Teleconference.2008.04.16/Minutes

From OWL
Jump to: navigation, search

These minutes have been approved by the Working Group and are now protected from editing. (See IRC log of approval discussion.)

See also: IRC log

Agenda amendments

Sandro Hawke: http://www.w3.org/TR/rif-rdf-owl/


Ian Horrocks: public working draft from RIF on OWL compatibility is out

Ian Horrocks: we should look at section 3 of this and comment

Jeremy Carroll: I believe I will need to review this for HP, so will also volunteer to do a WG review

Ian Horrocks: any volunteers?

Peter Patel-Schneider: can we comment on a document that we helped author?
Jeremy Carroll: yes

ACTION: JeremyCarroll to review http://www.w3.org/TR/rif-rdf-owl/


ACTION: Jeremy to review http://www.w3.org/TR/rif-rdf-owl/

trackbot-ng: Created Action 136 - to review http://www.w3.org/TR/rif-rdf-owl/ [on Jeremy Carroll - due 2008-04-23].
Bijan Parsia: I am the liason
Bijan Parsia: But not in the task force
Jeremy Carroll: (formally I am in RIF, but I am an absent second)
Bijan Parsia: Uli and peter have discussed the compat doc and Uli is planning to look at it even more.

Ian Horrocks: Uli and Peter will look at this anyway, so the action on Jeremy might be enough

F2F3

Sandro is looking into Bristol and Boston and dates and will be able to do a poll soon

Jeremy Carroll: has heard back from Bristol coordinator


Jeremy Carroll: ack JeremyCarroll

Sandro Hawke: will send out pointers to poll shortly

Minutes of Monday's UFDTF meeting

Alan Ruttenberg: they have been taken, and I am waiting for Evan to put it on the meeting page

F2F minutes

Peter Patel-Schneider: F2F2 minutes are minimally acceptable - Bernardo is not correctly listed as scribe, some tidying could be done

Ian Horrocks: any comments? I did some tidying

Alan Ruttenberg: they could use some more cleaning up

Rinke Hoekstra: I thought they looked rather good, overall. We've accepted minutes that weren't as tidy by far

Ian Horrocks: suggest that we approve, but ask scribes to take 10min to fix up

PROPOSED: accept F2F minutes

PROPOSED: Accept F2F2 Minutes

Ian Horrocks: +1
Uli Sattler: +1
Rinke Hoekstra: +1
Alan Ruttenberg: +1
Evan Wallace: second
Peter Patel-Schneider: +1 to accept F2F2 minutes
Jie Bao: +1
Michael Smith: +1
Evan Wallace: +1
Carsten Lutz: +1
Bijan Parsia: +1
Diego Calvanese: +1
Ivan Herman: 0

RESOLVED: PROPOSED: Accept F2F2 Minutes

Alan Ruttenberg: ack JeremeyCarroll

RESOLVED: Accept F2F2 Minutes

Alan Ruttenberg: ack JeremyCarroll
Zakim: JeremyCarroll, you wanted to speak on last weeks minutes

Jeremy Carroll: last week's minutes have only been cleaned up recently, so perhaps we shouldn't vote on them today

Peter Patel-Schneider: 4/16 minutes were minimally acceptable even before Jeremy's tidying

Ian Horrocks: I have tidyed them up already

PROPOSED: accept April 9 minutes

Rinke Hoekstra: +1
Ian Horrocks: +1
Michael Smith: +1
Uli Sattler: +1
Diego Calvanese: +1
Peter Patel-Schneider: +1 to 4/9 minutes
Bijan Parsia: +1
Jie Bao: =1
Jie Bao: +1

RESOLVED: accept April 9 minutes

Action item status

Action 76

Action 86

Jeremy Carroll: I haven't completed it yet, but 86 and 90 are now redundant

Action 100

Peter Patel-Schneider: q+ to ask about Action 100

Alan Ruttenberg: doesn't know what the status is since Jim has left the group

Zakim: pfps, you wanted to ask about Action 100

Peter Patel-Schneider: Jim has left and has never done anything regarding n3 rules


