RDFCore WG minutes for the Telecon 2002-02-15 from Dave Beckett on 2002-02-16 (w3c-rdfcore-wg@w3.org from February 2002)
From
: Dave Beckett <
dave.beckett@bristol.ac.uk
Date
: Sat, 16 Feb 2002 17:58:41 +0000
To
: RDF Core <
w3c-rdfcore-wg@w3.org
Message-ID
: <22881.1013882321@tatooine.ilrt.bris.ac.uk>
RDFCore WG minutes for the Telecon 2002-02-15
---------------------------------------------

Transcript:
--
(text version attached)

Agenda:
--
1: Allocate scribe: Dave Beckett

2: Roll Call
Participants:
- Brian McBride (chair)
- Eric Miller
- Dave Beckett (scribe)
- Dan Connolly
- Ron Daniel
- Jos De Roo
- Jan Grant
- Martyn Horner
- Frank Manola
- Stephen Petschulat
- Patrick Stickler
- Aaron Swartz
- Pat Hayes
- Sergey Melnik

Regrets:
- Daniel Brickley
- Graham Klyne
- Mike Dean

Absent:
- Frank Boumphrey
- Jeremy Carroll
- Bill dehOra
- Rael Dornfest
- Yoshiyuki Kitahara
- Michael Kopchenov
- KWON Hyung-Jin
- Ora Lassila
- Satoshi Nakamura
- Pierre G Richard
- Guha

3: Review Agenda
AOB: none

4: Next telecon - 10am Boston time, 22 Feb 2002

5: Review Minutes of 2002-02-08 telecon

See
Accepted

6: Confirm Status of Completed Actions

ACTION: 2001-11-16#7 Pat
ACTION: 2002-02-08#1 bwm
ACTION: 2002-02-08#5 Jeremy
ACTION: 2002-02-08#6 FrankM

Confirmed complete.

7: Face to face meeting

See:
Received regrets for the F2f from
Frank Manola
Stephen Petschulat
Aaron Swartz
Ron Daniel

Had a straw poll on interest in participating in the F2f by phone
and/or IRC and got interest from several people for phone access.

ACTION: 2002-02-15#1 EricM: Find details of phone access during F2F
ACTION: 2002-02-15#2 Brian: Consider holding teleconference(s) during F2F

8: Model Theory WD

OLD ACTION 2002-02-08#2: Publication of RDF Model Theory WD

Completed:

RDF Model Theory W3C W3C Working Draft 14 February 2002
Congratulations were given to Pat and thanks to EricM for helping the
with the publication process.

9: Status of Test Cases WD

OLD ACTION: 2001-11-30#3 JanG Get access to test case areas of W3C site
CONTINUED
OLD ACTION: 2002-01-11#2 JanG Post summary of Test Cases WD
outstanding updates to list.
CONTINUED
OLD ACTION: 2002-01-11#1 bwm Persue CVS access for Jan with EM
CONTINUED

Question was raised if Pat Hayes was subscribed to www-rdf-comments
and its purpose. He wasn't so would look into subscribing.

ACTION: 2002-02-15#3 Brian: Announce new Model Theory WD to
appropriate lists: rdf-logic, interest, ...

10: WG Status

Brian noted that the status of the working group needed to be
considered since it is chartered to finish early 2002. At present,
the group has approximaltey 28 open issues, 7 postponed and 26
closed. This will be on the agenda for the F2F.

11: Preparing for the f2f

* rdfms-seq-representation: The ordinal property representation of
containers does not support recursive processing of containers in
languages such as Prolog.

ACTION: 2002-02-15#4 PatH: Send a few paragraphs to the list to address this

* rdfms-assertion: RDF is not just a data model; an RDF statement is an
assertion.

Brian said the director has an architectural requirement that we say
something about this.

PatH said that he was already down to do this

ACTION: 2002-02-15#5 PatH, Graham: Will draft some appropriate words
on the nature of assertions

* Schema Issues

Postponed, danbri not present

* Contexts

Pat Hayes noted that webont are discussion such things and if it was
possible to say something on this, we should. This is mostly on how
it could be fixed, could do, rather than must-do and would be more
likely something for a future RDF Core WG with a new charter.

