Subject: Re: Exploring the limits of free software: Cygnus, and GPL
From: Craig Brozefsky <craig@red-bean.com>
Date: 26 May 1999 10:10:37 -0700

craig@jcb-sc.com writes:

> I do sometimes wonder, though, if we couldn't use some kind of
> *architecture* (or design) committee at that level.  There might
> be pressure to create/annoint one by the end of next year, if
> things like KDE v. GNOME (which I have nearly zero understanding of)
> continue to occur too often.  Not that competing *products* are a
> bad thing, but competing *platforms* with incompatible interfaces,
> each of which are intended to be "the" platform for 95% of the new
> code written for GNU (or GNU/Linux, whatever), is a great way to
> scatter our fire.

scatter the fire, scatter it to the four points of the compass.

As for organizations with a scope that extends beyond any single peice
of software, or distinct set of software tools, Debian provides a good
example.  They have a constitution, a well documented voting
procedure, a well defined social-contract, a respected and published
guideline for all licenses that they will accept into their main
distribution, and are actually producing something of immense value.

But Debian does not have positions or make recommendations on things
like KDE vs. GNOME, despite what you might have heard otherwise.
Their domain is strictly their distribution, and your software will be
in their distribution if it is packaged and its license meets their
published requirements.  Your software will not be in their
distribution if it is not packages and it does not meet their license
requirements.  This is why KDE is not in Debian main BTW, because
Debian cannot distribute it without violating the licenses of several
peices of GPL software KDE has used to build it's apps which link
against a non-GPL-compatible license (which is itself free by their
own standards).

Outside of  the assignment of copyright to a central organization to
make the protection of the entire system easier, Debian would seem to
be an incarnation of the GNU system, hence the title of their Linux
based distribution, GNU/Linux.  They are also working on a Hurd based
distribution, BTW.

There is no way to generate a comittee at that level which would be
able to deal with the issues you bring up, such as competing major
architectures, divergent protocols, and the other "inefficiences" of
unplanned and unguided labor.  I'm sure you could pretend to create
it, even give it a nice name and a bunch of members, but it would have
no real ability to resolve those issues.  

All the better IMO, because the last thing we need is some group of
wankers telling us we're not writing the right code.  I would wager
that more time and energy was spent arguing about how KDE/GNOME or
whatever are wasteful duplications of effort, than was actually
duplicated between the two.

-- 
Craig Brozefsky        <craig@red-bean.com>
Less matter, more form!      - Bruno Schulz
ignazz, I am truly korrupted by yore sinful tzourceware. -jb
The Osmonds! You are all Osmonds!! Throwing up on a freeway at dawn!!!