Re: [Gnu-arch-users] Why so many version control projects?

[Zack Brown]

CC list pruned

On Sun, Sep 21, 2003 at 10:38:21PM -0700, Tom Lord wrote:
>     >> I think that what you're missing when you reach that conclusion is an
>     >> understanding of arch as separable components that can be described as
>     >> interoperability specifications and fundamental capabilities based on
>     >> top of them -- arch as generic architecture rather than specific
>     >> realization of that architecture.
> 
>     > That's a great feature of arch, but it's not the holy
>     > grail. Other projects have other ideas, and they disagree with
>     > you. It's not about ego (as you claimed), it's about believing
>     > something else.
> 
>     > You seem to have the attitude of "but my idea is the best,
>     > therefore anyone who disagrees is doing so out of ego." You see,
>     > it's *your* ego that's involved, not theirs.
> 
> Ok, that's the other part you're missing: the explanation for why
> you're making up a myth.
> 
> You're assuming that I'm reacting to disagreement.  In the case of
> svn, I'm not reacting to disagreement.  I'm reacting to refusal to
> understand/consider the issues well enough to form a disagreement, in
> the face of ample overtures laying out the basics and offering to work
> through the details.
> 
> I'm reacting to a breakdown of how civilized engineers in positions of
> social responsibility (a condition both I and the svn engineers find
> ourselves in) should behave.   I don't see a disagreement:  I see a
> breakdown of those extra-economic processes by which engineers provide
> checks and balances on the economic authorities within whose domains
> they operate.

Again: your ego is preventing you from considering the issue clearly.
You see only stubbornness and inability in the people who don't do
things the way you feel they should be done.

You think you know the truth. And that may well be the case. But until you
actually convince the relevant people, then they don't "agree" with you. The
nature of the situation is one of disagreement. Of course you feel that
they are just unwilling or unable to see the truth, and you see it that way
because of your ego. And *your* inability to see that you're doing this, is
one of the best arguments for *all* of those projects to avoid arch altogether.

Why should the Subversion, darcs, and OpenCM crowd subject themselves to
your ego, when they have working development models of their own? I don't
see a reason, *regardless* of any technical point you may have to offer.

Note: I am not saying arch is a bad project that people should avoid.
Arch (and particularly tla) is great, and likely to make a big difference
to a lot of people.  But I am saying that you have problems communicating
your ideas to others in ways that are useful to them. And those problems
are significant enough that existing version control projects would do
better to continue what they're doing, than come along with you.

>     > I agree with you. But your claim that the other projects were avoiding
>     > this conclusion out of ego, is off the wall. 
> 
> Well, please let me retract the word "ego", then.   I use it in a
> particular way.   I think it is not the same way you use it.  Please
> just take me as pointing to a problem, where engineers have failed to
> live up to the duties imposed by their capabilities:
> 
> A bifurcation of effort and resources persists that is not well
> supported by a bifurcation of the design space.   The social
> implications of deployment under such a condition are, at the very
> least, deserving of critical attention.

Again, I agree with you that having one main project would be great. But
as long as you insist on looking at it only from the technical
standpoint, you will never understand that it is your own behavior that
keeps many people from wanting to follow you. Look at ArX. Have you
changed so much since those days? Yes. But not as much as you might
think.

As long as you treat people as stupid, or as willfully avoiding the "truth" you
present to them, you'll find them resistant to your ideas. People participate
in an open source project because it's fun, or because they need the tool. If
neither condition is met, they don't participate.

>     > I think a much more likely
>     > explanation is one of the following:
> 
>     > 1) the projects are already underway, and the developers don't want to
>     > abandon their work
> 
> But why not exhibit a better effort at trying to understand its
> relationship to related work?

Because the person trying to talk to them treats them like they're
stupid.

> 
> 
>     > 2) the developers of the other projects don't sufficiently understand
>     > how to map their problems onto the arch system
> 
> That is not a hard or terribly time-consuming problem to solve.

Apparently it is.

> 
> 
>     > 3) the developers of the other projects don't have faith that mapping
>     > their problems onto the arch system will result in as good a solution as
>     > what they are currently working towards.
> 
> "faith"?

I think it would take some amount of faith for anyone to say, "OK
fellows, let's ditch this code base and start over," or, "let's go work
on this other project instead." Yes. It takes faith in the ideas of the
alternative, in this case arch. And it takes faith that the project
leader (in this case, you) will lead them well.

> 
> 
>     > Once upon a time (around when you created the changeset mailing list) 
> you
>     > were working fairly closely with Subversion people, to integrate arch at
>     > the svn back end. Where did that go?
> 
> They were not participating, they were obfuscating.  They were not
> giving "uptake" to the issues, if that means anything to you.

I see. In other words, you couldn't convince them, so now you say they
intentionally prevented agreement. You're just passing the buck.

Tom, the problem isn't with all these many people who don't want to talk
with you. The problem is that you insist on putting ego before all else.
Why can't you take some responsibility for not convincing them? You
didn't make the case well enough, apparently.

It's not like people think arch is full of bad ideas. Yes, there are objections
to various aspects. But overall, I think people respect your design. But
your way of inviting people in is often the precise equivalent of pushing
them away. Not only that, but once you've failed to make your point to them,
you then turn around and insult them, and accuse them of willfully refusing
to understand your points.

> 
> 
>     > > OpenCM explicitly does _not_ chase a grand idea: it aims at a solid,
>     > > practically studied, and academically published exploration of a
>     > > handful of narrow ideas that arose out of the needs of Eros.
> 
>     > You say tomato, I say tomahto. To them, it's a grand idea. To you, it's
>     > something that should be abandoned in favor of arch.
> 
> Stop putting words in my mouth.   That is not even close to what I
> said about the OpenCM project.

Your point is that other version control projects should switch from their own
methods to the arch method. They should abandon what they're doing in favor
of arch. I'm not putting words in your mouth. That's what you said.

Be well,
Zack

-- 
Zack Brown