Peter Patel-Schneider: 102 was done

Ian Horrocks: I suggest to kill this one and see whether anybody will ever raise a similar one

Jeremy Carroll: +1


Action 102

Alan Ruttenberg: n3 is at http://www.w3.org/mid/6481386B-0C20-457C-871A-8BF9BB86B5F9@cs.rpi.edu

Michael Schneider: should be closed if Peter is happy

Alan Ruttenberg: as attachment
Michael Schneider: 102 http://lists.w3.org/Archives/Public/public-owl-wg/2008Mar/0206.html


Peter Patel-Schneider: the issue has been appropriately put to death

Action 115

Ian Horrocks: Boris says he has done it

Peter Patel-Schneider: 115 is done, but not documented correctly

Peter Patel-Schneider: later/second changes of Boris needs to be documented

Jeremy Carroll: I don't think more documentation is needed ...

Ian Horrocks: can you do this, Boris?

Ian Horrocks: ack 115
Carsten Lutz: boris is not here, is he?
Bijan Parsia: I didn't understand jeremy


Alan Ruttenberg: the mail you refer to does not contain the full OWL-R n3, only a part of it [Scribe assist by Ivan Herman]
Bijan Parsia: I.e., garbled voice

Jeremy Carroll: I don't think that detailed changes to the mapping rules need to be documented at the level of detail that peter is asking for


Jeremy Carroll: I said: I don't think that detailed changes to the mapping rules need to be documented at the level of detail that peter is asking for
Bijan Parsia: One provides a diff?
Michael Schneider: +1 to postpone
Jeremy Carroll: I said: that at and after last call such detailed tracking is needed

Action 116

Peter Patel-Schneider: 116 is OK
Jeremy Carroll: Peter disagreed

Ian Horrocks: this was uncontroversial?!

Action 117

Jeremy Carroll: yes

Ian Horrocks: done by Jeremy

Action 125

Bijan Parsia: Done

Action 126

Bijan Parsia: Also done

Action 130

Bijan Parsia: I just attached text and links to the action items

Ian Horrocks: is done as well

Due and overdue Actions

Action 43

Bijan Parsia: Tomorrow

Sandro Hawke: will do this in a couple of weeks

Bijan Parsia: I would write test cases

Ian Horrocks: isn't top priority, but would like to see them in the not too far future

Peter Patel-Schneider: various stuff could easily give rise to test cases - one reason they are not being generated is that there is no mechanism

Bijan Parsia: it would help some of my actions if we had test cases

Bijan Parsia: Even an incomplete version woudl be fine
Bijan Parsia: Yes

Ian Horrocks: asks Sandro to help us see test cases

Sandro Hawke: (Yeah, I didn't quite say "will do this in a couple of weeks" -- I said there seemed to be more urgent things, and asked what time pressure there was.)

Action 112

Evan Wallace: will re-schedule

Action 119

Ian Horrocks: believes that this occurred

Action 120

Bijan Parsia: See text in the primer on this (120)

Action 124

Sandro Hawke: is done

Action 127

Bijan Parsia: She did it
Evan Wallace: Done

Action 133

Alan Ruttenberg: http://www.w3.org/2007/OWL/wiki/Primer#OWL_1_Species "OWL Lite is a subset of OWL DL 2 and OWL Full 2 but is no longer a recommended profile." [Scribe assist by Bijan Parsia]

Ian Horrocks: believes this is work in progress and working on it

Action 134

Ian Horrocks: will be done soon

Michael Schneider: what about F2F3?

