[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: [xtm-wg] Swindon Minutes Module #2: RDF
Swindon Minutes Module #2: RDF
| This is a hunk of the draft minutes of the October 13-15, 2000
| meeting of the XTM Authoring Group in Swindon, England.
| Those who were present at the meeting are invited to suggest changes
| for the sake of recording the intent of the meeting more accurately.
***************************************************************
RDF and W3C "Summit" Meetings
There have been three 2-hour teleconferences in which Eric Freese,
Steve Newcomb, Michel Biezunski, Eric Miller, Ralph Swick, Dan
Connolly and Dan Brickley have participated. The records of what
transpired in these teleconferences, such as it is, in the form of
a transcript of the internet relay chat (IRC) record is available
at http://www.w3.org/2000/10/10-topicmap-rdf. (These records are
probably inscrutable to anyone who didn't participate in the
conference calls, and, in any event, not all participants in the
conference calls had access to IRC.) In any case, the W3C people
were very clear and unanimous that they would be pleased to be
able to recommend the XTM syntax for the interchange
("serialization") of RDF statements, pending their satisfaction
that XTM syntax is up to the task and that it would in no way
compromise or restrict the RDF model. At least one W3C
participant expressed enthusiasm for the possibility that Topic
Maps could ultimately serve as the basis for the W3C's "Semantic
Web" initiative. One requirement for any possibility of such an
outcome is that W3C-Recommended URIs (i.e., URLs and/or URNs) must
be the required syntax for addressing (referencing) all resources,
where the notion of "resource" includes everything that might ever
be addressed for any XTM-defined purpose.
Discussion:
The XTM Conceptual Model Subgroup (CMS) has it in mind to show the
relationship between the Topic Maps and RDF models.
The "essential triple" does appear in both RDF and Topic Maps,
viz:
The RDF statement:
http://www.w3c.org/rdf/... ---[creator]--> "Ora Lassila"
(resource) (arc) (resource)
... states that Ora Lassila is the creator of the resource that
is available at some URL.
In Topic Maps, exactly the same expressive goals might be
achieved by uttering an association link that establishes a
two-role relationship between two topics. The two roles of the
relationship would be
(1) "creator" and
(2) "thing-created",
and the subjects of the two topics would be
(1) the resource found at the indicated Web address (the
resource itself -- not the Web address, and not the notion
expressed by the resource), and
(2) Ora Lassila (Ora Lassila himself, not the string "Ora
Lassila").
It was also suggested that W3C's adoption of XTM might help to
clarify the RDF model by disambiguating the distinctions between
(1) names,
(2) addresses,
(3) resources, and
(4) the significance of resources:
... all of which distinctions are carefully defined and preserved
throughout the topic maps paradigm.
Two other possible mappings between the "RDF triple" notion and
the topic maps paradigm were also briefly discussed:
(1) One involved regarding the two resources as the occurrences of
a topic. While this possibility accounted for the fact that
two resources could be involved, topic occurrences are not
intended to establish relationships.
(2) The other possible mapping of RDF triples into XTM involved
the use of facets. The facet construct seems, in fact, to be
the single construct in the whole ISO 13250 standard that is
most similar to an RDF statement. However, there was little
enthusiasm in the AG for the idea of recommending facets as
the mapping point between XTM and RDF. RDF is already more
powerful than facets, facets are really not essential to topic
maps, and nothing that can be accomplished with a facet cannot
also be accomplished with topics and associations. Therefore,
RDF would not gain the significant benefits of topic maps from
being married to XTM in this way, and neither XTM nor the
topic maps paradigm in general would benefit from being
married to RDF in this way, either. (See also the report of
the Syntax Subgroup in these minutes, in which it was decided
that the "facet" construct that appears in ISO 13250 should
have no counterpart in the XTM Spec.)
One possible downside to marrying XTM and RDF was suggested by one
Participating Member whose organization has experienced political
and/or servicing problems in its efforts to register one or more
URN schemes. It is reportedly onerous to register such a scheme,
and the success of registering one cannot be predicted. It was
not known to anyone present at the meeting exactly what benefits,
if any, are obtainable by successfully registering a URN scheme,
and what penalties, if any, are incurred by using an unregistered
scheme. In order to support the Semantic Web, XTM will need the
ability to use at least one URN scheme, whether registered or
unregistered. Such a scheme will be needed in order to support
the public availability of public subject descriptors (PSDs), and
particularly in order to support direct comparison of public
subject identifiers (PSIs) by XTM-conforming applications.
(A new subgroup, the Public Subject Identifier Subgroup, was
chartered by the Syntax Subgroup to devise and propose a URN
scheme for PSIs, and to do such research on the subject as it sees
fit; see also the report on this elsewhere in these minutes.)
(See also the report on NewsML in these minutes for an interesting
sidebar about an economically significant industrial organization
that has recently devised and has fully adopted a yet-unregistered
URN scheme.)
***************************************************************
-Steve
--
Steven R. Newcomb, Consultant
srn@coolheads.com
voice: +1 972 359 8160
fax: +1 972 359 0270
405 Flagler Court
Allen, Texas 75013-2821 USA
-------------------------- eGroups Sponsor -------------------------~-~>
Get FREE long-distance phone calls on Tellme!
Dial 1-800-555-TELL, say "Phone Booth"
http://click.egroups.com/1/9816/4/_/337252/_/971902497/
---------------------------------------------------------------------_->
To Post a message, send it to: xtm-wg@eGroups.com
To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC