Subject: Re: stdio FILE extension
To: James Chacon <jchacon@genuity.net>
From: Greywolf <greywolf@starwolf.com>
List: tech-userlevel
Date: 10/17/2001 00:47:04
On Wed, 17 Oct 2001, James Chacon wrote:

# >But that's just me.  ACtually, no, from the discussion I've seen, it
# >isn't just me.
# >
#
# You're right...It's you and Greg. It's real easy to write off a hard problem
# if it doesn't affect you or you chose not to let it affect you.
#

It doesn't appear to be germane to me and Greg, either (and considering
that I consider myself to be generally at the diametric opposite end of
the universe from him on just about any issue that comes up should say
something....but I digress).

# Fine, you're willing to recompile everything for an upgrade...Most people
# that I've introduced NetBSD to have no intrest in learning the mechanics
# of the large scale make build process. They love things like pkgsrc and
# telling them:
#
# "to go from 1.5 -> 1.6 requires you to recompile everything in pkgsrc and
# you either have to learn the build system so you can prebuild before
# converting or lose all of /usr/pkg after upgrading before you rebuild..."
#
# just does them a disservice.

And carrying around a boatload of brittle dependencies is helping them
out?  Yeah, I know -- "if it ain't broke...".  Fine.

#
# The point of good system releases should be both new features and
# backwards compatibility. Not backwards compatibility necessarily to the Nth
# degree but needlessly breaking old binaries when there are other solutions
# (via symbol renaming, etc) that work in it's place IMO aren't worth the bad
# press/confusion that results from people who have gotten accustomed to old
# binaries "just working" on new system builds.

I think I can hold with that one a bit.  That said, I still think that
when we go to 2.x we ought to start fresh.  I do note that /emul might
be the best place to put this stuff at that point, considering that
everything else we have a COMPAT option for goes in there.

Conversely to others not having to rebuild a major library and up the
major number, is there any reason -- given free reign to rebuild as needed
-- that those of us inclined to tinker oughtn't attempt this?

# James

				--*greywolf;
--
NetBSD: My Computer Runs!