Issues

Raised Issues

Ian Horrocks: for each, we will have a short initial discussion to see whether we will open them

Issue 110

Peter Patel-Schneider: it's not an issue, just a comment

Zakim: pfps, you wanted to talk about 110

Alan Ruttenberg: perhaps the issue is that somebody else trying to use CURIEs has some problems

Peter Patel-Schneider: this should be made clear

Ian Horrocks: agrees: we have many issues

Issue 111

Bijan Parsia: it would be nice if we could signal, for an rdf graph, under what semantics this document is intended to be used

Michael Schneider: +1 to have some ontology property
Carsten Lutz: would they be allowed or forced to specify that?
Bijan Parsia: Carten, I imagine just allowing htem
Carsten Lutz: allowing is fine, IMHO

Sandro Hawke: agrees that we should have something like this

Jeremy Carroll: remembers that this was already discussed in web-ont

Rinke Hoekstra: And if the flag is incorrect?
Bijan Parsia: It can't be incorrect
Bijan Parsia: Issue 111

ACTION: Jeremy to look up discussion of Issue 111 in previous webont

trackbot-ng: Created Action 137 - Look up discussion of Issue 111 in previous webont [on Jeremy Carroll - due 2008-04-23].

Issue 112

Bijan Parsia: Except in how anything can be incorrect (i.e., I get my intent wrong)

Markus Krötzsch: it's uncontroversial from a semantic point of view, but we should find a good name for this universal property

Ian Horrocks: top-role is not really syntactic sugar as top-thing

Uli Sattler: didn't you say something about this being pseudo top role? [Scribe assist by Alan Ruttenberg]

Markus Krötzsch: for SROIQ, it sort of is

Carsten Lutz: agrees that it can easily be reduced out, but it is not really syntactic sugar

Michael Schneider: I have once managed to make it within OWL 1.1 itself <http://lists.w3.org/Archives/Public/public-owl-dev/2007OctDec/0166.html> :)
Alan Ruttenberg: call it "yente"

Jeremy Carroll: warns cautiously against new vocabulary if it isn't really used

Alan Ruttenberg: skos:relatedTransitive ;-)

Bijan Parsia: finds top and bottom role really useful from a tool developers' perspective

Bijan Parsia: and it would be useful for interaction with users

Alan Ruttenberg: curious about whether inclusion of top/bottom role means that roles will mean that reasoners will need to infer whether roles are equivalentproperty to them?

Markus Krötzsch: universal role might really add expressivity to the profiles

Issue 113

Alan Ruttenberg: this was from f2f, no?
Ian Horrocks: yes
Peter Patel-Schneider: from F2F2: RESOLVED: DL does not have certain OWL Full entailments. OWL-R
Peter Patel-Schneider: does not have certain OWL Full entailments. Vendors can
Peter Patel-Schneider: implement other/related languages if they want.
Bijan Parsia: Didn't we make a choice?

Jeremy Carroll: is an OWL-R reasoner allowed to infer OWL-Full entailments that are not OWL-R entailments?

Alan Ruttenberg: was a raised so that it could be pointed to in the documentation

Peter Patel-Schneider: this was resolved at the F2F

Alan Ruttenberg: so accept/resolve
Bijan Parsia: Shouldn't it be an action then, instead of an issue?

Michael Schneider: if you allow a reasoner to make additional entailments, then you have non-sound reasoning, because they can produce conflicts

Peter Patel-Schneider: what would the action be to do?
Alan Ruttenberg: if it wasn't an issue, presumably it wouldn't have made it in as editor note? Makes sense though

ACTION: document it

Carsten Lutz: seems related to the "signalling semantics" issue raised by bijan
Peter Patel-Schneider: the resolution was after the issue was raised

Jeremy Carroll: suggest to refer to next week to see what happened at the F2F regarding 113

Jeremy Carroll: s/refer/defer/

Issue 114

Alan Ruttenberg: good question

Michael Schneider: which of the different punnings do we want/not want

Peter Patel-Schneider: +1 to bijan
Alan Ruttenberg: Seems sensible to me.

