Re: Welcome to the RDF Interest Group

RDF IG members,

As your (until recently absentee) RDF Interest Group chair I'd like to
echo Ralph's welcome to the new interest group, and officially get this
thing kicked into life with a brief note raising a few issues I'm
hoping to see this activity address. 

First off, a brief dull aside on process issues. As you're hopefully
aware, the RDF Interest Group is part of the W3C Metadata Activity, and
like that activity itself is chartered to run until December 1999. At
that point we'll review the way things have been working, including of
course any feedback and suggestions from yourselves, the IG membership.
The RDF IG, following the PICS Interest Group precedent, is a public
forum, and as such has a mix of members from W3C Member organisations
and from the general public. I don't forsee this causing any problems,
but would like to remind you all of the existence of the W3C Process
Document [1] which provides a guide for members and non-members alike.
All the key RDF specifications are publically available [2]. If any
member of this group needs further advice on process issues (for
example relating to the Member confidentiality guidelines) please don't
hesitate to ask either myself or Ralph Swick, the Metadata Activity Leader.

A second item of administrivia concerns the relationship between this
Interest Group / mailing list and the other main RDF public forum,
RDF-DEV [3]. While there is clearly some overlap in scope (and
membership) I believe the relationship between these two lists is
something that'll shake out over time. One option that was discussed
prior to the creation of www-rdf-interest was the possibility of
simply 'blessing' RDF-DEV as the official mailing list of the Interest
Group. We chose instead to set up a separate list, and encourage IG
members to use RDF-DEV for detailed technical discussion for RDF
Developers, and www-rdf-interest for discussion focussed around the
formal charter[4] of this W3C Interest Group.

Which brings us rather verbosely to the 'what happens next' issue.

I'd like here to solicit opinions from the IG membership on directions
and priorities for the group's collective attention over the coming months.
We shall be setting up an RDF IG home page to track issues, proposals,
activity of sub-groups and so forth. I'm particularly keen to see a
concerted effort towards an interoperability test suite as an early,
practical and immediately  useful priority for the group. As Ora says,
implementation is everything. We have already seen on RDF-DEV a number
of subtle aspects of the RDF Recommendation discussed very usefully by
implementors. 


An RDF Interoperability Testbed?

My hope is that the RDF Interest Group progress this further by agreeing
a methodology or architecture for interoperability testing across
RDF-aware software components. There are now a variety of RDF parsers
and API proposals in existence. I can think of at least
SiRPAC, lib-www RDF, the WWW:Rdf perl module, the IBM parser, DATAX,
and Mozilla's RDF parser as possible participants in an RDF
Interoperability test suite. Over the past few weeks I've been
conducting a review of existing publically available RDF
implmentations and will (I expect late next week) post some sketchy
notes on approaches we might take for interoperability testing.
If you have suggestions on methodology, candidate data sets, or
software to test, please continue this thread with a new
subject line such as 'RDF Interoperability Testbed'. For that matter, do
post if you think this a dumb idea! (silence may be read as consent... ;-)



Web Architecture: XML and RDF

A more abstract but equally pressing  topic for consideration by this
group is the ongoing evolution of the RDF <-> XML relationship (see [5]
for overview of some of the issues from W3C staff). This is a matter of great
interest to a number of implementors from the Metadata community
(amongst others). A common request I've heard from many groups is for
the desired ability to adopt the abstract (and rather elegant) RDF data
model without being required to always use the RDF 1.0 XML serialisation
syntax. In other words, to use normal ('colloquial') XML syntax, with DTDs, XSL
etc., but to have somehow a way of derriving an RDF representation of
that data. If this anecdotal evidence is borne out by the experience of
this group, we should spend some time articulating more carefully how the
needs of metadata implementors bear upon the relationship between the
RDF and XML technologies. If your implementation experience with RDF
provides any insight into this issue, again please raise this as a new
thread of discussion with a subjectline of 'XML/RDF' or somesuch.


There are any number of other possible topics for this group's
attention. The IG charter certainly lists more topics than we could feasibly
address in a year! I propose we spend a couple of weeks weighing up
topics to focus on before tightening things up and giving one or two
such problems (eg. Interoperability Testbed, XML/RDF,   
your-issue-here...) the benefit of our collective attention. I've
focussed in this kick-off message on the two themes I hear most loudly
from RDF-DEV postings and from talking to developers. but these are
really just my guess as to topics you're all hoping to see addressed.

So... now's the time to voice an opinion on what you're hoping from the
RDF Interest Group. With my chair's hat on I'll be trying to focus
discussion to some extent on discrete, identifiable challenges. But
really as Ralph says, the agenda for this group _is_ wide open and it is
better to post than to not post.  An interest group should above all
interesting, so do let us know what you're interested in!


Dan
 
--
daniel.brickley@bristol.ac.uk
RDF Interest Group Chair


[1] http://www.w3.org/Consortium/Process/
[2] http://www.w3.org/RDF/
[3] RDF-DEV - a list for RDF developers
    http://www.ilrt.bris.ac.uk/discovery/rdf-dev/ 
[4] http://www.w3.org/RDF/IGcharter
[5] http://www.w3.org/1999/04/WebData

On Wed, 11 Aug 1999, Ralph R. Swick wrote:

> At long last we've found the resources to organize an Interest Group
> for developers, users, and designers interested in discussing the
> W3C Resource Description Framework (http://www.w3.org/RDF/).
> 
> Thanks to Dan Brickley for agreeing to facilitate this discussion
> group and thanks to the University of Bristol for permitting him
> the time to do so.
> 
> From the charter (http://www.w3.org/RDF/IGcharter):
> 
>   "The RDF Interest Group is a forum for W3C Members and non-Members
>    to discuss innovative metadata applications. ... [it] will address
>    both architectural/systems issues as well as application (e.g. T&S)
>    interests; it is a forum for everyone interested in storing,
>    exchanging, searching, and using metadata."
> 
> In other words, the agenda for this interest group is wide open;
> the group will be what you, the participants, make it.
> 
> So feel free to chime in with your ideas, your questions, and your
> answers to others' questions.  Judge for yourself which answers
> might be more authoritative than others.  We're all experts here
> in some area or another.  By sharing our expertise we'll help
> build the greater community understanding that will make the
> Web a better place.
> 
> -Ralph R. Swick
>  W3C/MIT
>  Metadata Activity Leader
>  Technology and Society Domain Technical Director
> 
>

Received on Monday, 23 August 1999 19:28:39 UTC