W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > November 2004

Re: UNSAID and OR use cases, from PRIME

From: Arjohn Kampman <arjohn.kampman@aduna.biz>
Date: Thu, 18 Nov 2004 10:51:49 +0100
Message-ID: <419C70B5.7040800@aduna.biz>
To: public-rdf-dawg-comments@w3.org
Cc: Bob MacGregor <bmacgregor@siderean.com>, Thomas Roessler <tlr@w3.org>, Rigo Wenning <rigo@w3.org>, Giles Hogben <giles.hogben@jrc.it>

Bob MacGregor wrote:
> The obstructionists keep shouting "use cases" whenever issues like
> these come up. I regard SQL was a mountainous use case that dwarfs
> the sum of all extant RDF use cases. That means that we should also
> be including ORDER BY and GROUP BY in SPARQL Full, an IN operator,
> and various other SQL flotsam that have been
> very heavily blessed.

I fully agree to that. Sesame's SeRQL[1] query language already supports
OR, NOT and optional path expressions for a while but this hasn't proven
to be enough for a lot of people. Hence we are currently working[2] on
adding set operations, subqueries and grouping and aggregation
functionality. Sorting will probably be the next step, but it is the
most complex feature to translate to an RDF context as it lacks strict
data typing.

On the other hand, I can imagine that the working group is aiming for a
limited version first in order to get the first specification out in
finite time. I consider the current SPARQL working draft to be a good
first step towards a standardized QL rather than as the single and final
step to that.


Arjohn Kampman

[1] http://www.openrdf.org/doc/SeRQLmanual.html
[2] http://www.openrdf.org/forum/mvnforum/viewthread?thread=245

Aduna BV - http://aduna.biz/
Prinses Julianaplein 14-b, 3817 CS Amersfoort, The Netherlands
tel. +31-(0)33-4659987  fax. +31-(0)33-4659987
Received on Thursday, 18 November 2004 09:51:53 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:52:05 UTC