IRC log of swbp on 2004-03-05

Timestamps are in UTC.

07:47:15 [RRSAgent]
RRSAgent has joined #swbp
07:47:28 [Zakim]
Zakim has joined #swbp
07:47:39 [RalphS]
RalphS has changed the topic to: meeting log is
07:53:28 [bijan]
bijan has joined #swbp
07:57:09 [DanC_jam]
DanC_jam has joined #swbp
07:58:29 [fabien_gandon]
fabien_gandon has joined #swbp
07:59:58 [RalphS]
zakim, this will be swbpd
07:59:58 [Zakim]
ok, RalphS; I see SW_BPD(tp2004)3:00AM scheduled to start in 1 minute
08:00:51 [Yoshio]
Yoshio has joined #swbp
08:01:20 [Zakim]
SW_BPD(tp2004)3:00AM has now started
08:01:20 [Zakim]
08:04:26 [libby]
libby has joined #swbp
08:05:11 [dajobe]
dajobe has joined #swbp
08:06:47 [Zakim]
08:06:57 [chaalsNCE]
chaalsNCE has joined #swbp
08:06:58 [dom]
dom has joined #swbp
08:06:58 [gk-mobile]
gk-mobile has joined #swbp
08:07:18 [Guus]
Guus has joined #swbp
08:07:56 [ericP]
ericP has joined #swbp
08:08:34 [bwm]
bwm has joined #swbp
08:09:17 [aldoG]
aldoG has joined #Swbp
08:09:52 [RalphS]
Meeting: SemWeb Best Practices WG
08:09:56 [RalphS]
Chair: Guus Schreiber
08:10:03 [RalphS]
-> agenda
08:10:20 [RalphS]
-> irc log from yesterday
08:10:54 [dlm]
dlm has joined #swbp
08:11:15 [dlm]
is there a url for the log from yesterday?
08:11:53 [RalphS]
-> notes from yesterday
08:12:02 [dirkx]
dirkx has joined #swbp
08:12:03 [DanC_NCE]
-> irc log from yesterday
08:12:05 [ChrisWelty]
ChrisWelty has joined #swbp
08:14:15 [dlm]
thx - got there
08:16:57 [JosD]
JosD has joined #swbp
08:18:46 [gk-mobile]
08:18:54 [gk-mobile]
... is list of topics from yesterday
08:19:13 [RalphS]
Guus: around the table, list your top 3
08:21:52 [chaalsNCE]
[which presumably implies that you should have some intention of putting resources into things you vote for, no?]
08:24:17 [areggiori]
areggiori has joined #swbp
08:24:56 [dlm]
it is hard to hear - are you voting from items on or somewhere else?
08:25:19 [RalphS]
straw poll ...
08:25:34 [RalphS]
1. Porting theaurii to the Web (5 votes)
08:26:37 [RalphS]
s/Portig theaurii to the Web/Thesaurus Porting Method/
08:26:51 [RalphS]
2. WordNet ontology conversion support (6 votes)
08:27:11 [RalphS]
3. Getty Thesaurus ontology conversion support (0 votes)
08:27:28 [RalphS]
4. Units and Measures (4 votes)
08:27:34 [RalphS]
5. Time Ontology (2 votes)
08:27:47 [RalphS]
6. XML Schema Datatypes (3 votes)
08:28:08 [RalphS]
7. MusicBrainz (0 votes)
08:28:26 [RalphS]
DanBri: this is important, but not in my top 3 for the next 6 months
08:28:39 [danbri]
danbri has joined #swbp
08:29:14 [RalphS]
8. Design Patterns (4 votes)
08:29:49 [RalphS]
Time Ontology s/2 votes/1 vote/
08:29:58 [RalphS]
9. Style conventions (4 votes)
08:30:12 [RalphS]
10. RDF in XHTML (in charter, not optional)
08:30:20 [dirkx]
+1 for the Style convention (@semantics)
08:30:23 [RalphS]
11. Ontology Design Issues (5 votes)
08:30:28 [RalphS]
12. Education (1 votes)
08:30:30 [gk-mobile]
GK has a "point of order": one of the topics noted yesterday isn't in the WIKI: "Integration of reasoning tools, and application-specific computation"
08:30:36 [gk-mobile]
GK has a "point of order": one of the topics noted yesterday isn't in the WIKI: "Integration of reasoning tools, and application-specific computation"
08:30:46 [RalphS]
13. World-view note (5 votes)
08:31:20 [RalphS]
Guus: Tools Page is links to parsers, reasoners
08:31:25 [fabien_gandon]
fabien_gandon has joined #swbp
08:31:40 [RalphS]
14. Tools Page (4 votes)
08:32:00 [RalphS]
15. Demos and applications (8 votes)
08:32:17 [RalphS]
16. Links to related techniques: MPEG (1 vote)
08:32:35 [RalphS]
17. Link to UML (1 vote)
08:32:42 [RalphS]
18. Link to TopicMaps (4 votes)
08:33:42 [danbri]
...which bullet was it subsumed by?
08:33:50 [libby]
08:33:52 [danbri]
* /me missed how i18n is handled; it looked like i'd get a chance to vote for it
08:33:55 [gk-mobile]
I'm an observer, so my vote doesn't count, but as someone who is interested in actually applying sweb, I select: "Design Patterns (part whole, defaults)", "RDF in XHTML", "Integration of reasoning tools, and application-specific computation"
08:37:30 [Yoshio]
I'm an observer, too, but I'm just wondering wheter we should divide making tools page and making demos and applications page, I think there should be one portal page (oh, is it becaues we've already heve one?)
08:39:22 [DanC_NCE]
I expect the tools and demos will end up on the same page too, Yoshio
08:41:43 [dlm]
it is much harder to hear on the phone tonight than last night. for example, i can not tell what is going on now so it would be hard to vote
08:41:52 [RalphS]
08:42:20 [RalphS]
second straw poll: which do you expect to participate in?
