds of individuals, dozens of
companies. There are lots of elements that can go into that, and the
best way I know to understand how to do it is to examine some of the
existing projects out there.

> The bottom line is that XFree86 continue to do the things that it wants
> to do, as an independent organisation of individuals.  We don't want to
> be forced to fit into someone else's grand plan.

If I could somehow control/demand/impose/force you to do what I
thought was best, that might be much easier than sending a lot of
email. ;-) I can't.

I'm not forcing you to do anything. I'm arguing that there are things
that are really useful to do, that you could do. That's it.

The main reason I'm persistent is that I think you're still reacting
to fears that from my perspective (where I know my own motivations,
and what I mean, and can guess at those of others) seem unwarranted.

Sometimes you seem to be against "demands" or "forcing," and I just do
not intend that at all. So I hope we won't fail to work together for
that reason.

Sometimes you seem to be worried about losing control, and I firmly
believe that control should be in the hands of contributors. So I hope
we won't fail to work together for that reason.

Sometimes you seem to be worried about "vested interests"; honestly,
I'm not sure what you mean by that. Some people have said to me that
you mean Linux vs. BSD, and some have said you mean corporate
influence. Based just on emails to this list, I'd assume you mean
Keith personally. I have no idea whether this is accurate. In any
case, I am not asking for control by any vested interest, and I don't
know what the concern is here, so I hope we won't fail to work
together for this reason.

You've seen my opposition to being able to buy X.org control, and 
that is because I'm not a fan of corporate control.

>  If you want improved coordination and communication with XFree86,
> then attempting to impose any additional structures on us, however
> "light weight" you might claim them to be, is counter-productive.

I'm all for improved coordination/communication with XFree86, but
that's a means or subgoal. The larger-scope goal is an X organization
that's scalable.

The even larger scope goal, in official market-speak, might be
something like: to find, create, encourage, advocate, implement
organizational and technical improvements that enhance the overall
progress of free software as a competitive workstation solution[1].

Market-speak enough for you? ;-) It's accurate though.

I see lots of possible improvements to what it looks like we're going
to end up with here if nobody does anything. I'd like to see us fix as
many things as people are willing to fix. When they aren't willing,
I'd like to understand why not so I can try to address or work around
those issues over time.

[1] (I realize X has uses outside workstations, such as embedded, and
those are also interesting IMO. I don't think they're in conflict with
workstation goals in any significant way, though.)

> And of course the rhetorical question always comes up, if it's so
> light weight why have it?

It's a solid step forward that gets people sharing trivial
nontechnical things, and symbolically puts them in the same boat.  It
means the barrier to adding modules is low, the barrier to making
arbitrary subset composite releases is low, and the barrier for adding
new contributors to an existing module is low. It begins to build a
full-scale critical mass of development momentum.

Yes one might want to do more eventually, in fact I'd advocate it, but
shared infrastructure is a good solid step forward that would let
people put a toe in the water without too much risk, and then evaluate
the next step later.

But it's just a strawman. There are lots of possible courses of
action. I was just trying to suggest something, anything, as a first
step.

> The "agenda" from "your group" is too diffuse for any of us to really
> figure out what you are thinking concerning the most important
> issues.

What is "my group"? I'm in the dark. I talk to a lot of different
people, and they don't all agree.

If I was speaking for a group, I'd say "we the undersigned" or something.

> It is your group making demands from XFree86, not the other way around.
> And I have yet to hear what benefit this gives XFree86.  I can see how
> the umbrella benefits, but what about us? I represent XFree86 and only
> XFree86, not the "community", not anyone else. My interest lies solely
> with what is best for XFree86 and those who have invested the most effort
> in XFree86 over the last eleven years.  I have no other agenda than that
> as leader of XFree86.

When you say "benefits XFree86", by XFree86 do you mean:
 a) the contributors to XFree86
or
 b) the technology and community interest that XFree86 de facto owns

The benefits to b) that I see are to speed up development, increase
contribution, improve quality, and increase the adoption of that
technology, thus promoting free software platforms in general (and
UNIX-like platforms overall as a side effect), and bringing the
technical, financial, and other benefits of these platforms to more
users that currently can't use them due to technical limitations.

In short, make the XFree86 code better and get it to more people.

The benefits to a) are mostly warm fuzzies about doing the right
thing, and the benefit of getting to use the improved code. I would
never pretend otherwise.

There are possible personal benefits in terms of employment, etc. I
suppose, just because I know a lot of people who found jobs as a
result of free software work. Maybe those opportunities will increase
as X-based platforms gain marketshare. But I don't know many people
who are in it for that reason. I guess you could enjoy seeing the
project move faster and better, as well. I don't know what your
motivations are so I don't know what you would consider a benefit.

However I don't see significant *cost* to a) either, though I can
understand fearing various costs, I don't think those fears would
really turn out to be justified in retrospect.

What *I* fear is getting squished into irrelevance by proprietary
interests while we goof around, leaving our platform about as
interesting as AmigaOS. That's why I want to move as quickly and
effectively as we possibly can on all fronts. The X platform is only
one piece of the big picture.

> BTW, if there is to be an umbrella, I personally like the "umbrella"
> under which you said the 3 BSDs sit.  It's news to me that anything
> formal exists, so this kind of notional umbrella which is totally unlike
> anything you've described sounds good to me!  By that analogy we're all
> under a notional "X" umbrella right now anyway :-) which sounds fine
> to me.

If I described BSD as an umbrella, I meant <insert some BSD here>, not
<insert all 3 BSDs here>. I only mentioned *BSD on the basis of
comments by others that they have some sort of process for becoming a
committer and so forth. I have no personal familiarity with *BSD and
don't even know how the 3 are different from each other in any detail.

> As for you Havoc, who do you speak for and what can you deliver? 

At the moment, I speak for myself. My mails are intended to carry
weight on the basis of the arguments made, not on the basis of 
who is writing the mail.

If it would advance this discussion for me to speak for a group I'm
involved with, or if a group I'm involved with could deliver something
to help XFree86 be part of the goals I've given, just say so. I am
willing to go back and drive the appropriate process to get you a
response, to the best of my ability.

Havoc