> Date: Tue, 23 Dec 2003 11:28:42 -0500
> From: Ray Denenberg <[log in to unmask]>
>
> > > Because I'm not sure where this "requirement" would (or whether it
> > > should) be expressed.
> >
> > Right there in the spec! You just quoted it!
>
> No, I doesn't work that way. Sorry to get too "formal" here (and in
> fact I'm trying to avoid excessive and gratuitous formality) but if
> you say "srw requires ..." in a document which is in effect a
> service definition, it begs the question where the cited requirement
> is stated.

I don't see the problem. The service definition is a requirement. If
a service does not satisfy tyhe definition, it's not providing the
defined service, so it fails the requirement. This is how things work
in IEFT RFC and ISO standards. Why should SRW be different? The
specification should damned well tell people what to implement.
That's what it's for.

 _/|_ _______________________________________________________________
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ "Do I understand you correctly, Professor Groves? Surely
         you're not advancing the astonishing theory that the mentality
         of Neanderthal man compares favourably with that organ which
         ten million years of evolution has developed in his modern
         counterpart?" -- dialogue from "The Neanderthal Man".

--
Listen to my wife's new CD of kids' music, _Child's Play_, at
        http://www.pipedreaming.org.uk/childsplay/