Ada: yes or no?

Bill Maltby LFS Related lfsbill at wlmcs.com
Sat May 18 06:02:29 PDT 2002


On Sat, 18 May 2002, Matthias Benkmann wrote:

> On Sat, 18 May 2002 08:47:19 +0200 Jeroen Coumans <jeroencoumans at gmx.net>
> wrote:
> 
> > On Friday 17 May 2002 22:50, Ian Molton wrote:
> > > Yeah. I was hoping someone else would comment, actually...
> > 
> > The best suggestion I've heard so far which circumvents all problems 
> > (both philosophy-wise and support-wise) is to install gcc-core and g++ 
> > instead of the whole package.
> 
> How does this solve the philosophy problem? You still need to grab a
> precompiled binary later if you want to add more compilers. THAT is the
> philosophical issue.

I agree it is the issue. But as someone pointed out earlier, we start
off building from "binaries". Every rule has an exception. Maybe this
one s/b an exception out of pragmatism?

Some folks will need Ada. LFS may want to *not* install it as part
of the defualt for many good reasons. What then is the *practical*
solution? There are several.

But it appears to me that the one permitting LFS to come closest to
one of its goals of providing a good distro for the *user*, while
satisfying another need, keep build times minimized is the good one.

I guess this means the hint route. The note in the book about the
build time increase shouldput most people off Ada if they don't
need it. So mail traffic increase s/b minimal.

And if we really want to keep the installed LFS distro pure, we
could instruct them to make another partition available and in-
stall the (*shudder*) binary there. mount, do-it-to-it and unmount.
Then torch the *offending* partition.  ;)

> 
> MSB
> -- 
> Indecision is the key to flexibility.

Bill Maltby
billm at wlmcs.com

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list