r8518 - in trunk/BOOK: . chapter01 prologue
Bruce Dubbs
bruce.dubbs at gmail.com
Fri Apr 11 15:20:41 MDT 2008
Randy McMurchy wrote:
> Sukucorp Sukucorp wrote these words on 04/11/08 15:52 CST:
>> On Fri, Apr 11, 2008 at 3:23 PM, Bruce Dubbs <bruce.dubbs at gmail.com> wrote:
>>> I was responding to some earlier comments about sh->dash causing problmes. I
>>> added the part about yacc and awk because I know thy are sometimes used, but I
>>> don't know for sure if they are essential for LFS. It is just making sure. I
>>> started to add lex, but I didn't see flex required.
>>>
>> Without any concrete problems I don't think it is advisable to force
>> the users to update the defaults on their host system.
>
> Agreed. And to some extent, telling users to update the defaults on
> their host system if it isn't necessary could lead to a perception of
> technical incompetence on our part.
Well I suppose that is a possibility, but not telling users of potential
problems does the same thing. We cannot test every possibility and telling
people that there may be problems is a resonable thing to do.
Actually, the problem comes from upstream where developers use implementation
specific features and generic names. For instance, glibc is known to use bash
specific features in scripts that start with #!/bin/sh -- even when told about
it. Now where is the technical competence there?
-- Bruce
More information about the lfs-dev
mailing list