paw/cwm dev meeting

28 Sep 2005

Agenda

Attendees

Present
Jordan, DK, vlad, TimBL, DanC
Regrets
Yosi
Chair
timbl
Scribe
jordan, DanC

Contents


 

 

Convene, schedule next meeting

<DanC> Scribe: jordan

<DanC> Date: 28 Sep 2005

Next meeting: 12 Oct?

next meeting on 12 of october - is everyone jhappy

fine by everyone present - next meeting available oct 12

Action Review

YK: bijan in thesis-writing mode

<timbl> Bijan not available for meetings just now.. no win california... action continues

<DanC> ACTION: YK to schedule a meeting with TimBL and Bijan to discuss n3 semantics [CONTINUES] [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action01]

DanC: corresponding with paolo on CWM intefecence web

<DanC> Cc: paw-team@lists.mindswap.org <paw-team@lists.mindswap.org>, pp@ksl.stanford.edu

<DanC> Subject: [Paw-team] cwm inferenceweb registration

<DanC> Date: Wed, 14 Sep 2005 13:07:40 -0500

inference web*

action transfers to danc...

<DanC> ACTION: dlm to get cwm generating PML that can be seen in iwbrowser, with timbl [WITHDRAWN] [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action02]

<DanC> ACTION: danc to get cwm generating PML that can be seen in iwbrowser, with paolo, timbl [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action03]

<jordan> Vlad does not have acct on w3C cvs

<jordan> Action for TimBL to find member account

<DanC> http://cgi.w3.org/MemberAccess/AccessRequest

<jordan> Note: none of the mindlab members have w3c cvs acct - but if we identify ourselves as such we will be authorized

<jordan> Vlad working on getting an account using Tim's instructions

<DanC> the /MemberAccess/ thing will get http access, but timbl has to take further action to do cvs auth

<jordan> Vlad filled application for w3c cvs acct - noted he's affiliated wtih mindswap

<jordan> Tim requests vlad's public key and username when received back from webmaster

<jordan> ACTION: Tim to get Vlad cwm CVS access [CONTINUES] [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action04]

Round the table updates

Jordan: checked why some test case fails -- fixed and will be checked in tonight (problem due to change in code to have proof tracing )
... almost done

<scribe> ACTION: check nodetests.py to Yarden [CONTINUES] [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action05]

<DanC> (title of the math course? sounds interesting)

<jordan> intro to model theory/logic

<DanC> how about a #?

<DanC> i.e. so I could look it up in a umd catalog?

<jordan> math712

<jordan> google David Kueker

<DanC> thanks

<vkolovski> (i can continue scribing)

<scribe> ScribeNick: vkolovski

<jordan> -jordan

DK: talked with Uche's brother... [something about sparql? I didn't catch much of it]
... started a wiki document with pointers to all of the APIs

<jordan> RDFInterfaces.txt

<DanC> (re api wiki pages, see also http://esw.w3.org/topic/WebDataInterfaceDesign )

DK: Worked on rdflib's Sparql

vlad: Continue on Pychinko... created a lot of unitests... noticed some performance problems
... Works fine with builtins now

<DanC> (can I get a commit log somewhere near http://svn.mindswap.org/pychinko/ ?)

<eikeon_> DanC, try: http://websvn.mindswap.org/

<DanC> thanks... http://websvn.mindswap.org/listing.php?repname=Mindswap%20Main&path=%2Fpychinko%2Fpychinko%2F&rev=0&sc=0

<DanC> what's an example of a file with a unit test that you wrote, vlad?

<vkolovski_> { ?x :hasValue ?valX. ?y :hasValue ?valY.

<vkolovski_> (?valX ?valY) math:sum ?z.

<vkolovski_> ("http://" ?x " + " ?y " = " ?z ) string:concatenation ?s

<vkolovski_> } => { ?s a :RESULT }.

<DanC> -- pychinko/ruletests.py

<scribe> ACTION: VK to work with DanK to enhance rdflib's n-triples parser to have a mode that allows literals as subjects [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action06]

DanC: I set up demo.policyawareweb.org, went to TAG meeting

<DanC> changePolicy.n3 and TAG

<DanC> terminology for languages/documents/versions.

<DanC> http://www.w3.org/2000/10/swap/util/changePolicy.n3

<DanC> http://www.w3.org/2001/tag/2005/09/22-diagram1.png

<DanC> (also started looking at http://hathawaymix.org/Software/TCPWatch for the proxy part of the demo, but didn't get very far)

DanC: when is the deadline for the policy demo?

DanK: sometime in Oct.

DK: Demo meetings are happening, just DK and LK last week

DanK: there are demo telecons, but only me and Lalana were present for the last one

TimBL: is everyone here on paw-team?

we're all subcribed to it

<DanC> paw-team roster

log:includes

<vkolovski_> { :a :b :c. } log:includes { :a :b :c. { :a1 :a2 :a3} }

TimBL: indeed, that's true. the nested { :a1 :a2 :a3} is the subject of a statement with zero properties

<timbl> { :a :b :c. } log:includes { :a :b :c. { :a1 :a2 :a3} :b1 :c1 }

TBL: this longer statement is not true.