Bijan Parsia: doesn't understand the issue there - it's not precise as it is

Markus Krötzsch: +1 to bijan that we should discuss concrete punning cases directly

Ian Horrocks: agrees

Bijan Parsia: we have ruled out 1 form of punning because we had good reasons to do so -- all others are still there

Bijan Parsia: There is a general argument for punning

Alan Ruttenberg: it seems sensible to me to look through remaining punning and see whether they are useful

Jeremy Carroll: MichaelSchneider and me could look into this in the FullTF

Michael Schneider: there is *no* problem with OWL Full wrt Punning!

Issue 115

Jeremy Carroll: but there might be divergence ...
Alan Ruttenberg: favicon
Uli Sattler: Rinke has a nice OWL2 picture!
Alan Ruttenberg: not an issue for publications

Alan Ruttenberg: 2 different questions: whether we want an icon always and in all browser

Sandro Hawke: not high priority, but fixable

Issue 116

Jeremy Carroll: +1 to fixing link checker problem

Ivan Herman: OWL-R-Full is currently not having some axiomatic triples: we need to see whether we want them or not

Peter Patel-Schneider: this has been decided through the semantics

Peter Patel-Schneider: your first triple follows from the Full semantics, so it should be there

Ian Horrocks: ack m_schnei

Ian Horrocks: so this means that there is a bug in the OWL-R-Full rule set

Michael Schneider: for the mentioned one, it should be there - but the question is whether there should be more..

Michael Schneider: all rules from the RDFS spec should also go into the OWL-R-FUll rules, for the other ones, this has to be decided

Ian Horrocks: ack JeremyCarroll
Bijan Parsia: The axiomatic triples don't seem to be part of the RDFS entailment rules
Bijan Parsia: See: http://www.w3.org/TR/rdf-mt/#rules

Jeremy Carroll: I accept that we need to take Ivan's question and work on the rules/document

Proposals to Resolve Issues

Issue 76 and Issue 77

Bijan Parsia: (so we're no worse off than the RDF semantics rec)
Carsten Lutz: si
Alan Ruttenberg: +1
Peter Patel-Schneider: mute DLP

Ian Horrocks: this is mute

Bijan Parsia: +1
Markus Krötzsch: +1
Uli Sattler: +1
Bijan Parsia: there are a bunch of triples at the beginning of section 4.1 of that document [Scribe assist by Ivan Herman]
Alan Ruttenberg: moot
Carsten Lutz: same for 77, 80
Bijan Parsia: Can we resolve all three with one proposal?

PROPOSED: resolve Issue 76, 77, 80

Bijan Parsia: 76, 77, 80
Michael Smith: 80 is a bit different, since DL-Lite is still there
Alan Ruttenberg: agree

PROPOSED: resolve Issue 76, 77

PROPOSED: resolve Issue 76, 77 as per http://lists.w3.org/Archives/Public/public-owl-wg/2008Apr/0165.html

Bijan Parsia: +1
Peter Patel-Schneider: +1 to mute 76, 77
Rinke Hoekstra: +1
Jeremy Carroll: +1
Carsten Lutz: +1
Michael Smith: +1
Evan Wallace: +1
Alan Ruttenberg: +1
Ian Horrocks: +1
Markus Krötzsch: +1
Diego Calvanese: +1
Jie Bao: +1

Ian Horrocks: because Issue 76 and 77 relate to no-longer existent fragments

Uli Sattler: +1
Michael Schneider: bijan, the RDFS axiomatic triples *are* belonging to the semantic conditions, and *also* to the entailment rules

RESOLVED: resolve Issue 76, 77 as per http://lists.w3.org/Archives/Public/public-owl-wg/2008Apr/0165.html

Issue 80

Ian Horrocks: we had to decide which flavour of DL-lite to have as a profile

Carsten Lutz: This is ongoing work, but I don't think we need an issue for this

Alan Ruttenberg: likes to keep issue since it is an ongoing work

Michael Smith: +! to bijan, carsten. this issue is too broad to know when to close it.
Peter Patel-Schneider: +1 to bijan

Bijan Parsia: would prefer to resolve it because this is the same as with all other works/documents in progress

Jeremy Carroll: +1 to bijan
Michael Schneider: from chapter 7, RDFS spec: RDF entailment lemma. S rdf-entails E if and only if there is a graph which can be derived from S *plus the RDF axiomatic triples* by the application of rule lg and the RDF entailment rules and which simply entails E.