08:42:37 [danbri]
[aside: please note that I am voting as I hope to join the group, but that further discussion w/ my colleagues and Guus needed before that can be confirmed]
08:42:59 [RalphS]
Demos and applications (5 votes)
08:43:11 [RalphS]
WordNet ontology conversion support (2 votes)
08:44:30 [areggiori]
volunteers to help out with WordNet conversion work (@semantics)
08:45:51 [danbri]
[is 'world view note' is things like relation between DL/Lite/RDFS/Full ?]
08:47:10 [DanC_NCE]
[evidently so]
08:47:58 [dirkx]
Style is asemantics+libby=2
08:48:28 [DanC_NCE]
(I hope to contribute webarch stuff to the "style" bucket, mostly via Ralph)
08:49:23 [dlm]
i do not see any listing here for the ontology design issues but i would expect to work on that.
08:50:28 [dlm]
now i see the list on the agenda page.
08:50:52 [dlm]
i would also contribute to demos and applications
08:52:15 [gk]
I understood ontology design issues to be (sort-of) covered by design patterns
08:53:54 [dlm]
i agree with gk - in past work - i have written on both ontology design issues and patterns in the same paper. that might make sense to consider here
08:55:03 [gk]
I'm puzzled why thesaurus porting is more important than, say, design patterns, w.r.t. getting sweb deployed
08:56:34 [chaalsNCE]
[gk, as a personal perspective, it is because I am doing it, where as I am dubious about the prospects of being given permission to work on collective writing about design patterns]
08:57:08 [gk]
dlm, for info, there was a brief discussion earlier about having "ontology design issues" and "design patterns" being together or not. (But I still hope that design patterns will provide some broader guidance on design principles)
08:57:43 [JosD]
08:59:01 [gk]
chaals, I think I understand. I guess I just wanted to inject a note of reference to the "deployment" aspect of this WG.
09:02:20 [RalphS]
ontology design issues is being merged with design patterns
09:03:19 [dirkx]
I worry about a lot of Owl/Ontologies things at the top of the list; while the world is still looking for practical guidance on simpler RDF issues - i.e. we may be running too far ahead of the crowd; and risk looksing them in the process.
09:03:23 [RalphS]
the transcription into the agenda of the straw poll vote tally that is being projected is no longer in sync
09:03:52 [RalphS]
now reverting to irc scribing rather than transcribing a second copy of what DaveB is projecting
09:04:33 [RalphS]
[speculation on how others who are not present would have voted]
09:05:11 [dom]
[I wonder if using WBS once the WG is in DBWG would help for this kind of exercise]
09:05:50 [chaalsNCE]
[/me noodles on an idea about potential task force leaders putting forward proposals, and having some threshold of people who are prepared to work on it in order to get going. But there's some complication because there are limited resources - perhaps more than the limits on participation]
09:05:52 [chaalsNCE]
09:05:56 [chaalsNCE]
09:06:52 [chaalsNCE]
Ralph: Background - RDF started out thinking that mixed namespaces would solve the mixing content problem, and when XML schema came out it didn't quite get it.
09:07:33 [chaalsNCE]
... so we have 3 activities trying to find a way forward. A year ago we asked people interested in this question to get together, and launched a taskforce under the Hypertext and SemWeb Coordination Groups.
09:07:43 [dajobe]
votes from session above
09:07:52 [chaalsNCE]
... originally led by Joseph Reagle, now by Dom Hazael-Massieux.
09:08:24 [chaalsNCE]
... part of the results are some work done by Dan Connolly and Dom, another input comes from Mark Birbeck.
09:08:40 [DanC_NCE]
"this meeting a year ago" -- the tech plenary... the RDF IG...
09:08:44 [chaalsNCE]
... Point of this session is that we have had this issue for a long time, and we are finally getting critical mass...
09:09:24 [chaalsNCE]
DHM Task force has been around 8 months. Gathered requirements - eg from FOAF, Dublin Core, Creative Commons. THen we went into a couple of proposals -
09:09:47 [chaalsNCE]
... one came informally from HTML group, another was based on stuff Dan Connolly had talked about.
09:10:20 [chaalsNCE]
... In October/Nov I tried to reformulate Dan's work - he will present it, and recently we got a different proposal from Mark Birbeck
09:10:46 [chaalsNCE]
SP (which is better). [said with only a bit of tongue-in-cheek --scribe thinks]
09:10:55 [dlm]
i could hear ralph but i can not hear the current speaker
09:11:00 [dajobe]
slides irc
09:11:39 [dlm]
now i can hear
09:11:49 [MDubinko_]
MDubinko_ has joined #swbp
09:12:23 [chaalsNCE]
DanC Will stop for questions before the proposed solution
09:12:51 [chaalsNCE]
... often you want to store machine data inside documents meant for people. First need is information about the doc. Etc.
09:13:11 [chaalsNCE]
... where it turnns out that etc is a pretty-much infinite requirement, as seen in 1993.
09:13:30 [chaalsNCE]
... [this is hiistory overview]
09:13:46 [chaalsNCE]
... 1996 - PICS came out in response to an acute need.
09:14:12 [chaalsNCE]
... The subject-predicate-object structure of RDF goes back to PICS (well, that's when W3C adopted it)
09:14:37 [chaalsNCE]
... PICS also gave the idea of external people being able to say stuff, not just the author.
09:14:56 [chaalsNCE]
... PICS using round brackets means no conflict to put stuff in HTML
09:15:35 [chaalsNCE]
... 97/8 XML becomes "the one syntax to bind them all".
09:15:50 [chaalsNCE]
... some interesting calculations on a slide (expressed without MathML :(
09:16:22 [chaalsNCE]
... Design of RDF was early adopter of namespaces, and skipped DTDs to anticipate XML schema.
09:16:59 [chaalsNCE]
... slide RDF anticipates...
09:17:21 [chaalsNCE]
... People note that you get jumped on by the validator if you follow the initial non-normative recommendation.