ACTION: 2002-02-15#6 PatH: Draft something on contexts for the WG to
consider in order to indicate to others our position

12: Issue rdfms-propElt-id-with-dr

Clarify the interpretation of an ID attribute in the propertyElt
production within a Description element with a distributive
referrant.

Propose:

o the WG resolves that this issue be closed on the grounds
that with the removal of rdf:aboutEachPrefix and rdf:aboutEach
there are no distributive referrants and the issue is mute.

See:
--

APPROVED

[scribe note: typo in proposal? "moot" not "mute"?]

13: Issue rdf-terminologicus

The RDF community needs a precise terminology to enable it to discuss
issues. (Martyn Horner)

Propose:

o the WG resolves that this issue is addressed by the primer and
that this issue be closed.

See:
--

Discussion later brought up the note that the current editors draft
of the primer has no glossary section so original working amended to
the above.

APPROVED

14: Issue rdfms-graph

Formal description of the properties of an RDF graph.

Propose:

o the WG resolve that the model theory is a formal description
of the properties of an RDF graph and that this issue be
closed.

See:
APPROVED

15: Issue rdf-formal-semantics
The RDF Model and Syntax Rec and RDF Schema CR do not
provide a formal specification of the semantics of RDF.

Propose:

o the WG resolves that the model theory defines formal
semantics for RDF and that this issue be closed.

See:
APPROVED

16: Issue rdfms-fragments

Propose:

o The WG resolves that the meaning of absolute
URI's with fragment ID's is a matter of web architecture and
beyond the scope of this WG and that this issue be closed.

See:
--

This item was skipped.

17: Issue rdfms-literals-as-resources

Consider replacing literals with resources whose URI uses the
data: URI scheme.

Propose:

o the WG resolve that the proposed change would be a major
change to the RDF specification and is out of scope for
this WG.

See:
APPROVED

18: Issue rdfms-literalsubjects
Should the subjects of RDF statements be allowed to
be literals?

Propose:

o the WG resolves that the current syntaxes (RDF/XML,
n-triples, graph syntax) do not allow literals as subjects.

o the WG notes that it is aware of no reason why literals
should not be subjects and a future WG with a less
restrictive charter may extend the syntaxes to allow
literals as the subjects of statements.

See:
--

The above resolution was updated (one word change from agenda text).

Clarification was asked by DanC whether for any entailments in the
model theory, statements ended up with literal subjects. DanC's
example was does

{ rdfs:range rdfs:Literal. "23". }

entail

{ "23" rdf:type rdfs:Literal }

JanG and PatH responded that no, it would not happen and that there
are guards in the model theory to stop the above case.

APPROVED

19: Issue rdf-containers-otherapproaches
The design of the RDF Model collection classes exhibit various
awkward features. Might these be augmented with a 'better' design?

Propose:

o the WG resolves this issue is out of scope for this WG
but places the issue on the list of to be considered by a
future WG.

See:
APPROVED

20: Issue rdfms-uri-substructure

xmlns, uri+name pairs or just uris..? Clarification needed

Propose:

o the WG resolves to close this issue on the grounds that
changing how resources are named on the web is a web
architecture issue and beyond the scope of our charter.

o Whereas:

(a) the RDF 1.0 spec says that property and class names
are computed from element and attribute names
by concatenating their namespace names with their local names

(b) it's useuful to be able to process RDF with
XPath and XSLT, where even though
concat(namespace-name(qname1), local-name(qname1))
is the same as
concat(namespace-name(qname2), local-name(qname2))
the qnames themselves may not compare equal in XPath expressions.

(c) lots of implementors have looked for advice on how to serialize
RDF, and, in particular, how to compute a namespace name and
localname from the name of a property or a class.

the WG advises RDF schema/namespace/vocabulary designers

(d) choose namespace names that end in non-xml-name-characters
such as / # ?

and we advise implementors of RDF serializers:

(e) in order to break a URI into a namespace name and a local name,
split it after the last XML non-name character. If the URI ends in a
non-name-character throw a "this graph cannot be serialized in RDF
1.0" exception.

