Re: Gauge interest in an XDR implementation

[Ralf Wildenhues]

Hello Bruno, Richard, all,

* Bruno Haible wrote on Wed, Nov 05, 2008 at 02:03:06AM CET:
> Richard W.M. Jones wrote:
> 
> > The request brings up the issue of what the limits of what can be in
> > gnulib.  I'm not sure it is a useful discussion...
> 
> My opinion: If it's "common" in the sense that more than one GNU package
> is likely to use the code, it's welcome in gnulib. And if it's a collection
> of small codes, rather than a big atomic chunk of code, it also technically
> makes sense to have it in gnulib.

However, if it has a stable API, and well-defined semantics, you might
as well spare every gnulib cloner the needed disk space and make XDR
a stand-alone library, since we just heard again that a stable gnulib
API is not anywhere close.

You can still decide whether to make it static or shared or both, and a
distribution can pick the shared build to avoid security fix hassles.

> Yes, this definitely can have its place in gnulib, because it is glibc
> functionality that is not present on all other systems.

Notable difference is that glibc provides a stable API and ABI, unlike
gnulib.

I may sound like ranting, but that's not my intent; rather, the freedom
gnulib gains by not providing stability, has a cost, and it is often
paid by people not developing gnulib: users of gnulib and distributors
of packages that use gnulib.  This setting may lead to interests being
represented on this list in a skewed way.

Cheers,
Ralf