09:18:02 [chaalsNCE]
... and it turns out not to work a something you can write in XML schemas
09:18:30 [chaalsNCE]
... "RDF Core"
09:18:59 [chaalsNCE]
... dotted i's and crossed t's but avoided doing too much new work beyond datatypes. (Yes, this problem is hard... we'll agree on that.)
09:19:37 [chaalsNCE]
... There's lots of stuff around, and I started scraping things out of the data that's around
09:19:53 [chaalsNCE]
... "we need features more than standards..."
09:20:21 [chaalsNCE]
So people put stuff in comments.
09:20:31 [chaalsNCE]
s/So/... so/
09:21:24 [chaalsNCE]
... W3C started a task force in SW-CG, the TAG meanwhile collected issue 13 and split it into pieces
09:21:39 [David]
David has joined #swbp
09:23:10 [chaalsNCE]
... gets to GRDDL slide.
09:24:54 [chaalsNCE]
... there are many ways of extracting different data - you pick the one that gets the stuff you want and turns it into RDF. Then you put a link to it, with a rel grounded by a profile.
09:25:37 [chaalsNCE]
... so "choose your favourite conversion, put a link to it with a particular rel/profile. Price to pay includes that you use XSLT for the conversion".
09:26:07 [chaalsNCE]
... XSLT has issues, but in userland people do actually have it on their machines.
09:26:10 [dirkx]
[do you actually have to fetch the whole resource itself (and be allowed to do so) in order to be able to extract the rule to get to the about data].
09:26:32 [chaalsNCE]
... Syntax then gets to be chosen by the particular sub-community of one or more.
09:26:40 [karl]
karl has joined #swbp
09:26:45 [dom]
[yes, dirkx, you have to]
09:27:21 [chaalsNCE]
Mark Birbeck presenting his stuff.
09:27:29 [gk]
[/me wonders if the <head profile= ...> in GRDDL might use a URI form based on a hash of the transformation source, to avoid possible security concerns from subversion of the XSLT]
09:27:43 [dom]
[remember, the goal is to be able to embed RDF in a document, not to solve the general problem of attaching metadata to a document]
09:27:45 [ivan]
09:30:18 [gk]
[... a similar idea is used in]
09:30:18 [karl]
[file is not valid... ;) QA obsession]
09:30:24 [chaalsNCE]
Ralph: There was a question about where GRDDL fits in. It is a way or an author to declare an explicit transform to get from their personal convention of choice to RDF/XML...
09:30:56 [chaalsNCE]
Mark: how many people were in the SWIG discussion on this?
09:31:02 [chaalsNCE]
Ralph: 50%?
09:31:16 [chaalsNCE]
Mark: HTML have slightly different goals to the RDF group.
09:31:51 [chaalsNCE]
... were working on XHTML 2, and wanted to knwo what to do with meta. At the same time we tried to see if we could make RDF stuff easier to get out of HTML.
09:32:38 [chaalsNCE]
... We tried putting RDF attributes in RDF namespace so you could attach an RDF parser. But it didn't work because you had to keep indirecting the predicate stuff.
09:32:39 [dom]
[it was mentionned that the HTML WG proposal is GRDDL-compliant, but I doubt so, at least with GRDDL using only XSLT 1.0, given that XSLT 1.0 can't deal well with qnames]
09:33:21 [chaalsNCE]
... So we tried to make the same structure, without having to do anything terribly complex fromm the perspective of HTML authors.
09:33:49 [chaalsNCE]
... we suggested Qnames - dc:author not (creator)
09:34:34 [chaalsNCE]
... So we have a meta and link - one has object as a literal, the other as an href / resource. This is what people are used to doing already in HTML.
09:35:17 [chaalsNCE]
... We renamed a couple of things - property instead of name, for specifying the predicate. It can go inside.
09:35:56 [DanC_NCE]
(ooh... I hadn't considered the QName parsing issue.)
09:36:15 [chaalsNCE]
... then we allowed nesting. Which is one of the things that XHTML already has as a general evolution pattern from HTML.
09:36:18 [danbri]
[is xhtml2 is already committed to qnames in attribute values elsewhere?]
09:36:39 [dajobe]
[ danbri no, not as far as I know]
09:36:49 [chaalsNCE]
... went back to the RDF notion, to work out how to interpret the nesting structure - where do bnodes appear, etc.
09:37:11 [DanC_NCE]
(XSLT parsers are no more or less dangerous than any other programming language interpreter. perhaps somewhat safer because they don't have "delete all files on the disk" in the standard library. but bugs can lead to security issues pretty much just like javascript.)
09:37:36 [chaalsNCE]
... Introduce an "about" attribute - can be about a part of the document (something that was missing in meta).
09:38:28 [chaalsNCE]
... shows an example of how to use this to replace the cite attribute. A goal is to reduce the number of special-purpose constructs in HTML if they can be grounded in the semantic web.
09:39:14 [chaalsNCE]
... [scribe's words - internalised it trying to keep up :( ]
09:39:39 [chaalsNCE]
... shows some examples of different constructs.
09:40:25 [chaalsNCE]
... We think you can't go wrong. You can have triples that nobody can do anything with (total anonymity...) but there is no way to create a real error (beyond syntax)
09:41:16 [chaalsNCE]
... Put together a document to see if it can be converted, GRDDL-style. No problem.
09:42:46 [chaalsNCE]
...[shows an XSLT / demo]
09:43:18 [dom]
[in fact, the proposal is only GRDDL-compliant to produce N3 rather than RDF/XML]
09:44:16 [DanC_NCE]
[do you think producing RDF/XML isn't trivial, once you can make N3?]
09:44:38 [chaalsNCE]
a short XSLT (even including comments) that produces N3 (ergo, should be trivial to convert to RDF/XML instead).
09:45:17 [dom]
[it is trivial - that just points that we may need to make GRDDL more flexibile wrt output if we want to combine both techs]
09:45:46 [DanC_NCE]
[no, I'm thinking about changing the XSLT, not GRDDL.
