Re: Questions To Answer

[jg pa dec com (Jim Gettys)]

> Sender: mjs@pythagoras.eazel.com
> From: Maciej Stachowiak <mjs@eazel.com>
> Date: 12 Jul 2000 14:06:29 -0700
> To: Alan Cox <alan@lxorguk.ukuu.org.uk>
> Cc: jg@pa.dec.com (Jim Gettys), frank@collab.net (Frank Hecker),
>         foundation-list@gnome.org
> Subject: Re: Questions To Answer
> -----
> Alan Cox <alan@lxorguk.ukuu.org.uk> writes:
> 
> >
> > The gnome code is irrelevant. The gnome foundation, the LSB, the vendors
> need
> > APIs not code. APIs are protocol specifications, APIs are largely
> independant.
> > Code is just implementation detail on the interface.
> 
> Standards may be important in the future, but right now we don't have
> any, and have no plans to create any formally. What we have now is a
> bunch of hackers who want to figure out how to organize their project
> better.

Whether you have formal standardization, or informal is relatively irrelevant
(right now).  Once you have a significant pile of applications built on
your interfaces, stability becomes a really major concern.  You are very
rapidly reaching the point where it will become infeasible to remove existing
interfaces (without careful transition strategies).

> 
> Maybe standardization will be useful later, but that hardly makes the
> GNOME Foundation primarily a standards body. In fact, if there are
> standards, it would be better if they were promulgated by some
> independent entity, e.g. LSB.

This is clearly a solution on the formal standards side.

> 
> I don't understand why people keep bringing up this standards thing
> now. More formally organizing the project is not going to make people
> magically switch from writing code to writing standards, no matter how
> desirable that would be (and personally I don't think it would - the
> Windows API has been very successful without any of it ever being
> formally standardized). What we need is a structure to organize what
> we do NOW, not to handle someone's vision of what we really ought to
> be doing.
> 

I bring it up primarily since I believe the governance of the IETF has
alot to offer Gnome, not that it should be adopted lock, stock, and barrel,
including all the standards processes.  This level of standards process
might become reasonable some day (and as you say, the LSB may be the right
venue:) but you are in fact attempting to set a de-facto standard whether
you acknowledge it or not.

Somehow, the IETF governs itself, with about the most transparent and
open process imaginable, despite allowing participation of one and all.
This is the area where one can learn greatly from the IETF.
					- Jim


--
Jim Gettys
Technology and Corporate Development
Compaq Computer Corporation
jg@pa.dec.com