[Proposal] Create a lfs-scripts package

Bill's LFS Login lfsbill at nospam.dot
Sat Mar 20 12:04:48 PST 2004


On Sat, 20 Mar 2004, Tushar Teredesai wrote:

> [Please reply to blfs-dev only]
>
> There are some useful scripts that would be useful for a LFSer. A few
> days back I suggested creating a package to collect all the scripts that
> are referenced in the books and also some additional ones into a
> lfs-scripts package. See
> <http://archives.linuxfromscratch.org/mail-archives/lfs-dev/2004-March/043035.html>.
>
> Any comments or suggestions?

Trying to find negatives first ...

I can only easily see one objection. That is the long-espoused desire to
install only what is needed for a basic platform (even anything beyond
an ifconfig for a network card was resisted for a long while) and
minimal utilities/user work environment. I'm not sure this would have
much strength in this case though. See "... positives" below.

Racking brain ... another possible arg might appear that we don't want
to maintain our own custom packages. I believe lfs-utils was disparaged
on this premise (and for other reasons) in a few posts.

Possible argument that exposure to the more advanced things in there
(things that are really BLFS now) might cause some newer users to "jump
the gun", install something before their knowledge/system is ready to
use it and result in more support issues?

Looking at the positives ...

We should probably come into closer proximity to reality. Probably 99%
of our users have network connectivity, would prefer to start working
from their shiny new LFS install ASAP, need to have a few more scripts
(and more guidance) about getting at least the more common connectivity,
routing, DNS needs met (yes, I know there are some pitfalls about the
variety of possible configs). And there's more than just networking.
Since our intended users are "intermediate-to-advanced", it would seem
reasonable to have *available* for them a few extra components that
"intermediate-to-advanced" users could immediately use effectively to
minimize having to reboot back into the host to continue their trek.

ISTM that support requests from users not quite into our intended
audience yet is just a "cost of doing business". We haven't been
successful in limiting users to that category with any of the little
things we've tried in the past, so maybe we should abandon that as a
consideration in these sorts of things. We still target who we target,
but we know that raw n00bs will still appear on the scene.

If that attitude is adopted, then the support activity issues about that
class of user disappear, and we need only consider support issues for
"intermediate-to-advanced" users. I don't see a big possibility of
substantial increase for those users if we offer the proposed package.

If for whatever reason, it does not belong in the "official" book, even
as a reference, it can certainly be offered as a hint.

We have folks willing to maintain it, most are editors and they seem
willing. So why not?

For me +1 to do it.

-- 
NOTE: I'm on a new ISP, if I'm in your address book ...
Bill Maltby
lfsbillATearthlinkDOTnet
Fix line above & use it to mail me direct.



More information about the blfs-dev mailing list