See:
--

The proposal in the agenda was amended in (e) in order to get the
maximally-allowed XML name to be made from the characters at the end
of the URI as Brian discussed in
APPROVED

ACTION: 2002-02-15#7 DaveB: Add sections addressing this resolution
to the Syntax WD.

ACTION: 2002-02-15#8 JanG: Add test cases based on those suggested by Dan in:
21: Issue rdfms-boolean-valued-properties

No standard vocabulary is defined for representing boolean valued
properties. The author of this suggestion proposes the introduction
of two new properties, rdf:is and rdf:isNot. To represent the fact
that someone likes chocolate, their resource could have the property
rdf:is with a value of foo:ChocolateLover.

Propose:

The WG notes that since a boolean-valued property can be
identified with a class, rdf:type can be used to
represent boolean valued properties. Thus:

.
.

can be represented by

.
.

The WG notes that RDF(S) defines no built in mechanism for expressing that
ChocolateLover and ChocolateHater are disjoint classes. The WEBONT WG
are defining mechanisms for such expressions.

The WG resolves to close this issue.

See:
--

The second of the alternative proposals in the agenda (above) was
discussed.

APPROVED

22: Model Theory for Containers

Discussion of PatH's proposal for the semantics of rdf:Bag:
and DanC's noted an idea on bags (isn't really a bag, but a bag
nearby) that he would write up for the group:

ACTION: 2002-02-15#9 DanC: Write up rdf:Bag idea and send to list

23: Reification

Items from FrankM's Reification "subagenda" in
23.1 Entailment #1

Does?

.
.
.
.

.
.
.
.

.

entail:

.

Discussion of this entailment; FrankM proposed that the answer is
NO. Some people are confused or don't care too much about it.

APPROVED: Answer to above entailment is NO.

23.2 Interpretation of decision

Discussion of what words to say about this, where and if
reification syntax now has a meaning. No consensus on agreeing if
the above entailment (and reification decision) was cast into
English right.

23.3 Entailment #2

Does?

.

entail

_:r .
_:r .
_:r .
_:r .

APPROVED: The answer is NO

ACTION: 2002-02-15#10 JanG: Add both entailments #1 and #2 to the
test case suite

JanG noted he has to work with Jos on manifests and support for
entailment tests in the test cases areas.

23.4

Not explicitly discussed.

Sergey asked if we have a mechanism for reification or not, it seems
we are deprecating it, so we should at least clarify a mechanism for
serialising reified statements. DanC noted the syntax is still there
and it is DaveB noted the mapping is described in the syntax WD.

24: Datatypes
Review status and plan.

Lots of discussion till meeting close and afterwards.

DanC asked had S-B had gone away?
wants to constrain the string values of a statement to the lexical
range of the datatypes. Discussion continued as the meeting closed.

END OF MEETING

Some sketchy post meeting chat notes:

DanC's question and ways to implement it - new machinery /
properties or can do with existing ideas? DanC wants to talk of
the lexical space of the datatype.

DanC gave this example:
"apple"
dc:date rdfs:range xsdt:date.lex

would be datatype-illegal

Somebody suggested
xsd:date rdfs:range
would allow URI xyz to be used as the lex space of the xsd:date datatype
but might need some MT tweaks (not rdfs:range but rdfs:drange?)

Sergey replied to this later in more detail:
Supporting only W3C XML Schema datatypes as built-ins.

What does it mean to implement RDF datatypes as currently specified

Renaming rdf:value - having opposite properties

----

[scribe: See attached IRC log for some notes, those after meeting end
are more intermittent]