09:45:47 [DanC_NCE]
09:45:49 [chaalsNCE]
[I would prefer not to open it up to two or more result syntaxes...]
09:46:09 [dom]
[hmm... you're right]
09:46:17 [chaalsNCE]
Ralp: So we have several appraoches, people working to meet constraints in different places/ways and our charter is to help them find an answer...
09:46:50 [chaalsNCE]
SP: Our intention is to make this a module that any language can use. You can decorate your content with statements about it.
10:05:59 [David]
David has joined #swbp
10:12:11 [libby]
libby has joined #swbp
10:12:30 [mdubinko]
mdubinko has joined #swbp
10:13:43 [josd_scribe]
next 20 min is trying to summarize RDF - XHTML
10:14:25 [josd_scribe]
SteveP: do it without changing language; getting triples out
10:14:59 [josd_scribe]
SteveP: leverage existing documents and get semantics out
10:15:25 [gk]
jjc: one approach or both?
10:15:30 [josd_scribe]
JeremyC: likes both approaches
10:16:09 [josd_scribe]
MarkB: kind of dislikes adding header stuff in GRDDL...
10:16:11 [gk]
I think that the likelihood of tool support (in HTML document editors) should be a consideration.
10:18:14 [josd_scribe]
RalphS: stresses that the point is "what do the 2 groups need to share"
10:18:51 [josd_scribe]
KarlD: develop a series of use cases - see req doc
10:20:50 [josd_scribe]
RalphS: what about participation of BP group? practical deployment question
10:21:17 [Guus]
Guus has joined #swbp
10:21:32 [danbri]
[bijan popped in to ask whether any chance could talk in this group today about OWL implementation, eg. improving/expanding test suite]
10:21:56 [DanC_NCE]
(a collection of use cases:
10:22:16 [josd_scribe]
Liddy: critical for some groups to move forward; also stresses importance for WAI community
10:25:05 [danbri]
[I'm torn: concerned that qnames in attr values may come back to haunt us, but i want to see this go into xhtml2 and there may be no other pretty-enough syntactic alternative]
10:25:33 [josd_scribe]
SteveP: aggressive schedule for XHTML 2.0; GuusS: is in charter for BP so let's do
10:27:22 [areggiori]
areggiori has joined #swbp
10:27:36 [David]
David has left #swbp
10:28:44 [areggiori]
areggiori has joined #swbp
10:29:00 [bijan]
bijan has joined #swbp
10:29:35 [josd_scribe]
4 to 5 people interested in this taskforce; DaveB: having some review by May 2004? SteveP: right
10:29:52 [DanC_NCE]
10:30:13 [dajobe]
I find the threads on www-html - new to me
10:30:44 [josd_scribe]
DanC: collected som technical comments during the coffe break and will circulate them to the mailing list
10:31:26 [josd_scribe]
RESOLVED: DanC will lead that taskforce
10:31:58 [DanC_NCE]
for at least two weeks ;-)
10:32:36 [josd_scribe]
this afternoon Ralph will come to the the more formal TF mechanisms
10:33:50 [DanC_NCE]
(I expect somebody... steven? to mail www-html and let them know what's up, but I'm not sure anybody accepted that action)
10:33:50 [josd_scribe]
this closes the session with the XHTML WG; thanks very much
10:34:13 [chaalsNCE]
[SP departs, leaving us a little package of consensus :-)]
10:35:48 [josd_scribe]
GuusS: back to the taskforce list
10:36:26 [MDubinko_]
MDubinko_ has joined #swbp
10:37:33 [josd_scribe]
GuusS: what can we start at the short term; evt prepare the ones left out for the next term
10:39:39 [josd_scribe]
GuusS: discussing "top-3's received so far"
10:40:17 [josd_scribe]
JimH nr 2 covered by world view
10:40:56 [josd_scribe]
JimH nr 3 N3 discussion...
10:40:59 [gk]
Re. JimH comment about MIME types, I don't think they're broken, but maybe MIME is being expected to do more than it was designed to do.
10:42:00 [Yoshio]
Yoshio has joined #swbp
10:42:01 [danbri]
[also we've just in rdf-xhtml discussion agreed to work on two rdf syntax approaches already; grddl+xhtmlsyntax]
10:42:47 [josd_scribe]
ACTION DanB: investigate publication of Turtle as SWIG note
10:42:58 [dirkx]
Danbri - feel free to prod/rely on me/David for help here.
10:43:42 [danbri]
dirkx, thanks, i'll take you up on that
10:44:04 [josd_scribe]
now discussing NN's message
10:44:53 [josd_scribe]
now discussing NN's message
10:45:36 [josd_scribe]
have to look for not yet covered points....
10:45:43 [rigo]
rigo has joined #swbp
10:46:03 [josd_scribe]
Pat's message
10:48:44 [gk]
I think Pat's message also alludes to an issue of deciding whether a 3rd-party ontology(author)'s intended usage is compatible with one's own intended usage.
10:48:47 [josd_scribe]
RaphS: give people tools so that they can find existing concepts and check them out
10:49:48 [josd_scribe]
GuusS: "how to work with multiple ontologies" seems to be a new point
10:49:56 [RalphS]
DanC: Best Practices for versioning are desperately needed
10:50:59 [DanC_NCE]
10:51:03 [josd_scribe]
GuusS: proposal to add "versioning" as additional taskforce
10:51:11 [josd_scribe]
GussS: seconded
10:51:33 [josd_scribe]
10:52:06 [DanC_NCE]
re namespace versioning, a W3C namespace policy:
10:52:48 [josd_scribe]
RESOLVED add to the TF list an item "how to cope with evolving ontologies"
10:53:05 [libby]
<gk>I think Pat's message also alludes to an issue of deciding whether a 3rd-party ontology(author)'s intended usage is compatible with one's own intended usage.