PROPOSED: close Issue 80 as moot

Peter Patel-Schneider: +1 to mute 80
Evan Wallace: +1
Uli Sattler: +1
Ian Horrocks: +1
Carsten Lutz: +1
Alan Ruttenberg: 0
Michael Smith: +1
Jeremy Carroll: +1
Sandro Hawke: +1
Diego Calvanese: +1
Ratnesh Sahay: +1
Rinke Hoekstra: +1
Markus Krötzsch: +1

RESOLVED: close Issue 80 as moot

Issue 67

Ian Horrocks: reification in axiom annotation

Bijan Parsia: Thanks michael...I my search didn't find it
Peter Patel-Schneider: proposal is from Alan
Bijan Parsia: I have an action on this topic
Bijan Parsia: http://www.w3.org/2007/OWL/tracker/actions/129

Bijan Parsia: I have an action related to this, so we cannot resolve it before I have done this action

Jeremy Carroll: at the F2F, alanr said that annotation and reification both are semantically difficult

Alan Ruttenberg: we were going to wait for Bijan
Zakim: JeremyCarroll, you wanted to speak against closing this issue, but OK with next

Jeremy Carroll: is anxious regarding this issue, especially with negative property assertions

Bijan Parsia: I'm supposed to compare it to other proposals
Bijan Parsia: Isn't it silly to discuss this when I have an action to come up with stuff?

Peter Patel-Schneider: there is a proposal on the table

Jeremy Carroll: on neg prop assertions at f2f some people spoke against alan's proposal, and I found arguments compelling
Michael Schneider: +1 to wait for bijan
Bijan Parsia: This is why I have *my action*(
Jeremy Carroll: I didn't feel my coutnerarguments were as strong
Ivan Herman: +1 to wait for bijan
Diego Calvanese: +1 to wait for bijan
Peter Patel-Schneider: ha ha ha

Ian Horrocks: let's move on to an issue we can resolve

Bijan Parsia: This is part of my action :)

Issue 81

Jonathan Rees: neg prop, best argument against was introduction of nominals which raised the expressivity ante unnecessarily [Scribe assist by Alan Ruttenberg]
Bijan Parsia: Action 129
Bijan Parsia: open
Bijan Parsia: Come up with proposals for Issue 67 and Issue 81.

Ian Horrocks: we had a proposal at the F2F


Alan Ruttenberg: we need to wait for Bijan for this as well

Issue 9

Bijan Parsia: The point was that we didn't have agreement on *any* proposal,s o I have action to enumerate and compare them

Ian Horrocks: this should be easy/resolvable

Ian Horrocks: the statement/worry this issue refers to is no longer in the document

Jeremy Carroll: I am happy

RESOLVED: Issue 9

Alan Ruttenberg: moot

Issue 60

Alan Ruttenberg: also not culturally universal :)
Ivan Herman: +1
Alan Ruttenberg: +1
Jeremy Carroll: happy families certainly aren't culturally universal :(
Bijan Parsia: We replaced wine with a sterotypical, imperialistic, western 50's style family
Jie Bao: +1

Ian Horrocks: this is no longer an issue since no more wine in primer

Diego Calvanese: +1
Rinke Hoekstra: +1 very eager
Bijan Parsia: Every happy family is the same. Every unhappy family is unhappy in its own way.

PROPOSED: close issues 9 and 60 as moot

Bijan Parsia: +1
Peter Patel-Schneider: +1 to moot 9 and 60
Jeremy Carroll: +1
Uli Sattler: +1
Evan Wallace: +1
Carsten Lutz: +1
Markus Krötzsch: +1
Michael Smith: +1
Ivan Herman: +1
Jeremy Carroll: so at least unhappy families are interesting
Bijan Parsia: So says tolstoy

RESOLVED: : close issues 9 and 60 as moot

Other Issue Discussions

Michael Schneider: if we have time, we should consider talking about f2f3

Ian Horrocks: these issues have been with us for a while

Bijan Parsia: How about without the static!
Carsten Lutz: not understandable
Bijan Parsia: Much better!

