Just a thought that popped into my head

Bill Maltby LFS Related lfsbill at wlmcs.com
Wed May 8 02:56:56 PDT 2002

On Tue, 7 May 2002, Gerard Beekmans wrote:

> On Tue, May 07, 2002 at 12:03:37PM -0400, Jeffrey B. Ferland wrote:
> > consider upgrading to be the best of ideas until either most distributions
> > ship with it, or the kernel team uses that version by default.
> Until most distro's ship with it...
> Okay, that means somebody has to be the first to release a distribution
> with it. Why not let us be among the first ones to do it (eventhough LFS
> isn't a distribution, we can still provide instructions on how to use it,
> that's the closest we'll come to a distro anyway)

Rumors, e-mails, compiler developers concerns - all a valid
*considerations* in making the decision. But, they are the
*least* significant.

The criteria I would use:

 1) What am I doing. I, in this case, being the collective LFS/BLFS/ALFS
 2) Does it work for what I am doing? Are the *workarounds* of an
    acceptable level and nature?
 3) Is the performance acceptable, overall?
 4) Is it stable for what we do?
 5) Will it enable or impede anticipated future activities? I.e.
    will we spend so much time doing workarounds when we try to add
    stuff later (like BLFS X) that it impedes our progress? Are there
    things we'll be doing *soon* that we can't do unless we have it?
 6) Will the performance be acceptable for future stuff?
 7) How long will we live with it?

Other considerations?

Then regradless of other distros, gcc developers, rumors et al,
we go or not. Example, I've seen C++ enhancements referenced. We,
as a community, don't do C++. But some of the packages we install
do. So the only *collective* concern (we must draw lines - indivi-
dual C++ developers needs in the community are lesser considerations)
is "for the packages we use, do we need the enhanced features *now*
and if we need them in the future, when". Other than that, C++ has no
relevance to the decision of if and when we move.

We have the luxury of something that works now, and for the near
term, reliably. We can take as much time as needed to make the
decision - although it is like flogging a dead horse. We will
move forward - it's only when and with what.

We also have the luxury of having co-existing compilers if need be,
although it may be more of a burden than a luxury, in reality.
This can be used to transition over a longer period of time with
reduced risk, *if needs be*..

> -- 
> Gerard Beekmans
> www.linuxfromscratch.org

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