Dave
14:28:03 logger_3 has joined #rdfcore
14:28:03 topic is: RDF Core WG Meeting 15:00-16:00 UTC Friday Agenda:
14:28:03 Users on #rdfcore: logger_3 bwm @dajobe
14:28:19 dajobe has left #rdfcore
14:29:51 jan_g has joined #rdfcore
14:38:06 bwm has quit
14:38:06 jan_g has quit
14:42:23 dajobe-scribe has joined #rdfcore
14:52:04 em has joined #rdfcore
15:00:34 * dajobe-scribe waves
15:00:48 * em waves
15:00:51 * em dialing
15:01:04 * em wonders where everyone is.... ?
15:04:14 ircleuser has joined #rdfcore
15:04:14 meeting starts
15:04:16 roll call
15:04:22 -danbri
15:04:27 regrets danbri
15:04:31 +bwm +em
15:04:33 +daveb
15:04:34 -frankb
15:04:39 jc regrets
15:04:40 danc regrets
15:04:44 rond absent
15:04:48 billd absent
15:04:52 jos present
15:04:54 -rael
15:04:56 +jang
15:05:00 +martynh
15:05:05 -yoshiyuki
15:05:08 graham regets
15:05:16 michael absent
15:05:19 kwon absent
15:05:22 ora absent
15:05:25 frankm present
15:05:41 satoshi absent
15:05:44 stevep present
15:06:02 pierre absent
15:06:05 patrick present
15:06:12 aaron present
15:06:18 miked regrets
15:06:20 guha absent
15:06:23 path present
15:06:30 sergey absent
15:06:49 +rond
15:06:54 3 review agenda
15:07:01 no aob
15:07:05 4 next telecon
15:07:16 same time next week
15:07:20 item 5 prev minutes
15:07:24 approved
15:07:30 item 6 completed actions
15:07:36 approved
15:07:38 item 7 f2f register
15:07:40 DanCon has joined #rdfcore
15:07:47 * DanCon realizes he's late
15:07:57 please check list and send regrets if not comming to bwm: ACTION
15:08:03 or tell now
15:08:18 frankm: not coming, regrets
15:08:32 stevep: regrets
15:08:35 aaron: regrets
15:08:45 rond: regrets
15:08:56 are folks interested in participating in part of the ftf by phone?
15:09:00 and/or IRC?
15:09:15 * dajobe-scribe forwards Q
15:09:25 pathayes: yes, if poss - exact times please
15:09:38 frankm: yes, me too
15:09:49 aaron, rond +1 each
15:09:51 prefer phone
15:09:54 how much time? I find 2 hrs is about the longest I can be productive by phone
15:10:06 * dajobe-scribe forwards that
15:10:19 bwm askes em to work out phone facility
15:10:28 the facilities are availble, EricM.
15:10:34 action em: phone link
15:10:49 JosD has joined #rdfcore
15:10:52 bwm will think about this
15:11:16 discussion of time, timezones and f2f time w.r.t telecon
15:11:21 action bwm: think about
15:11:54 item 8 model theory wd
15:11:57 JosD, let's write some N3 rules about who's interested in which agenda items, when the sun is up where, etc. (hmm... actually, simmulated annealing is probably better for scheduling stuff)
15:11:58 published
15:12:01 well done pat
15:12:04 thanks to em
15:12:07 q+
15:12:08 item 9 test cases wd
15:12:16 DanCOn: ?
15:12:25 oops... no zakim queue-manager
15:12:27 jang: cvs access now, working on it. will be done by f2f
15:12:40 wet towels
15:12:49 DANC now on telecon
15:13:04 danc notes path not subscribed to rdf-comments
15:13:19 action path: register on comments
15:13:33 discussion of status to www-rdf-comments
15:13:48 Well, DanC, that't an excellent idea!
15:14:12 action path: announce to rdf-logic, interest, ...
15:14:35 "pat hayes'" model theory - no, the group model theory
15:14:53 above action on bwm not path
15:14:56 item 10 wg status
15:15:02 bwm: more at f2f
15:15:17 bwm: looking at issues list, 28 still open, 7 postponed, 26 cloed
15:15:30 and out of time - charter to finished early 2002
15:15:34 item 11 preparing for f2f
15:15:54 seq-representation issue
15:16:00 nneed volunteer
15:16:09 or put on next time list
15:16:18 DanC, I meant to use N3 for that (not SA or GA)
15:16:43 path: will write para on that, or find it and send again
15:16:56 ACTION path: will send a few paragraphs to address this issue
15:17:03 issue rdfms-asseriton
15:17:16 the director needs something on this as an architectural requirement
