static libraries' policy

[Ralf Corsepius]

On Sat, 2005-11-12 at 13:05 -0500, Daniel Veillard wrote:
> On Sat, Nov 12, 2005 at 08:21:28AM -0800, Toshio Kuratomi wrote:
> > On Fri, 2005-11-11 at 18:36 +0100, Christian.Iseli at licr.org wrote:
> > > buc at odusz.so-cdu.ru said:
> > > > In  other words - what I should do  with static libraries today?
> > > 
> > > So far, I have seen pretty good arguments why static libraries should *not* be 
> > > included by default:
> > >  - they introduce security risks
> > >  - they add bulk to devel packages
> > >  - some features of modern OS/compiler combos only work with shared libs
> > > 
> > > I don't think I have seen a good argument why static libraries *should* be 
> > > included *by default* (except maybe "convenience", but convenience to whom ?).
> > > 
> > Daniel Velliard's LSB argument was accepted by Jens as a reason to
> > modify his original proposal.  I've never looked into the LSB until
> > today so I'm not certain of the argument but this is what I've found:
> > 
> >   A strictly conforming application shall not require or use any
> >   interface, facility, or implementation-defined extension that is not
> >   defined in this document in order to be installed or to execute
> >   successfully.[1]_
> > 
> > [1]_
> > http://refspecs.freestandards.org/LSB_3.1.0/LSB-Core-generic/LSB-Core-generic/application.html
> > 
> > If this isn't the relevant section, Daniel or jens will have to speak
> > up :-)
> 
>   yes that was my point basically.
There is no fixated Linux ABI. This renders all attempts to build LSB
compliant packages moot and renders the LSB moot as a whole.

Ralf