SIMILE Minutes April 04th Call

Agenda:

1. Roundtable status - feedback on definition document (all)

2. Overview of changes to definition doc (butler, bass)

3. Introduce strawman prototype/demonstrator proposal [1]
   request for demonstrators. (butler)

4. Review agenda, logistics, prep for April 9th Face to Face. (bass)

	April 9th
	12-6pm Eastern (lunch included)

	1. Review & Discuss version 0.4 SIMILE Research Drivers (this version will include edits from PI feedback)
	   - common themes, prioritization

	2. Review & Discuss strawman proposal for 1st year demonstrators/prototypes (to be distributed by Wed 7 April).

	3. Revisit: requirements for MIT job postings.

=======

attending:
	eric miller
	mick bass
	rob tansley
	john erickson
	mark butler
	david karger (12:45)
	kevin smathers (12:55)
	mackenzie smith (1:00)	

=======

hygiene discussions:

mick acknowledges and apologizes for late notice on US/UK time delta
(this week only).  We agreed last week to start an hour early, but
most of us didn't note it - including mick when sending meeting notice
:( Meeting begins with miller, bass, tansley, erickson, butler.

can we use irc for scribing.  if scribe speaks another can take over.

need simile issues list.  to include action items.

naming of documents and soon to be discussed others.  need a current
version for documents.  W3C TR approach.  for every version do
previous and next links.

links at bottom of issues lists to source xml and transforms.

=======

1. Roundtable status - feedback on definition document (all)
2. Overview of changes to definition doc (butler, bass)

Mark: 37 issues, some small and almost editorial.  Others significant.

main changes since last time:
     - closed the easy to close issues.
     - Also walked through document and annotate with references where
       that is possible.  Need convention for citations within the
       document?  How should we organize references in the document?

discussion on issues list:
  <http://web.mit.edu/simile/www/resources/issues.html>

Mark - expectation that some issues can be closed easily, via
conversation between editor and individual.  Other issues are more
fundamental, and we will need consensus among the group.

Let's bin issues: those that require consensus, those that can be
addressed One on One.

Mark -
1) karger's website ingest issue [34]
2) distributed resources [33]
3) prioritization of use cases

Eric: [34] may create some tension. MacKenzie's point of view will be
interesting/important

Eric - +1 on prioritization of use cases, and shaping next years work.

Eric - One of the objectives of the SIMILE work is in part
documentation/best practices that allow these technologies to work in
a particular context.  e.g. de-referenceability of namespaces and what
they resolve to (from Mark).  W3C technologies are enabling but not
complete.  Different communities will need best practices to augment
these to enable applications in a particular domain.

approach: figure out what hurts - write that down figure out what
	works - write that down

Mark: maybe incorporate this as a separate use case in the document?

em - JSE, any progress on "noodling" about policy issues?  jse - yes,
but nothing written yet.  Parallel conversation with Bristol asset
utility team.  RDF and policy expression, what work is going on in
this space?  em - for example indecs, creative commons

**action jse to send initital thoughts on policy issues to
  www-rdf-dspace.

em - hygiene, in each meeting need to review open action items.

em feedback on "research drivers" document (just mailed to list) -
many issues raised already in the list.  some editorial, some lack of
clarity, some collapsing in use cases.  points to stress - patron/end
user perspective on getting access to the data.  consider human
annotations as well as machine-oriented annotations.  lessons learned
- take advantage of research aspects, but also recording lessons so
future groups can best take advantage.

em - I've offered in my feedback to sponsor a couple of use cases.

=======

3. Introduce strawman prototype/demonstrator proposal [1] request for
   demonstrators. (butler)

based on use case document, we've produced a new document which tries
to outline some possible prototypes that we could build.

idea: break up into 3 month tasks or less, the kind of thing that can
be given to a student and try techniques.  also to remove
dependencies.

at the moment, have 5 of these tasks.

I. RDF + DAML/OIL schemas.  how do we move frome existing schema reps
	to RDF + DAML/OIL how do we translate instance data.

	build corpus of schemas and instance data

II. Archival of legacy websites

III. Interoperation between multiple schemas use processing tools
	DAML+OIL or OWL, interoperate between diffrerent pieces of

IV. Dissemination Kernel

V. Distribution


   em - distribution connects with SWAD-Europe semantic blog work.  in
   essence - RDF diff.  How do we get the diffs on information stores,
   two versions at two different times.  if this is the case, it is a
   very attractive low-hanging fruit demonstrator.  solves the problem
   of many communities trying to do distribution.

   sample communities: RSS, OAI, wiki update, blogging systems

   rdf diff general solution approach, OAI - more of a point
   solution.

Another from em: joseki-backed semantic-web service: take an existing
subject classification and make available persistent URI's for these
via a josecki interface.  useful but no eye candy.

<karger joins>

Then: integrate with something like haystack to allow simple
navigation.  useful demonstrator for navigating a corpus based on
these URIs.  For example take a collection of data that we've acquired
that uses same URIs for the authority terms, geographic places, etc.
Haystack allows easy navigation of the results using these various
facets.  Karger - already have Metadata-based result-set navigation.

David: need right communication link from SIMILE store and haystack.
required task: connect haystack with Joseki so info in SIMILE store is
available to haystack client.

=======

4. Review agenda, logistics, prep for April 9th Face to Face. (bass)


david - I'll be out for 1hr at 4pm on the 9th.


	April 9th 12-6pm Eastern (lunch included)

	1. Review & Discuss version 0.4 SIMILE Research Drivers (this
	   version will include edits from PI feedback)
	   - common themes, prioritization
	   - issues requiring consensus among PIs

	2. Review & Discuss strawman proposal for 1st year
	   demonstrators/prototypes

	3. Revisit: requirements for MIT job postings.

<smathers joins>

=======

Offline discussion on collaboration hosting:

<all depart save Eric Miller, Mark Butler, Rob Tansley>

em: wiki option from danbri - but more of an open, general
  collaboration

  have the ability to provide cvs access to specific parts of the W3C
  website.  certain hoops to jump through, but fairly well greased.

  hosting viewcvs not currently an option at W3C.

  **action em distribute reference to primer on W3C authentication /
    publishing infrastructure.
  **action bass distribute reference to current athena authentication

<smith joins> brief update on meet, discussion of time mixups.  <smith
departs>


possible hosting alternatives
	mit athena
	w3c
	sourceforge
	ipssources - ?? (connected with collabnet?)
	hpds3
	tigris.org
		back-end sw for collabnet's collaborative development features.

<all depart>

=============================================
Mick Bass


Manager
Research and Business Development
HP Laboratories
Hewlett-Packard Company
1 Cambridge Center
Cambridge, MA 02142


617.551.7634 office    617.551.7650 fax
617.899.3938 mobile    617.627.9694 residence
bass@alum.mit.edu      mick_bass@hp.com
=============================================

Received on Friday, 4 April 2003 15:12:51 UTC