15:17:36 action path, gk: already down to do this. On the nature of assertions
15:18:13 issue schemas
15:18:18 postponed, no danbri
15:18:20 issue contexts
15:18:34 bwm: broad support for moving on, but path got something?
15:19:28 path: big big storm on webont and if it were possible we could say something to calm waters, we should ...
15:19:40 ... but say something about how it could be fixed, would be useful at this point
15:19:50 ... would like to give us that chance, hold on
15:20:04 AaronSw` has joined #rdfcore
15:20:08 bwm: ok, labelled as could do rather than must-do
15:20:22 path: technically is can't do, but for next wg, could say something
15:20:41 action path: post something on contexts in order to indicate to others, future plans
15:20:45 item 12
15:20:57 propelt-id-with-dr
15:21:15 bwm: since abouteeach* gone, is moot - closed?
15:21:20 danc: seconded
15:21:24 APPROVED
15:21:27 item 13
15:21:32 terminologicus
15:21:54 bwm: glossary in primer resolves this - closed?
15:22:05 APPROVED
15:22:10 item 14 rdfms-graph
15:22:23 bwm: model theory is a formal desc of rdfms-graph - close?
15:22:38 path: comment, if model theory rewritten, then will track back to re-address this
15:22:58 bwm: want a category "Queued for closure" but just more work, rather nopt
15:23:02 bwm: propose again?
15:23:07 APPROVED
15:23:35 frankm: resolution of 13 re glossary in a primer
15:23:41 ... but there isn't one now
15:23:59 discussion of glossary and where it is
15:24:32 proposal to change item 13 resolution s/the glossary in the//
15:25:24 _will_ be addressed by the primer, no? hmm... I suppose everpresent "_is_" is ok
15:25:38 bwM; approved ammended item1 3 resolution
15:25:43 item 15 rdf-formal-semantics
15:26:02 proposed resolution - MT defines it?
15:26:03 APPROVED
15:26:08 item 16 rdfms-fragments
15:26:13 skipped
15:26:19 item 17 rdfms-literals-as-resources
15:26:31 (16 moved to later in agenda, bwm)
15:26:40 consider the proposal for 17
15:26:43 APPROVED
15:26:48 item 18 rdfms-literal-subjects
15:27:16 bwm reads resolution in agenda
15:27:31 jang: amendment, s/resources/subject/ in 2nd paragraph
15:27:39 i.e. "not be subjects..."
15:28:04 danc: clarification re entailments in MT, do any of the MT entailments end up with literals as subjects
15:28:11 jang: no, not unless they start with literal subjects
15:29:08 { rdfs:range rdfs:Literal. "23". } => { "23" rdf:type rdfs:Literal }
15:29:36 path: ther eare guards to stop above happening
15:29:40 there are
15:29:43 danc: ok
15:29:49 bwm: call for dissent?
15:29:58 APPROVED
15:30:01 hm... that pretty much blows S-B out of the water.
15:30:05 item 19 containers-other-approaches
15:30:27 propose to rule this out of scope for this wg, punt to later wg?
15:30:29 APPROVED
15:30:35 item 20 rdfms-uri-substructure
15:30:50 ... long long summary .... :)
15:30:51 DanCon, can you elaborate a bit on this?... are you saying that S-B is no longer feasible?
15:30:54 No, I think S-B is OK, but you just can't use rdfs:Literal to any real extent.
15:31:02 ugh. phone died...
15:31:12 bwm reads proposing first para?
15:31:21 aaronsw: comments on para 1?
15:31:22 Whoops, that's PatH called 'ircleuse'
15:31:40 try /nick PatH
15:31:53 +aaronsw (phone again)
15:32:09 bwm re-reads 1st para
15:32:26 APPROVED 1st paragraph
15:32:32 (before "whereas...")
15:33:01 nick PatH
15:33:09 patrick: need (e) change as previously posted to list,f rom bwm
15:33:33 bwm: deleted "and no other ..."
15:33:36 up to the comma
15:33:40 ircleuser, need a / in front of nick
15:34:05 change to "after the last non xml-name character"
15:34:11 ircleuser is now known as PatH
15:34:30 hi Pat!!!!!!!!!!!!!
15:34:35 Hi Jos
15:34:36 patrick: maximally long name
15:34:42 dajobe: yeah, I think I get it
