Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 01 Aug 2002 14:59:19 -0700
From:      Terry Lambert <tlambert2@mindspring.com>
To:        Zak Johnson <zakj-freebsd-arch@nox.cx>
Cc:        freebsd-arch@FreeBSD.ORG
Subject:   Re: OpenSSL vs. -lmd
Message-ID:  <3D49AF37.C7E1E272@mindspring.com>
References:  <200207311641.g6VGfRWj099655@freefall.freebsd.org> <20020801143059.GA536@nevermind.kiev.ua> <200208011151.55478.mi%2Bmx@aldan.algebra.com> <3D498FB4.6987B696@mindspring.com> <20020801195640.GQ26797@madman.nectar.cc> <3D4998F9.A736EA85@mindspring.com> <3D499CF3.4030601@ntlworld.com> <3D49A37B.BA3C2982@mindspring.com> <20020801212004.GC6856@opiate.nox.cx>

next in thread | previous in thread | raw e-mail | index | archive | help
Zak Johnson wrote:
> On Thu, Aug 01, 2002 at 02:09:15PM -0700, Terry Lambert wrote:
> > > If I had the time, I would be so sorely tempted to roll my own *BSD
> > > distribution based on that idea.... FreeBSD without the fluff.
> > >
> > > I'd have to roll some kind of binary distribution channel as there is no
> > > system compiler ;)
> >
> > There have been a lot of us who have felt this way.
> 
> I'm inclined to say that an entirely new distribution is a bad idea,
> because it will involve a great deal of duplicated effort.

I'd agree, if it were duplicating the effort that we all
felt so strongly about.  But then if that were the case,
there would be no need for it, since the code would be
there already, and it wouldn't be a problem.

So the part people are concerned with is certainly not
something which would be duplication.


> What was done recently with Perl was a tremendous step in the right
> direction, and a good compromise for those of us who want a smaller base
> system.  I would like to see this sort of packaging extended to other
> unnecessary packages in the base system.  The largest barrier seems to
> be differing definitions of "unnecessary" in this context.

No, the largest barrier is that *everything* is not a
package, so that *anyone* can make a decision on what
"unnecessary" means to them, without shooting anyone else
in the foot.

In other words, if you are trying to define "unnecessary",
then you are addressing a symptom, rather than the problem.

The "make installworld" target defines what is and is not
in "The Base System".  For better or worse, this is an
Ultimate Truth.

Without tackling the problem at that level, it's unlikely
that it will ever be solvable.

If you were around when John Dyson decided to start an SMP
kernel project, as a drop-in replacement for the FreeBSD
kernel, you'll remember that the project flopped.  John was
also trying to address a symptom, without addressing the
base problem.  He wanted to create an SMP system, but the
major attribute of such a system is kernel reentrancy.  This
is also the major attribute of an RT system (as one example
of problem domain expansion).  But John did not want to
permit RT (again, an example) in the resulting system.  In
other words, he did not expand the set of problems that the
resulting system would be able to solve.

The net effect was that the newly declared project did not
attract volunteers, since it wasn't really solving any new
problems, other than John's own perception of the messyness
of the existing code.  Cleaning up the mess is a goal that
could have been achieved through simple code refactoring,
if people would permit what would be, without argument,
"gratuitous changes" (making code readable is *always* a
gratuitous change; worse: it's subjective).

With three strikes against it starting out, it's no wonder
that it went nowhere.

The "OpenSSL vs. -lmd" discussion is, similarly, just a
symptom of the real problem.  The problem in this case is
third party mega-packages, where you have to make an all or
none choice between a set of components, rather than on an
individual component basis.

Ask yourself "Why is it an issue in the first place?"  The
answer is that the "-lmd" library is seperate in the legacy
code, but glommed into the OpenSSL library, along with other
cruft, making it no longer severable.  The issue is one of
severability.


I think the idea of fundamentally breaking the system into
optional and highly granular components has a large amount
of "grass roots" support; discussions like the packaging
discussion would not attract nearly so much participation,
were that not the case.

To readdress your idea of "an entirely new distribution":
if that's the only way the problem will be permitted to be
solved, then it's a likelihood; one way or another, the
problem *will* be solved, eventually, likely sonner than
later.

-- Terry

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3D49AF37.C7E1E272>