Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 1 Jul 1999 19:00:33 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        jesus.monroy@usa.net (Jesus Monroy)
Cc:        dwilde1@thuntek.net, freebsd-advocacy@FreeBSD.ORG
Subject:   Re: [Re: [Re: [Re: [FreeBSD Man Pages]]]]
Message-ID:  <199907011900.MAA09564@usr06.primenet.com>
In-Reply-To: <19990701085309.4697.qmail@nw176.netaddress.usa.net> from "Jesus Monroy" at Jul 1, 99 01:53:09 am

next in thread | previous in thread | raw e-mail | index | archive | help
> > I don't consider what you called a patch worthy of the name.
>
>     I have not written a patch to date.
>     The early PatchKit should have been called
>     "The Greate Pumpking Packet Kit", but it missed it's calling. :-(
> 
>     As for the current patch system it seems to be faring
>     no better.


Not to tempt you out of your self imposed seven day email celebacy,
but...

The original patchkit was done the way it was done because Bill
Jolitz did not expose a source repository of any kind.

I realize that the strictures of the patchkit were onerous, but
they derived from a human being (me) being the equivalent of
a CVS merge.

The ordering dependencies only triggered on overlapping patches,
but when they triggered, there was a single order-of-application
mutex that triggered (gated through a human -- again, me) to
ensure that the patches did not fail to apply.


Yes, I take full responsibility for the emergent properties of
this system, starting with the fact that the serialization was
such a bottleneck that people became uncomforatable enough with
the choke-hold that they started a completely new repository
using a real source management system.  And Voi'la, we have
NetBSD, the first 386BSD schism.


Similarly, I believe the current CVS system, with the inability
to run simultaneous views on the repository (what Linus Torvalds
and Larry McVoy call "LOD"'s or "Lines Of Developement"), has
similar emergent properties.


You can make fun of the patchkit, as Lynne Jolitz (who I respect,
and whose obvious territoriality at the time is excusable under
the circumsatnces of the time) did, and in a similar vein ("one
third of the patches are good; one third of them are ineffective
but harmless; one third of them are wrong" -- with the implied
"I won't tell you which third is which").  However, without my
effort, and the subsequent efforts of the people I foisted the
job off onto (Nate Williams, Rod Grimes, Jordan, et al.), the
world would be a much poorer place.


Yes, it's obvious that there are problems that result from
emergent properties of the current system.  Those of us with
sufficiently advanced mathematical tools, an understanding of
games and complexity theory, and non-linear dynamics even have
mathematical models that are predictive of the systems overall
behaviour.

It is one thing to call attention to the problem occasionally,
to ensure that it is not forgotton; it is another entirely to
pour salt in the wound.


The problem before us is not to salt wounds, but to design a
system such that it has the resulting emergent properties which
we designate as desirable.  This task is non-trivial in the
extreme, to put it mildly.

Unless you have concrete proposals, I believe that you have
exceeded the threshold of "occasionally" for this particular
thread.


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


> 
> > This is getting really disgusting, and you're wasting a lot of
> > bandwidth with your obtuseness. 
> >
>      I'm obtuse? I reporting facts. You don't like them?
> 
> > SHUT UP if you've got nothing valid to add.
> >
>     I've always SpeakUP when I have valid things to add.
> 
> 
> ---
> "I'd rather pay for my freedom than live in a bitmapped, 
> pop-up-happy dungeon like NT."
> http://www.performancecomputing.com/features/9809of1.shtml
> 
> 
> 
> ____________________________________________________________________
> Get free e-mail and a permanent address at http://www.netaddress.com/?N=1
> 
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-advocacy" in the body of the message
> 



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-advocacy" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199907011900.MAA09564>