15:35:13 APPROVED "whereas..." (a) .. (e) plus modifications
15:35:28 ACTION daveb: add to syntax wd
15:35:35 ACTION: test cases editor to add test cases based on dancs
15:35:41 item 21 booleanv-vallued-properties
15:35:50 order? shall we move on to reification/datatypes?
15:35:59 bwm: propose the 2nd of the two ("alternatively..>")
15:36:19 * dajobe-scribe notes DanC order
15:36:58 bwm proposes
15:37:00 APPROVED
15:37:16 item 22 model theory containers
15:37:22 AaronSw` has quit
15:37:24 bwm: danc you have some ideas here?
15:37:42 danc: bag isn't really a bag, but a bag nearby, we can explain that
15:37:48 pathayes: sounds interesting
15:38:03 ACTION danc: write up this up today
15:38:11 s/this/his idea on the explanation/
15:38:13 item 23 reification
15:38:23 subagenda
15:38:28 reif item 1
15:38:35 (frankm)
15:38:42 entailment 1
15:38:50 frankm: do we agree the item is no?
15:39:03 some agreements
15:39:13 jos: i have a problem with this; subject should be more defined
15:39:18 * dajobe-scribe not recordign this well
15:39:41 jos: could say this entailment is true
15:39:47 frankm: I'm confused by this
15:40:03 the source that pathayes is refering to is the source of the original stratement
15:40:06 (path) yes
15:40:18 .. so I could write
15:40:27 ... why would I need that reification syntax anyway?
15:40:38 ... I could just make all teh statemsnts I want about that thing
15:40:38 JosD, as far as I'm concerned, this rdf:subj/pred/obj stuff is worthless, but as graham said, it's "mostly harmless".
15:40:43 path: talk about the subject
15:40:53 ... and provides an explict rdf map to the triple
15:40:59 ... oh, nevermind!
15:41:06 AaronSw has joined #rdfcore
15:41:17 josd: I don't care too much. It could work but it is meaningless
15:41:31 bwm: if there is no actual dissent, lets decide and move on?
15:41:37 frankm: Agreed the answer is no?
15:41:43 APPROVED entailment 1 answer is NO
15:41:50 reification agenda item 2
15:41:59 hmm... I'd like to suggest that reification gets no screen-space in the primer.
15:42:08 frankm: have we decided an isntance of rdf:Statement is a stating?
15:42:15 danc: what is the impact?
15:42:25 ... what docs will change?
15:42:35 pathayes: some of the words in M&S seem to say the opposite
15:42:40 danc: where will we write this down?
15:42:50 ... in model theory / primer, that replace m&s
15:42:58 frankm: are we contradicting something int he current docs?
15:43:10 danc: my prefication is that we leave it in the syntax doc and that is all
15:43:20 pathayes: in primer maybe for reassurance?
15:43:50 daveb: already in syntax doc, in simplest form
15:43:59 bwm: rdf community familiar with the Stating term
15:44:04 frankm: should be explained
15:44:12 patrick: an occurance of the statement
15:44:20 aaron: in the glossary
15:44:49 frankm: resolve to agree that they are statings?
15:45:05 bwm: are we clear on what we agreeing?
15:45:15 frankm: guidance for documetn writers
15:45:21 pathaes: something in MT doc
15:45:45 +sergey
15:45:48 discussion of what goes where
15:45:56 sergey: this vocab is undefined everywhere?
15:46:13 frankm: not discusing reification cvocab?
15:46:17 sergey: yes we are
15:46:27 danc: yes answer is they turn into non-magical properties
15:46:37 pathayes: not meaningless, no special meaning
15:46:43 sergey: don't get the difference
15:47:17 frankm: want to make a decision on "the meaning of rdf:Statement" and seperate issue of the 4-triple vocab
15:47:45 rond: uncomfortable approving statment versus stating, want something more concrete
15:47:52 the most I can do with "an rdf:Statement represents a stating" is abstain. I don't think it's an interesting thing to decide.
15:48:12 frankm: talking about appearances of various things in rdf