10:53:21 [libby]
this is the bit I was talking about
10:54:54 [josd_scribe]
GuusS: nothing yet done wrt to outreach to other groups/areas/communities
10:55:22 [gk]
I wonder if the outreach can be incorporated as source of examples for educational/design pattern/etc materials
10:56:49 [josd_scribe]
RESOLVED add to TF list "working with multiple ontologies"
10:58:34 [DanC_NCE]
(note to self: brain-dump namespace versioning experience)
10:58:36 [josd_scribe]
4 hands raised to work on versioning guidelines?
10:59:19 [dlm]
if one raises a hand if one is or has worked on the issue, add me to the count on versioning as well as multiple ontologies
10:59:49 [RalphS]
(if we can find a sweet spot in documenting an engineering best practice without trying to solve all the hard research problems, that will be useful progress)
11:00:02 [josd_scribe]
GuusS: what's realistic to do supposed we have to deliver at the end of the summer
11:04:52 [josd_scribe]
DanC: RDF - XHTML is one of those (should be on the TF list)
11:05:07 [danbri]
[[re datatypes and DAWG interaction, some datatype (via XQ functions+operators) related text from DAWG charter, "At this stage, it is not clear to what extent XQuery technology is applicable to the task of querying RDF datasets. ... There is a requirement for RDF data to be accessable within an XML Query context. ... The working group should specify at least one mechanism for exposing RDF q
11:05:07 [danbri]
uery facilities in an XQuery environment; that is, a way to take a piece of RDF Query abstract syntax and map it into a piece of XML Query using some form of extension to XQuery. "]]
11:05:47 [josd_scribe]
GuusS: all of top 5 low hanging fruit?
11:07:19 [josd_scribe]
ChrisW: has version of time ontology in OWL
11:09:27 [dlm] is what most people point to
11:09:36 [dajobe]
11:09:43 [dlm]
i have not heard all of this conversation but there is still some work to do on the time ontology
11:09:55 [dlm]
there are multiple axiom sets at the moment
11:10:27 [josd_scribe]
ACTION ChrisW present his time ontology in the next telecon
11:10:36 [DanC_NCE]
ok, it's not done because ~ferguson conflicts with
11:10:48 [dlm]
for example, there is a temporal reasoner component of jtp that is being used by some
11:11:01 [dlm]
and it is not exactly the same as hobbs and colleagues
11:11:55 [josd_scribe]
DanC: present it to get it to the groups satisfaction...
11:12:08 [DanC_NCE]
DanC: this time ontology seems to be done to ChrisW's satisifaction. I'd love him to present it to the rest of us to see if it's done to our satisfaction. Either it is, [in which case we can endorse it somehow] or it isn't, in which case we shoudl write up why and maybe help fix it.
11:13:17 [josd_scribe]
we are actually discussing the non top 5 points; the low hanging fruit in there
11:14:18 [josd_scribe]
pointer to the list is
11:15:13 [DanC_NCE]
didn't chrisw take an action to present the time ontology to us?
11:16:07 [josd_scribe]
GuusS: asking for TF leaders - who's willing to lead "Demos and applications"?
11:16:46 [AndyS]
AndyS has joined #swbp
11:17:34 [ChrisWelty]
zakim, who is on the phone?
11:17:34 [Zakim]
On the phone I see Tp_iles_a, Deb_Mcguinness
11:17:51 [josd_scribe]
DanC: Libby would you be prepared to bring the "demo stuff" in front of the group?
11:18:59 [josd_scribe]
no TF leader yet :-(
11:19:53 [josd_scribe]
people seem to be happy to get *all* mail in their inbox ?!
11:20:43 [josd_scribe]
GuusS: an ALL subject prefix (but this is for the afternoon...)
11:20:55 [gk]
[Mail-tagging... could this be an sweb application; different email addresses (xxx+topic@xxx) get fed to the same mailing list with added metadata which can be used to highlight accordingly.?]
11:21:30 [DanC_NCE]
(the w3c mail servers do currently support +topic, fyi)
11:21:32 [dom]
[any address get also mails sent to AFAIK and FWIW]
11:22:10 [gk]
(danC, and do what with it?)
11:22:29 [DanC_NCE]
(as dom said, goes to the same place as
11:22:38 [josd_scribe]
GuusS: leader for "Worldview Note"? (5 pages)
11:23:28 [DanC_NCE]
Fri, 05 Mar 2004 03:23:18 PST
11:24:16 [josd_scribe]
Deb: did you define what it means "to coordinate a TF?"
11:25:03 [josd_scribe]
ACTION GuusS asking JimH wether he's wanting to coordinate "Worldview Note" TF
11:25:59 [josd_scribe]
GuusS: asking for coordinator "Porting thesaurii to the Web"
11:26:09 [dlm]
yes.... it is approaching 3:30 am.... but i will still probably be willing to run or co-run the ontology design issues...
11:26:33 [dlm]
however i will drop off at your lunch break so it would be good if someone there would co-run with me and take notes in the afternoon.
11:26:48 [dajobe]
just 5 mins before lunch
11:29:11 [danbri]
[eek, I started work on the rdf/ontology/thesaurus vocab in 1998 describes an implementation w/ f-logic tools; definitely time to finish that]]
11:29:12 [DanC_NCE]
(I think I missed the point on XMI/UML etc. ... perhaps follow up in email?)
11:29:25 [josd_scribe]
RESOLVED Deb and Chris to corun "Design Patterns and Issues"
11:29:52 [danbri]
[[I look forward to migrating foaf's wordnet usage to some product of this WG]]
11:30:58 [DanC_NCE]
(I like pairs. deb/chris sounds like a plan for success,to me.)
11:31:07 [dajobe]
11:31:30 [josd_scribe]
Deb: will contribute to other TF's as well
11:32:33 [danbri]
[[wordnet can be used in rdf either in thesaurus-style, or in ontology-style; i expect the thesaurus tf to article those options, after which "doing Wordnet" will be a more clearly defined task]]
11:32:50 [josd_scribe]
RESOLVED JeremyC to lead XML schema datatype after May
11:34:15 [josd_scribe]
RESOLVED OscarC to lead "Tools page" and/or "education"
11:37:49 [josd_scribe]
further asking for volunteering to coordinate: BernardV for "Link to TopicMaps" and MarcoN for "Links to related techniques: MPEG"
11:39:13 [josd_scribe]
BenA: concerned about the "deployment coordination"
11:39:38 [josd_scribe]
JeremyC: maybe a "Deployment chair..."
