Explanation of the Network Inference DAWG Strawman Objection

Dear WG Members, W3C Leadership et al,

At a face-to-face meeting in Carlsbad, CA on 15 July, 2004, the RDF Data
Access Working Group (DAWG) voted to select a 'strawman proposal' to be
used as the basis for continuing standardization work. Several
XQuery-based approaches had been suggested to the group (including the
'XQuery as a strawman' option suggested by the group's charter).
However, none have generated significant interest among group members
other than Network Inference and invited expert to the group Howard
Katz. 

Network Inference has suggested that a reduction in scope to a simple
objective that the DAWG language be compatible with the XQuery surface
language. Such an approach would allow implementations which support the
limited algebra of the DAWG language fragment, while also providing
users the opportunity to more fully integrate the two languages.
Additionally, agreement on an XQuery surface language approach would
provide a foundation upon which to build further query interfaces to OWL
and an eventual Rules layer. 

Despite this compromise, the working group outright rejected any
requirement or objective which expressed any commitment, at any level,
to XQuery.

We believe that the DAWG working group is making an egregious error by
rejecting any level of commitment to XQuery at this critical juncture.

By choosing to move forward without any requirement or objective to take
this specification towards XQuery compatibility, the group not only
ignores a specific mandate in their own charter, but also risks
producing yet another query specification that could be deemed
irrelevant by major vendors. Significant vendor support from IBM,
Microsoft, and BEA, to name a few, indicates that XQuery is indeed
commercially viable and will very likely be a successful query language
with broad adoption potential.

Besides the strength and momentum behind XQuery, Network Inference fears
that a one-off query language overly coupled to an RDF view of the world
will result in further fragmentation of the Semantic Web family of
languages. We believe that this is a golden opportunity to unite the
Semantic Web with a common query surface layer. Drawing upon many
lessons from past standardization efforts (ANSI SQL for example), we
believe that a query language can serve as a catalyst to spur widespread
user adoption and acceptance.

While regrettable that Network Inference was the lone objector in this
working group's strawman proposal, the company stands firm in support of
the DAWG charter in the belief that XQuery, with minor extensions, would
be the best overall foundation on which to enable query-based access to
the family of Semantic Web languages.

Regardless of outcome, Network Inference will remain devoted to our
customer feedback by continuing our XQuery support for query-driven
inferencing across RDF and OWL data inside our Cerebra Server product
family.

The purpose of this letter is to make sure that DAWG participants, W3C
leadership, and Chairpersons of relevant WG's are apprised of the
rationale behind our objection to the recent strawman language decision
within the DAWG. We feel that our support for an XQuery context to DAWG
is consistent with the group's charter, W3C architecture goals, and
indeed the needs of the wider commercial adopters of all Semantic Web
technologies.

Sincerely,

Jeff Pollock
Vice President, Technology
W3C AC Rep
W3C DAWG Group Member

Received on Friday, 16 July 2004 15:46:01 UTC