Re: The Future of Work

From: Adrian Tymes (wingcat@pacbell.net)
Date: Tue Sep 19 2000 - 20:20:30 MDT


Eugene Leitl wrote:
> Adrian Tymes writes:
> > If a large number of the developers get dissatisfied with the decisions
> > of the manager, any of the developers can take the code base and fork
> > it. The central manager would have to screw up big time in order to
> > provoke such a reaction, but it is possible - and that, alone, can keep
> > the manager honest.
>
> Yes, but then we're talking about two different projects. Since the
> fork was provoked by a screwup on manager's part, one of them seems
> destined to crash and burn. There *are* a number of OpenSource
> screwups out there.

Granted. But the projects start from the same base, so it is possible
for them to share improvements much more than any random two projects,
almost as if they were still the same project...

> > > Anarchic software development does not work.
> >
> > Define "anarchic". No body of law ties, say, Linux developers to Linus
>
> Headless. No central arbitrator. Everybody can touch everything, and
> distribute the end result under the same name. I'm not saying it can't
> work theoretically, but lacking angels it doesn't seem to work in
> practice.

People are gonna trust a newbie stranger as much as they trust someone
they already know? To argue for that ignores human nature.

But that's not the nature of anarchy, and that's not the nature of Open
Source. That's not even really what it's being sold as, though it is
similar enough to be very confusing. (For one, if a project is forked,
then the forks have their own identifiers - imposed by the users if not
by the forker - so it's not the case that the *exact* same name
identifies both forks.)

> > Torvaldis. Linus can not hire or fire developers; he can only make
> > acceptance of their code more or less likely. (He can't even absolutely
>
> A great damn deal more or less likely. You try putting something into
> the kernel what Linus (or Alan Cox, or any of the core team) thinks is
> wrong.

In my own copy of the kernel? No sweat; having the source means that I
can do that, and owning my own hardware means that no one else can tell
me what I can do with it.

In the "official" copy of the kernel? Not right away, true. But if I
run a copy with the mod over time, and demonstrate the advantage, I can
eventually convince the core team and get it in there...whereas, with
proprietary software, I could never conduct such a demonstration, so the
only criteria with which the core team could judge would be their
initial negative speculation.

(This is assuming that it is a feature that provides value to users. If
it isn't...well, ideally, proprietary software wouldn't have "features"
that only hurt the user, but we all know that isn't the case, especially
with the VP of Sales' pet feature. With Linux, "just because Linus says
so" isn't good enough if everyone else disagrees...and there have been
cases like that, to my knowledge.)

> > reject software on his own, though his negative review can go a long way
> > towards rejecting certain code.) Promises, trust, and working code are
>
> Code that works for you might not work for me, and vice
> versa. Architectural decisions are frequently incompatible.

Code that accomplishes a certain function in a certain way does so for
everyone. Take, for example, this mailing list: the software that runs
it obviously works well enough that we can both use it...

> > all that unify the developers - and this lack of law is, technically,
> > "anarchy", in the form that some promoters of anarchy envision as their
> > utopia.
>
> I still see no evidence that there are nontrivial OpenSource projects
> which work by pure anarchy. Can you name an example?

By the meaning of "anarchy" in my dictionary, yes. By the meaning in
yours - which corresponds, in mine, to "unorganized chaos" -
unsurprisingly no.

But even if there are no laws, that doesn't mean you have to let me pick
your pocket. If you allow the concept that some people can declare
certain modules as their own, such that others can see and make their
own copies of, but not directly alter, then one could argue that the
content of the Web - the *content*, mind you, not necessarily the
languages and programs that allow it to be viewed - is, itself, one big
open source effort. I can easily view the HTML on your site, and put up
my own exactly identical. You can try various ways to get me to stop if
you want, but there's no central authority that governs everyone. (Try
getting, say, a US court order enforced in Canada.) I can copy all your
tips and tricks, and come up with my own improvements on them; you are
free to do the same right back. The audience (users) choose which is
the superior one, and spread word of it; the loser usually remains
obscure.



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:31:04 MST