Jeremy Carroll: the language tag issue arose from some prior work. When dealing with natural language literals in OWL, we cannot talk about these things

Carsten Lutz: have to leave, sorry; bye

Jeremy Carroll: I would like to add some primitives: language tags and language range

Alan Ruttenberg: fwiw, I have recently wanted this in some ontology development
Bijan Parsia: If we had XML schema lists, couldn't we handle this?

Jeremy Carroll: it would be similar as the literal handling for XSD derived types

Bijan Parsia: Add a bit of syntactic sugar for langed literals

Michael Schneider: dislikes this because it is domain-specific knowledge

Alan Ruttenberg: the use case is to distinguish bar-codes from comments

Michael Schneider: I dislike it in RDF, too
Zakim: alanr, you wanted to say why
Zakim: JeremyCarroll, you wanted to respond

Jeremy Carroll: it's not domain specific knowledge - it is to relate a language-specific tag to its language

Bijan Parsia: agrees that this is not domain knowledge

Bijan Parsia: it's a xsd datatype, and this is a sensible proposal to use it

Michael Schneider: ok, then I will wait for a concrete proposal,
Bijan Parsia: I'd happily work on one with jeremy

Ian Horrocks: could somebody to come up with a proposal

Michael Schneider: might well be that I misunderstood this issue

Jeremy Carroll: I can go back to my previous work and come up with one

Jeremy Carroll: see my scratch proposal above [Scribe assist by Bijan Parsia]

ACTION: JeremyCarroll to come up with a proposal to Issue 71

Issue 71

Michael Smith: Issue 16 discussion at f2f2 http://www.w3.org/2007/OWL/wiki/F2F2_Minutes#Issue_16_entity_annotations

ACTION: Jeremy to come up with a proposal to Issue 71

trackbot-ng: Created Action 138 - Come up with a proposal to Issue 71 [on Jeremy Carroll - due 2008-04-23].

ACTION: jjc to drive this issue forward to resolutio

Michael Smith: sorry, that was me quoting, not assigning new action
Michael Schneider: didn't peter and boris had proposals for this?

Ian Horrocks: there was an action generated at the F2F2 on Jeremy, but this was clearly to short a time for this

Evan Wallace: Hurray!
Bijan Parsia: easykeys!

Ian Horrocks: end of agenda

Michael Schneider: we have missed F2F3 meeting on this agenda

Ian Horrocks: disagrees - we have agreed that Sandro will put a poll out

Ian Horrocks: would people be willing to discuss easykeys?

Bijan Parsia: Boris actually

Alan Ruttenberg: Peter and Bernardo wanted the more careful semantics

Zakim: pfps, you wanted to ask about Monday meeting for imports task force

Peter Patel-Schneider: will we have an imports TF on monday?

Bijan Parsia: That's fine
Bijan Parsia: The proposal is more fleshed out...please reveiw

Alan Ruttenberg: we discussed having an imports TF on monday

Alan Ruttenberg: has spent some time looking at XML catalogue

Alan Ruttenberg: we could meet and discuss Peter's proposal

Alan Ruttenberg: would think that it would be more productive to not have a meeting next week

Alan Ruttenberg: we can put in Peter's proposal and discuss it in the WG

Peter Patel-Schneider: I cannot put in my proposal because Boris has a lock currently

Alan Ruttenberg: why don't we meet for 10min and see where we are at

Ian Horrocks: suggests to have Boris there as well

Jeremy Carroll: +1 to adjourn

Ian Horrocks: anything else?

Ian Horrocks: +1
Alan Ruttenberg: claps for Ian!
Rinke Hoekstra: thanks
Alan Ruttenberg: bye