11:42:12 [josd_scribe]
GuusS: after 3pm people could convene to discuss specific TF matters
11:43:10 [dlm]
one thing before lunch - if anyone is interested in having a f2f co-located with aaai in july in san jose, it would good timing if i know before saturday am my time. I am running the program committee meeting then and we will be working on logistics
11:44:44 [josd_scribe]
==================== time for lunch
11:44:45 [DanC_NCE]
Guus: July doesn't work for me.
11:44:57 [DanC_NCE]
... but thanks for the offer.
11:45:32 [dlm]
dropping off the call now. thanks
11:45:38 [Zakim]
11:46:53 [Zakim]
12:34:36 [karl]
karl has joined #swbp
12:42:59 [dom]
dom has joined #swbp
12:58:16 [dirkx]
dirkx has joined #swbp
12:59:45 [bijan]
bijan has joined #swbp
13:05:25 [ocorcho]
ocorcho has joined #swbp
13:05:30 [libby]
libby has joined #swbp
13:05:40 [Zakim]
13:06:10 [chaalsNCE]
chaalsNCE has joined #swbp
13:06:18 [DanC_NCE]
====== Afternoon session: planning session
13:06:24 [DanC_NCE]
scribe: DanC
13:07:02 [DanC_NCE]
Guus: practical things... communications channels...
13:07:37 [DanC_NCE]
... mailing list: in WebOnt, we had a WG mailing list and a comments list; I gather that's normal for REC-track work.
13:08:07 [DanC_NCE]
... this WG has just one list so far: public-swbp-wg. world-READable archive
13:08:26 [areggiori]
areggiori has joined #swbp
13:09:16 [Yoshio]
Yoshio has joined #swbp
13:09:17 [DanC_NCE]
... does it seem like a good idea to use the label "ALL" for stuff all the WG should read.
13:09:57 [DanC_NCE]
JJC: note replies to ALL messages are likely to get ALL labelled by default
13:11:25 [DanC_NCE]
DanC: you'll want to tune your practices as you go. Labels can help, but don't reply to heavily on them...
13:11:48 [DanC_NCE]
... also, the chair will help a lot with references from the agenda to things you really should have read.
13:11:57 [RalphS]
Guus: try to avoid taking decisions on items that were not on the agenda
13:12:07 [DanC_NCE]
Guus: yes. For example, decisions to be taken will normally be in the agenda.
13:12:49 [DanC_NCE]
Guus: so I'll use "ALL" for mail to the whole WG. please edit the subject line if you follow-up.
13:13:17 [DanC_NCE]
Guus: how about post-access to the list?
13:13:41 [DanC_NCE]
JJC: let's not put technical barriers, but establish a norm that it's WG members who post.
13:13:54 [DanC_NCE]
... but I don't want to preclude brief cross-WG discussions.
13:14:28 [DanC_NCE]
Ralph: we can use "anybody subscribed to any W3C mailing lists can posts"
13:14:45 [DanC_NCE]
Guus: OK, so chair/team contact will gently enforce the "only WG members should post" norm.
13:15:09 [DanC_NCE]
JJC: e.g. conference announcements should not go to the list.
13:15:45 [DanC_NCE]
Guus: as much as possible, use plain text email. Occasionally HTML, valid please. Maybe PDF in extreme cases. Don't send proprietary attachments.
13:16:18 [ChrisWelty]
ChrisWelty has joined #swbp
13:16:20 [ericP]
will the example be in pdf?
13:16:20 [DanC_NCE]
ACTION JJC: send to the WG list an example of using www-archive (and perhaps some relevant advice)
13:16:46 [DanC_NCE]
Guus: don't send html-only email. it's ok as an alternative.
13:17:05 [DanC_NCE]
W3C Guidelines for Email Attachment Formats
13:17:13 [chaalsNCE]
[please note W3C policy on attachments to mailing lists (which says basically what Guus is saying) - and particularly attachments policy at ]
13:17:23 [ChrisWelty]
Oscar, Aldo: are you OK with "Ontology Patterns and ENgineering" as the name of our Task Force?
13:17:36 [gk]
gk has joined #swbp
13:17:46 [DanC_NCE]
Guus: with my disability, it really makes a difference.
13:17:56 [chaalsNCE]
[including mentioning accessibility of documents as a requirement for working with people when you don't know if they have a disability]
13:17:57 [DanC_NCE]
ACTION RalphS: add attachment policies to list archive cover page
13:18:25 [DanC_NCE]
RalphS: do we care who's allowed to subscribe to the list?
13:19:07 [bwm]
bwm has joined #swbp
13:19:33 [gk]
I'd like to subscribe to see messages, even if I don't post
13:20:20 [DanC_NCE]
JJC: why limit whether they can use HTTP or SMTP?
13:21:35 [DanC_NCE]
DanC: I suggest the team contact will bear the costs of this decision; I have some advice I'm happy to give offline.
13:21:51 [DanC_NCE]
-- telecons
13:22:04 [DanC_NCE]
Guus: charter says approx. every 2 weeks.
13:22:36 [DanC_NCE]
... tradition is to use a fixed time in the week.
13:22:51 [DanC_NCE]
... agenda is due 24 hrs in advance. will contain links to other stuff.
13:23:07 [aldoG]
I like the OPEN Task Force
13:23:18 [DanC_NCE]
... teleconference bridge is based in Boston (+1-617-...), "Zakim".
13:24:01 [DanC_NCE]
... then you key in a key code.
13:24:22 [DanC_NCE]
Bernard: pls give it in UTC.
