IRC log of rif on 2008-11-17

Timestamps are in UTC.

15:59:28 [RRSAgent]
RRSAgent has joined #rif
15:59:28 [RRSAgent]
logging to http://www.w3.org/2008/11/17-rif-irc
16:01:33 [DaveReynolds]
DaveReynolds has joined #rif
16:02:32 [Zakim]
SW_RIF(CORE)11:00AM has now started
16:02:36 [Zakim]
+[IBM]
16:02:39 [ChrisW]
zakim, ibm is me
16:02:40 [Zakim]
+??P12
16:02:44 [Zakim]
-??P12
16:02:46 [Zakim]
+??P12
16:02:50 [Zakim]
+ChrisW; got it
16:04:38 [Zakim]
+??P15
16:05:04 [ChrisW]
zakim, ??P15 is AdrianP
16:05:04 [Zakim]
+AdrianP; got it
16:05:54 [Zakim]
+kifer
16:06:17 [Michael_Kifer]
Michael_Kifer has joined #rif
16:08:49 [Zakim]
+Gary
16:09:01 [AxelPolleres]
AxelPolleres has joined #rif
16:09:06 [Gary_Hallmark]
Gary_Hallmark has joined #rif
16:09:08 [AxelPolleres]
moment... dialing
16:10:10 [Zakim]
+??P30
16:10:52 [DaveReynolds]
topic: discussion re #/##
16:11:08 [DaveReynolds]
Gary: needs to be known at translation time the class of every object
16:11:30 [ChrisW]
ChrisW has joined #rif
16:11:42 [ChrisW]
zakim, who is here?
16:11:42 [Zakim]
On the phone I see ChrisW, DaveReynolds, AdrianP, kifer, Gary, AxelPolleres
16:11:44 [Zakim]
On IRC I see ChrisW, Gary, AxelPolleres, Michael_Kifer, DaveReynolds, RRSAgent, Zakim, sandro, trackbot
16:12:05 [DaveReynolds]
Gary: or allow anywhere but have some compliance exception for PRD
16:14:19 [DaveReynolds]
Michael: Patrick wanted to restrict to creation but that can't be done within CORE
16:15:59 [Harold]
Harold has joined #rif
16:16:56 [Zakim]
+Harold
16:18:45 [DaveReynolds]
Dave: could we leave it as currently resolved and cover data models in a future extension?
16:19:31 [DaveReynolds]
Gary: but then it makes little sense to have # at all, because we then can't define it. Better to have externally defined.
16:20:45 [DaveReynolds]
Chrisw: It means the class memberships are not defined in rules but defined in some out of band object system and # is just a way to test that in rules.
16:20:55 [DaveReynolds]
Gary: so should # be wrapped in an External wrapper?
16:23:55 [DaveReynolds]
Dave: two clean options (a) unrestricted, (b) drop completely for FPWD
16:24:30 [DaveReynolds]
Gary: ree: b but many production systems require membership in patterns and so no PR rule would be a core rule.
16:24:35 [DaveReynolds]
s/ree/re/
16:25:24 [DaveReynolds]
Gary: would rather leave it in unrestricted.
16:26:22 [DaveReynolds]
Gary: to be discussed in PRD telecon
16:26:39 [DaveReynolds]
ScribeNick: DaveReynolds
16:29:25 [DaveReynolds]
Gary: PRD has special syntax for object creation which is where # can be asserted in PRD.
16:29:45 [DaveReynolds]
Chris: so does a # in a fact in Core correspond to an object creation?
16:33:53 [DaveReynolds]
Gary: hopes that if allow # in facts in Core then static analysis can capture duplicate assertions and trap attempts to reassign membership at translation time
16:39:21 [DaveReynolds]
Gary: is we restrict to facts and conditions is that OK for core?
16:39:35 [DaveReynolds]
Michael: yes
16:39:58 [DaveReynolds]
Harold: c.f. abox in OWL
16:41:12 [DaveReynolds]
Dave: what about subclass, is that needed?
16:41:34 [DaveReynolds]
Gary: yes, to represent e.g. complex type hierarchy in XML Schema
16:42:43 [Harold]
Using #,facts to define class membership is a good match for Aboxes in OWL DL.