15:48:34 jang: point2 is casting into English what is meant by the entailment in point 1
15:48:43 danc: hard to agree to that?
15:49:14 bwm: move on for now to reification item 3?
15:49:41 frankm: believe item 2 follows from 1
15:49:44 (sounds of dissent in bg)
15:49:49 reification item 3
15:49:55 and an entailment
15:50:01 frankm: do we agree the answer is NO?
15:50:22 ... does a triple entail it's reification?
15:51:04 action on the test case editor?
15:51:05 APPROVED the answer is no to entailment
15:51:11 no 2 under item 3
15:51:25 bwm: no more reification issues left?
15:51:35 ACTION jang: add to test suite?
15:51:40 s/?//
15:52:09 jang: need to work with jos on this
15:52:15 sergey: do we have a mechanism for reification or not?
15:52:22 ... seems we are deprecating it
15:52:36 ... so we should at least clarify a mechanism for serialising reified statements
15:52:41 xfering across the network
15:52:48 danc: syntax still there
15:53:01 pathayes: yes, but where is the connection between the original triple and the 4-triple reificaiton
15:53:07 danc: in teh syntax
15:53:36 daveb: yes, there is a section on this in the syntax
15:53:57 bwm: to sergey, move this to the list
15:54:10 .. use cases based on how do I do this?
15:54:31 frankm: something w.r.t. frank's point of identifiying the original triple
15:54:33 take what?
15:54:35 action on who?
15:54:37 s/frank/path/
15:54:39 whatever.
15:54:56 item 24 main agenda - datatypes
15:55:19 still some bits incomplete in last thing (V3?) posted to list
15:55:22 (pathayes)
15:55:34 ... true local typing
15:56:06 ... magic triple asserts everything is local in item 3
15:56:07 (!)
15:56:18 ... gives the user more options
15:56:34 ... not more complexity, just draws users attention to more options
15:56:46 s/item 3/section 10 of v3 datatyping doc/
15:56:54 danc: has S-B gone away?
15:57:09 danc: ie. range of this property is constrained to literals that are the lexical representation of dates
15:57:12 pathayes: still there
15:57:26 for a datatype?
15:57:27 bwm: no
15:58:02 danc: such as one for the xsd:date
15:58:35 pathayes: says yes
15:58:42 bwm: misunderstanding the Q?
15:58:57 bwm: "10"
15:59:05 bwm:
15:59:08 and have xsd:date.lex
15:59:19 danc: not what I mean
15:59:35 ... want to constrain the strings you have as property values
15:59:44 pathayes: how do you do that?
15:59:53 danc: want to constrain to the lexical range of the datatypes
16:00:03 pathayes: no, can't do that - can't label lexical spaces of the datatypes
16:00:28 literals as subjects could do this
16:00:36 (danc, pathayes)
16:00:55 bwm: out of formal meeting time
16:01:08 ok; I already sent "where did S-B go?" mail.
16:01:09 ... ahve a document, try to use that as the basis for our datatypiung solutoion
16:01:19 bwm: and moving fwd
16:01:26 ... can continue to discuss here
16:01:38 END OF MEETING
16:01:45 ----
16:01:53 datatypes chat:
16:02:54 patrick: literals as subject and untidy literals
16:03:16 two different properties prop1 and prop2 with two different ranges
16:03:19 p1 range decimalInteger
16:03:22 one is decimalInteger and other is octalInteger
16:03:25 p2 range octalInt.
16:03:28 x p1 "10".
16:03:32 y p2 "10".
16:04:10 - "10" is in lexical space of both datatypes (danc)
16:05:04 patrick: and an app that says give me triple and the property hanging of the literal node is going to g et the wrong value since the literal can be the wrong one - decmial or octal
16:05:25 danc: "10" is both an octal and a decimal numeral
16:05:55 patrick: but if you are doing a queyr and get back a literal node, and then ask for the peopreties of tat literal, you get both datatypes since literals are tidy
16:06:00 (am I recording this righ?)
16:06:17 pathayes: you want literals to denote their values?
16:06:32 ..l. wwould the datatyping barf if given this?