13:24:32 [DanC_NCE]
Guus: yes, we'll give it in UTC plus a few other convenient timezones.
13:24:52 [danbri]
danbri has joined #swbp
13:25:29 [DanC_NCE]
Guus: we use an IRC as a supplement to the teleconference, for notes, queue management, etc. Though we realize not everybody has real-time access to it due to local IT policies etc.
13:26:16 [libby] is useful for times
13:26:45 [DanC_NCE]
... it might seem impossible to have a reasonable conversation with 25 participatns, but it actually works. Chair is the only one with the privilege to interrupt. And the scribe.
13:27:04 [DanC_NCE]
Guus: and now for the teleconference time...
13:27:34 [DanC_NCE]
Guus: how about 1500Z?
13:28:18 [libby]
- meeting planner for multiple tz:
13:29:56 [DanC_NCE]
[... discussion of us/europe/asia logistics...]
13:30:13 [JosD]
JosD has joined #swbp
13:31:37 [DanC_NCE]
Guus: hmm... 0900Boston... 1300Z ... 2100 in [japan?]
13:31:54 [dajobe]
& 2300 in AU/NZ
13:32:14 [RalphS]
actually, 0900 Boston is 1400Z (during normal time)
13:36:24 [em]
em has joined #swbp
13:39:34 [chaalsNCE]
[/me thinks this problem is a fundamental limit to the usefulness of telephones as a medium for communication. Good minutes and working style permitting useful participation via email is a very important redundancy]
13:40:11 [DanC_NCE]
poll: 1 repeating time vs. alternating
13:40:17 [DanC_NCE]
6 or 7 for 1 time
13:40:30 [DanC_NCE]
3 or alternating
13:41:03 [Zakim]
13:41:39 [DanC_NCE]
Guus: for the 1st two months, I'm going to propose 1500Z...
13:41:50 [DanC_NCE]
... I can do either Wed or Thu
13:42:01 [DanC_NCE]
jjc: I have a conflict with Wed 3pmZ
13:42:28 [DanC_NCE]
RESOLVED: to have telcons at 1500Z thursdays for the 1st 2 months.
13:42:34 [DanC_NCE]
i.e. 4 telcons.
13:42:42 [DanC_NCE]
every other thursday.
13:43:13 [RalphS]
next 4 telecons, then review
13:43:34 [DanC_NCE]
Guus: in particular, Thu, 18 March 2004 at 1500Z
13:44:04 [DanC_NCE]
... for 60 to 90minutes.
13:44:39 [DanC_NCE]
JJC to scribe.
13:45:19 [DanC_NCE]
ACTION Guus: forward some tips for scribes (from Dan to webont)
13:45:57 [DanC_NCE]
Jos: where should we send regrets?
13:46:14 [libby]
telecon time in all timezones:
13:46:18 [DanC_NCE]
Guus: I'm ok if you send them to the chairs. Ralphs: and to the team.
13:47:16 [DanC_NCE]
Chaals: sometimes it's useful for other folks than the chair to know about regrets. Ralph: I'll try to make the info available
13:47:54 [DanC_NCE]
Guus: if the regrets come 24hs in advance, I'll try to reflect them in the agenda.
13:48:01 [DanC_NCE]
------ Face to face meetings
13:48:55 [DanC_NCE]
Guus: W3C process calls for 2 ftf meetings per year. Ralph/Dan/Challs: really? [w3c process humor...]
13:49:06 [DanC_NCE]
RalphS: process does require 8 weeks notice.
13:49:23 [DanC_NCE]
Guus: if we assume about 2 per year...
13:49:38 [DanC_NCE]
... and take the 2 year duration from the charter, that's 4 ftf meetings...
13:50:02 [DanC_NCE]
JJC: how about Hiroshima?
13:50:13 [DanC_NCE]
Guus: there's also the tech plenary in March 2005 in the Boston area.
13:50:17 [gk]
[/me notes a thesaurus FAQ recently posted at:]
13:50:56 [DanC_NCE]
Guus: Hiroshima seems a bit late. [when is it?]
13:51:31 [DanC_NCE]
semweb conference in Hiroshima seems to be 7-11 Nov 2004.
13:51:54 [RalphS]
also a W3C AC meeting early in Nov 2004
13:52:23 [DanC_NCE]
Guus: I have a conflict for July, and it's soon, so Deb's AAAI suggestion doesn't seem workable.
13:52:35 [DanC_NCE]
Guus: Aug/Sep [?] seems ideal.
13:52:55 [DanC_NCE]
Libby: SWAD-Europe is doing a closing [shin-dig of some sort]... so I can offer some resource to help [?]
13:53:12 [DanC_NCE]
JJC: PROPOSED: to meet at the tech plenary next yera.
13:53:21 [DanC_NCE]
Guus: I welcome any proposals to host.
13:53:32 [DanC_NCE]
... perhaps to discuss at the next telcon.
13:53:38 [libby]
swad-e shindig would prtobbaly be sept, would have to be europe
13:53:46 [libby]
(pref bristol!)
13:54:03 [DanC_NCE]
ChrisW: I'm happy to host at IBM Watson in NY.
13:54:26 [DanC_NCE]
JJC: HP's always happy to host, but I don't think Bristol, UK is geographically appropriate.
13:54:31 [DanC_NCE]
MikeU: I can look into hosting.
13:54:38 [DanC_NCE]
JJC: I could host in Palo Alto too.
13:55:12 [DanC_NCE]
(missed something about Japan/Australia)
13:56:19 [dajobe]
Guus suggested that maybe if time stays 1500 UTC, have both f2f on west coast, japan or Australia time
13:56:25 [DanC_NCE]
------ end of planning discussion...
13:57:09 [DanC_NCE]
JJC: I have some anxiety about public review of our docs... we should do a 2nd version after publication as a note.