16:43:10 [Harold]
s/#,/ #/
16:43:58 [DaveReynolds]
Chris: what was the issue csma raised with pattern matching?
16:45:00 [DaveReynolds]
Michael: only diff between unification and matching, when no function symbols, there is no difference unless there are variables in facts and safety rules that out
16:46:13 [DaveReynolds]
Gary: isn't about external functions in conclusions? Not unification but about the need to rewrite a rule to handle builtins.
16:46:55 [Harold]
Gary, do you mean something like unifying +(2,3) with 5?
16:47:15 [Harold]
So, not just syntactic unifcication?
16:48:19 [Gary]
P(x) :- P(x - 1) vs. P(x + 1) :- P(x)
16:48:22 [sandro]
sandro has joined #rif
16:48:31 [DaveReynolds]
Dave: what if restrict builtin fns to just equality?
16:48:47 [AxelPolleres]
not sure, whether that was what christian meant.
16:49:24 [AxelPolleres]
+1 to discuss that with christian tomorrow again.
16:49:31 [DaveReynolds]
Need clarification from Christian on what he meant
16:50:25 [DaveReynolds]
Dave: raise in plenary tomorrow?
16:50:58 [Harold]
+1 (even in the plenary)
16:51:34 [DaveReynolds]
Michael: issues (a) #/## (b) external functions (c) shorten compliance section (d) EBNF needs fixing
16:55:06 [Harold]
s/safety /safeness/
16:55:12 [DaveReynolds]
Action: dave update EBNF to match BLD
16:55:12 [trackbot]
Created ACTION-655 - Update EBNF to match BLD [on Dave Reynolds - due 2008-11-24].
16:56:40 [AxelPolleres]
We still have "safety" in some places, don't we?
16:57:33 [Harold]
Yes, should be changed.
16:58:05 [Harold]
Is it realistic to have a freezable version before next Monday?
16:58:07 [DaveReynolds]
Michael: can look at compliance section to see how to reformulate
16:59:18 [DaveReynolds]
Possibles topics for plenary: (i) #/## new restrictions (ii) Christian pattern match
17:00:22 [DaveReynolds]
Sorry - have to go
17:00:27 [Zakim]
-DaveReynolds
17:03:07 [AxelPolleres]
shall we adjourn?
17:03:14 [Zakim]
-Gary
17:03:36 [AxelPolleres]
rrsagent, make minutes
17:03:36 [RRSAgent]
I have made the request to generate http://www.w3.org/2008/11/17-rif-minutes.html AxelPolleres
17:04:11 [Harold]
PRD colleagues, could you pleas briefly bring up if http://www.w3.org/2005/rules/wiki/Core#Intersection_of_RIF-PRD_and_RIF-BLD is fine for the FPWD of Core?
17:04:16 [Zakim]
-AdrianP
17:04:38 [Michael_Kifer]
rrsagent, make public
17:04:38 [RRSAgent]
I'm logging. I don't understand 'make public', Michael_Kifer. Try /msg RRSAgent help
17:05:12 [Zakim]
-AxelPolleres
17:05:14 [Zakim]
-kifer
17:05:36 [Harold]
RRSAgent, make logs public
17:05:57 [Harold]
RRSAgent, make minutes
17:05:57 [RRSAgent]
I have made the request to generate http://www.w3.org/2008/11/17-rif-minutes.html Harold
17:07:12 [Zakim]
-ChrisW
17:07:13 [Zakim]
-Harold
17:07:14 [Zakim]
Attendees were ChrisW, DaveReynolds, AdrianP, kifer, Gary, AxelPolleres, Harold
17:07:34 [ChrisW]
rrsagent, make minites
17:07:34 [RRSAgent]
I'm logging. I don't understand 'make minites', ChrisW. Try /msg RRSAgent help
17:07:37 [ChrisW]
rrsagent, make minutes
17:07:37 [RRSAgent]
I have made the request to generate http://www.w3.org/2008/11/17-rif-minutes.html ChrisW
18:19:46 [Zakim]
Zakim has left #rif
20:01:38 [StellaMitchell]
StellaMitchell has joined #rif
20:01:41 [sandro]
hey
20:01:49 [sandro]
SWC#Annotation_properties
20:01:59 [sandro]
<a href="....">SWC#Annotation_properties</a>