16:06:37 * DanCon isn't too worried about the record
16:06:47 * dajobe-scribe will slow down then
16:06:53 discussion
16:08:57 pathayes: if literals could be subjects...
16:08:57 JosD has quit
16:09:40 danc: handy to give our users names for lexical spaces, mapping , ...
16:09:52 bwm: feedback from users, too complex to explain?
16:10:17 patrick: someone going to ask xsd to do it?
16:10:19 danc: we can do it
16:10:53 I'd be quite happy to make up a name for "the lexical space of a (property used as a) datatype"
16:11:07 discsuion of sergey's proposal for union
16:11:25 sergey: also like danc's proposal
16:11:45 dc:date rdfs:range [ is lexicaSpace of xsdt:date ].
16:11:46 ... have the values and the lexical spaces
16:12:05 patrick: get rid of the value idiom?
16:12:57 wasn't that 4 months ago?
16:13:02 discussion
16:14:23 danc: let me build an expression for the property I want, need a lexicaSpaceof property
16:14:46 ... tricky since there are many "the ranges" for a datatype
16:15:42 danc: want to conclude "apple" is false
16:15:55 phahes: but no negation
16:16:03 danc: but datatypes give us this
16:17:21 discussion of rdf reasoners and datatypes
16:18:28 and application space
16:18:47 patrick: using the knowledge in rdf and doing more
16:20:31 * dajobe-scribe wonders what it means to implement RDF datatypes, as currently speciifed
16:21:37 to me, to implement datatypes is to parse a document and say "nope; that document is broken. it says 'apple' is in the range of the XML Schema date datatype"
16:22:32 so a document that generated "apple" (with datatype checks) would be illegal?
16:23:21 if that document also said dc:date rdfs:range xsdt:date.lex, then yes, it would be datatype-illegal
16:23:28 that's what I meant
16:23:51 datatype-illegal: hmm
16:29:59 discussion of just having xsd datatypes only
16:32:01 phayes: said, I think: the name for the lex space of the dc:date rdf:type of the range of the datatype
16:32:09 s/of the/is the/
16:32:23 ugh, try 2
16:32:34 phayes: said, I think: the name for the lex space of the dc:date rdf:type is the rdfs:range of the datatype
16:36:21 examples based on xsd:date
16:36:27 danc: says as a property; patrick as a class
16:37:03 used with a pair (subject, object), danc interested in the range - the lexical space
16:37:57 so can make a property [xsd:date xsd:range xyz]
16:38:06 and xyz can be used as the lex space of the datatype
16:38:07 (all danc)
16:38:23 jang: but no guarantee that it is just what you want
16:38:32 danc: want that guarantee
16:38:35 phayes: could od this
16:38:59 sergey: not sure range does this
16:39:20 phayes: could but an extra condition, when it is a datatype
16:39:26 sergey: ugly?
16:40:10 ... but sometimes want to use the whole range [not sure about recording this right]
16:41:55 * DanCon takes another call briefly.
16:42:03 * DanCon returns
16:43:31 does DatatypeSummary3 say what the class extension of xsdt:date is?
16:44:51 pathayes doc:
16:47:24 PatH has quit
16:47:44 * DanCon must run soon
16:47:56 several people want to rename rdf:value as used here for datatypes
16:49:19 telecon slot runs out in 10 mins I think
16:50:16 rdf:value and lform - opposite properties?
16:50:48 pathayes suggested ^
16:51:00 bwm: some m&s examples wouldn't fit that?
16:51:18 discussion of what to call new property(s)
16:51:30 in order to stay away from rdf:value
16:52:55 ciao, all.
16:58:07 have we invented anonymous predicates?
16:58:22 dajobe-scribe has quit
16:59:34 em has quit
16:59:49 * AaronSw suspects that telecon is over...
16:59:51 AaronSw has left #rdfcore
Received on
Saturday, 16 February 2002 12:58:44 UTC

Same domain → Similar titles →

C U Cyber History — Public Interest Web Archive

Preserving fading web memories. Discover history that once existed.