13:57:29 [ChrisWelty]
q+ to discuss task force name
13:57:48 [RalphS]
Guus: task force drafts will be readable by public
13:58:07 [RalphS]
... also during the lifetime of the WG some Notes may have second versions
13:58:30 [RalphS]
DanC: Note is though-of by some as an end state; not revised
13:59:18 [DanC_NCE]
[... discussion of publication of WD then Note... patent implications of WDs...]
14:00:57 [RalphS]
DanC: it is useful if people can know when they need to _start_ paying attention to the deliverable of a Task Force
14:01:06 [chaalsNCE]
[/me notes that WAI explicitly expects to regularly publish revised versions of a number of documents published as Working Group Notes]
14:01:27 [DanC_NCE]
ACTION DanBri: ask SemWeb CG about constraints and advice re WD, Note, etc.
14:01:45 [danbri]
[thanks, better phrased than my attempt i just deleted]
14:01:53 [DanC_NCE]
14:02:25 [DanC_NCE]
14:03:04 [DanC_NCE]
Guus describes T&S domain, SemWeb Activity, SemWeb CG
14:03:36 [Zakim]
ChrisWelty, you wanted to discuss task force name
14:03:36 [DanC_NCE]
ack Chris
14:03:58 [DanC_NCE]
Chris: "Ontology patterns and engineering"
14:04:06 [DanC_NCE]
... email keyword OPEN
14:04:22 [danbri]
re Note action, a relevant Process doc section: 7.1.2 Maturity Level When Ending Work on a Technical Report
14:04:22 [DanC_NCE]
... I propose, on behalf of this taks force about design patterns and such
14:04:36 [danbri]
14:04:37 [danbri]
A Working Group MAY publish a Working Group Note with or without its prior publication as a Working Draft. W3C MAY also publish "Interest Group Notes" and "Coordination Group Notes" for similar publications by those types of groups. Interest Groups and Coordination Groups do not create technical reports that advance toward Recommendation.
14:04:38 [danbri]
14:04:57 [DanC_NCE]
Bernard: [... external organization... ?]
14:05:23 [DanC_NCE]
... how about saying "hello" to some other groups?
14:05:49 [DanC_NCE]
Guus: let's talk about that at an upcoming telcon. sounds interesting.
14:06:29 [DanC_NCE]
JJC: would you care to draft something, Bernard?
14:06:54 [danbri]
[[charter, re IPR cites -> ]]
14:07:02 [DanC_NCE]
ACTION Bernard: draft a message to be sent from this WG to some other group for discussion at an upcoming teleconferece
14:07:30 [RalphS]
DaveB: what dependencies exist between SWBPD and DAWG?
14:08:06 [danbri]
rrsagent, pointer?
14:08:06 [RRSAgent]
14:08:11 [RalphS]
DanC: one possibility envisioned was that BPD saw querying as a high priority for a Best Practice, that hasn't seemed to have happened?
14:09:15 [DanC_NCE]
Guus: size seems relevant to "style" task force... nobody...
14:09:24 [DanC_NCE]
dirkx: actually, I expressed interest
14:10:06 [DanC_NCE]
[... connection with a DL workshop...]
14:10:42 [DanC_NCE]
Guus: there's a DL workshop 2-4 June... co-locate a meeting? only if we'd do 3 ftf's per year... [idea gets little support]
14:11:30 [DanC_NCE]
Mike: how about things that we didn't decide to start?
14:11:40 [DanC_NCE]
Guus: I intend to revew those at the end of some telcons.
14:12:24 [DanC_NCE]
Mike: application reasinging integration [?] didn't get on the list. [ I thought it did.]
14:13:02 [DanC_NCE]
Mike: if I'm going to do that anyway... should I do that independently?
14:13:16 [DanC_NCE]
Guus: you can propose a new task force, and we'll try to keep a list...
14:13:44 [DanC_NCE]
JJC: I've got a few things that didn't particularly interest this group; I'm working on them elsewhere; perhaps I'll bring them back here when they're more baked.
14:14:05 [DanC_NCE]
------ Closing
14:14:23 [DanC_NCE]
Guus: thanks much for the participation despite the short notice!
14:14:34 [DanC_NCE]
... between the call for participation and this meeting
14:15:24 [em]
well done everyong
14:15:29 [DanC_NCE]
RESOLVED to adjourn, with applause to Guus for chairing.
14:15:29 [danbri]
[[ Constraints/advice re WD vs Note for SWBP WG and SWIG From: Dan Brickley <>
14:15:30 [danbri]
Date: Fri, 5 Mar 2004 09:14:46 -0500
14:15:30 [danbri]
14:15:30 [danbri]
14:15:40 [danbri]
(I believe that closes my action above)
14:16:05 [Zakim]
14:23:22 [Zakim]
14:23:29 [RalphS]
zakim, bye
14:23:29 [Zakim]
leaving. As of this point the attendees were Tp_iles_a, Deb_Mcguinness, EMiller
14:23:29 [Zakim]
Zakim has left #swbp
14:23:38 [RalphS]
rrsagent, bye
14:23:38 [RRSAgent]
I see 6 open action items:
14:23:38 [RRSAgent]
ACTION: DanB to investigate publication of Turtle as SWIG note [1]
14:23:38 [RRSAgent]
recorded in
14:23:38 [RRSAgent]
ACTION: JJC to send to the WG list an example of using www-archive (and perhaps some relevant advice) [2]
14:23:38 [RRSAgent]
recorded in
14:23:38 [RRSAgent]
ACTION: RalphS to add attachment policies to list archive cover page [3]
14:23:38 [RRSAgent]
recorded in
14:23:38 [RRSAgent]
ACTION: Guus to forward some tips for scribes (from Dan to webont) [4]
14:23:38 [RRSAgent]
recorded in
14:23:38 [RRSAgent]
ACTION: DanBri to ask SemWeb CG about constraints and advice re WD, Note, etc. [5]
14:23:38 [RRSAgent]
recorded in
14:23:38 [RRSAgent]
ACTION: Bernard to draft a message to be sent from this WG to some other group for discussion at an upcoming teleconferece [6]
14:23:38 [RRSAgent]
recorded in