Re: numeric-sequence predicates

Could I suggest that subsequent drafts of XQuery 1.0 only references
Formal Semantics but does not depend on it. XQuery 1.0 should, IMHO,
be written in a way that is abstracted from Formal Semantics; all concepts
should be concisely described in XQuery 1.0 without requiring the reader
to drill into Formal Semantics. I suspect that it must be an NP problem
trying to synchronise the specs ! This way, I think it is feasible to let
XQuery 1.0 (and plausibly Data Model; Function/Operators) to move
on quickly without risk of delays.

----- Original Message -----
From: "Jerome Simeon" <simeon@research.bell-labs.com>
To: "Howard Katz" <howardk@fatdog.com>
Cc: <www-ql@w3.org>
Sent: Tuesday, February 12, 2002 21:08 PM
Subject: Re: numeric-sequence predicates


> Howard,
>
> As pointed out in our latest draft, the Formal Semantics document is out
> of sync with the rest of our publications for now. We are working hard
> on make it consistent.
>
> - Jerome
>
> On Mon, 2002-02-11 at 14:56, Howard Katz wrote:
> > The Formal Semantics document provides a mapping in "6.2.1.7 Predicates"
for
> > range predicates. The rules for predicate-expression evaluation in the
> > XQuery 1.0 documentation however ("2.3.3.1 Predicates") imply that more
than
> > a single numeric value is an error. Why the discrepancy?
> >
> > In fact, I'm curious why numeric sequences in general can't serve as
> > predicates. I had a brief chat with PeterF, who indicated this was
> > semantically possible (although incompatible with XPath 1.0). I'd like
to
> > see the more general case, if it's doable.
> >
> > Best,
> > Howard
>
>

Received on Tuesday, 12 February 2002 22:37:01 UTC