Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 12 Mar 1997 11:19:15 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        nate@mt.sri.com (Nate Williams)
Cc:        terry@lambert.org, msmith@atrad.adelaide.edu.au, freebsd-platforms@freebsd.org
Subject:   Re: Kernel layout reshuffle (Was: Re: FAQ?)
Message-ID:  <199703121819.LAA27688@phaeton.artisoft.com>
In-Reply-To: <199703120528.WAA10606@rocky.mt.sri.com> from "Nate Williams" at Mar 11, 97 10:28:45 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> > It would be nice if the structure of the source tree did not
> > actively oppose porting to other platforms.
> 
> That has nothing to do with Richard's proposals.

Thankfully, Richard has corrected you better than I could have on
his behalf.

> Terry, I find it *really* annoying when you exaggerate peoples
> ideas and somewho shoe-horn them into your 'grand scheme' for
> unifying the world.  You've done this with my work somehow making
> the PCCARD patches the salvation to autoconfiguration,

This isn't true.  They aren't because they aren't in a larger framework;
they also aren't because they require too much opportunity for pilot
error in configuration, and they aren't because they are not a subelement
in a larger design.

Whish is not to say that they shouldn't be a subelement ina larger
design.

What they *are* is a first model for a device arrival/departure
framework in which the arrival and departure of devices will not
affect the overall ability of the machine to continue running.

Even if that model is not the final model, the goal of allowing
dynamic reconfiguration is a worthy one, and should be pursued in
other areas of the kernel, and the PCCARD stuff should be a component
in the overall framework designed to meet this goal.

> you're doing it with Richard's (which is/was keep the build tree
> only dependant on the sources in the build tree, which may in fact
> break multi-platform support),

I think you give Richard too little credit.  If I remember correctly,
Richards suggestion that he fix the make system *came out of* a
porting discussion.

I will also point out that this discussion was never sent to Richard
on a side band: he very obviously has a presence on the -platforms
list, where this discussion is taking place.


> and many others projects.

Name them; I will be happy to refute them on a case-by-case basis.


> Exagerrating the significance of things only makes you look silly, and
> angers me because it makes the developers who don't 'INTEGRATE THESE
> OBVIOUS GOOD THINGS' look petty and foolish, which in fact you are for
> implying this.

There is "obvious to Terry" and there is "obvious to the rest of the
world".  I freely admit this.  I believe I am a deep thinker, and
that believing this is not hubris.  I believe that a number of things
that have not been integrated *are* 'OBVIOUS GOOD THINGS', and I
believe that some of the developers involved in the decisions *are*
foolish for not seeing the "obvious to Terry" (not seeing the obvious
does not make one petty, however).


> Grow up and act professionally, and not like someone whose not allowed
> to play with someone else's toys and mopes in a corner,

On the contrary; I could easily start a "TerryBSD", or go over to
the OpenBSD or NetBSD camp and start integrating FreeBSD code left
and right.  To a large extent, porting FreeBSD to the Alpha and
HP3xx platforms is exactly that.  A merger by fiat.  But I *am* a
professional, and I have a good understanding of the social
structures on which I would be operating, so I have chosen the
path most likely to succeed in the long term.  To do otherwise
would *truly* make me out the childish fool you imply.


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199703121819.LAA27688>