<timbl> { :a :b :c. } log:includes { :a :b :c. "a1 b1 c1" }

TimBL: when you have a zero property subject, you're not asserting it's true

<DanC> you're just saying it exists.

<timbl> (1 2 3 ( 4 5 "dftgyhujk")).

<timbl> [ mother anne ].

<timbl> [ mother anne ] father fred.

RESOLUTION: We should have a warning if it is not a square bracket thing and it has no data.

<DanC> "warning: existence of literal is rarely what you meant"

<timbl> notation3.py warning

<DanC> I'm mailing

TimBL: let's move this issue to cwm-bugs, it's not on the critical path for this meeting

<DanC> RFE: warn about zero predicates with anytyhing other than []s

<vkolovski_> {

<vkolovski_> { :fred a :Person.

<vkolovski_> {?x a :Person} => { ?x :hasFather [a :Person] } } log:conclusion

<vkolovski_> [ log:notIncludes { :Vlad :knows :Jim } ]

<vkolovski_> } log:implies {:test :a :success}.

<timbl> I think there's a log:notSupports

<DanC> log:conclusion, log:notIncludes, and infinite formulas

<DanC> (I'm curious about the claim that cwm gives false answers)

<DanC> (VK misspoke; it didn't terminate.)

TimBL: should we support --datalog in cwm?

<DanC> (seems like a reasonable idea, but I don't see any specific benefit that motivates me to write the RFE)

<DanC> Scribe: DanC

H. Nary operations and built-ins

TimBL: cwm uses ... (1 2) math:sum ?X.
... but others use named things... [ :arg1 1; :arg2 2 ] math:sum ?X. # [er.. ]
... cf semweb bpd wg note, rules charter

F. RDFlib fork

TBL: what agreement did we come to last week?

DK: to join the forks, I think...
... difference is [missed]
... and yes, to cross-link from rdflib.org and mindswap
... how?

TBL: on rdflib, "rdflib is supported by mindlab/mindswap". on mindswap, "projects include rdflib"

<vkolovski> TBL: ACTION DanK to cross-link rdflib/mindlab sites

<DanC_> I see Redfoot is already on http://www.mindswap.org/projects/

A. the cwm-pychinko integration break

DanC: do you know where --why is needed in the demo?

TimBL: I've seen where log:includes and log:semantics fit into REIN...

Subject: [Paw-team] PAW demo telecon today

Date: Tue, 20 Sep 2005 13:43:25 -040

[[

A Rein policy example for the girlscout troop is up at

http://groups.csail.mit.edu/dig/2005/09/rein/examples/

]]

<vkolovski> wm <input> <request> --n3 http://dig.csail.mit.edu/2005/09/rein/engine.n3 --think

<vkolovski> --filter="http://dig.csail.mit.edu/2005/09/rein/filter.n3" <output>

<DanC> pbjam:~/paw/www/2005/rein connolly$ python2.4 ~/w3ccvs/WWW/2000/10/swap/cwm.py requestE.n3 --chatty=20 --think --filter=engine.n3 >,x

<scribe> ACTION: TimBL to work on --why toward log:includes, log:conclusion for rein [CONTINUES] [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action07]

<timbl> http://www.w3.org/2000/10/swap/doc/plans.html updated

plans.html,v 1.37 2005/09/28 19:34:16

DanC: can you run the girl scout REIN example with pychinko?

VK: no; missing log:includes, log:semantics
... should we do quoting?

TBL: or maybe quads.

<timbl> K: F includes G

<timbl> G: f1 f2 f3

<timbl> F g1 g2 g3

<timbl> F: g4 g5 g6

<timbl> K: F type TestCase

vkolovski: we have no Formula datatype in pychinko

<timbl> First, develop Formula as a datatype.

<timbl> Second, develop log:notiIncldues as a python builtin

<timbl> Second, develop log:notiIncldues as a builtin implemented in python

<DanC_> (log:semantics is pretty much like math:sum, but with memoization/cache)

<timbl> Then look at ways of doing log:incldues

<vkolovski> no nested lists in pychinko now

<timbl> Zeroth, implement lists as a datatype

"RESOLUTION: all paw-team telcon Fri 28 Oct 10amET postscript: perhaps 1.5hrs." -- http://www.policyawareweb.org/2005/ftf2/paw-mtg

http://www.policyawareweb.org/2005/ftf2/paw-mtg-cal.ics

Summary of Action Items

[NEW] ACTION: danc to get cwm generating PML that can be seen in iwbrowser, with paolo, timbl [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action03]
[NEW] ACTION: VK to work with DanK to enhance rdflib's n-triples parser to have a mode that allows literals as subjects [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action06]
 
[PENDING] ACTION: check nodetests.py to Yarden [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action05]
[PENDING] ACTION: Tim to get Vlad cwm CVS access [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action04]
[PENDING] ACTION: TimBL to work on --why toward log:includes, log:conclusion for rein [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action07]
[PENDING] ACTION: YK to schedule a meeting with TimBL and Bijan to discuss n3 semantics [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action01]
 
[DROPPED] ACTION: dlm to get cwm generating PML that can be seen in iwbrowser, with timbl [recorded in http://www.policyawareweb.org/2005/pf-dev/09-28-paw-minutes#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2005/08/16 15:12:03 $