IRC log of rif-prd on 2008-10-07

Timestamps are in UTC.

16:58:13 [RRSAgent]

RRSAgent has joined #rif-prd

16:58:13 [RRSAgent]

logging to http://www.w3.org/2008/10/07-rif-prd-irc

16:58:33 [csma]

Meeting: RIF-PRD telecon Oct 7

16:58:39 [csma]

Chair: Christian de Sainte Marie

16:58:53 [Zakim]

SW_RIF(PRD)1:00PM has now started

16:58:55 [Zakim]

+??P2

16:59:00 [csma]

Agenda: http://lists.w3.org/Archives/Public/public-rif-wg/2008Oct/0005.html

16:59:43 [csma]

csma has changed the topic to: 7 Oct RIF-PRD telecon; agenda: http://lists.w3.org/Archives/Public/public-rif-wg/2008Oct/0005.html

17:00:31 [Zakim]

+??P3

17:00:32 [Zakim]

-??P3

17:00:32 [Zakim]

+??P3

17:00:52 [ChanghaiKe]

ChanghaiKe has joined #rif-prd

17:01:02 [csma]

zakim, ??p3 is me

17:01:02 [Zakim]

+csma; got it

17:01:12 [Zakim]

+Paul_Vincent

17:01:18 [csma]

zakim, ??p2 is ChanghaiKe

17:01:18 [Zakim]

+ChanghaiKe; got it

17:01:58 [Hassan]

Hassan has joined #rif-prd

17:02:01 [PaulVincent]

PaulVincent has joined #rif-prd

17:02:31 [csma]

agenda+ Organisation (telecon day/time, etc)

17:02:48 [PaulVincent]

scibenick PaulVincent

17:02:48 [csma]

agenda+ Actions in PRD

17:02:59 [Zakim]

+StuartTaylor

17:03:08 [PaulVincent]

scribenick: PaulVincent

17:03:14 [csma]

agenda+ Principles, scope and priorities

17:03:24 [Zakim]

+Gary

17:03:30 [csma]

agenda+ AOB

17:03:39 [Gary]

Gary has joined #rif-prd

17:04:22 [csma]

zakim, who is on the phone?

17:04:22 [Zakim]

On the phone I see ChanghaiKe, csma, Paul_Vincent, StuartTaylor, Gary

17:05:28 [PaulVincent]

scribe: PaulVincent

17:06:25 [Hassan]

mute me

17:06:52 [csma]

ack st

17:07:03 [csma]

zakim, stuarttaylor is hassan

17:07:03 [Zakim]

+hassan; got it

17:07:34 [csma]

next item

17:08:15 [PaulVincent]

CSMA: check if dayofweek and timeofday is OK: any complaints

17:08:31 [PaulVincent]

CK: OK if we keep to 1hr

17:09:54 [csma]

next item

17:10:00 [PaulVincent]

CSMA: keep DOWeek and TODay

17:11:11 [PaulVincent]

CK: is this bind - need strawman

17:11:24 [ChanghaiKe]

do is left as it is, we use bind instead to bind new variable to a new frame

17:11:35 [PaulVincent]

Action: GH to propose bind

17:11:57 [Gary]

did anybody get a chance to read http://www.w3.org/2005/rules/wiki/PRD#Operational_semantics_of_actions

17:12:47 [PaulVincent]

CSMA: other actions include modify, execute, ...

17:13:02 [Hassan]

q+

17:13:09 [Gary]

q+

17:13:10 [csma]

ack has

17:14:32 [PaulVincent]

HAK: assign is finer grained than assert (cf Java)

17:14:49 [ChanghaiKe]

q+

17:14:52 [PaulVincent]

HAK: remove and add object in assert/retract is not atomic

17:15:05 [PaulVincent]

GH: assert can also be at the slot level

17:15:10 [csma]

(bind ?x object(attr->?x) do(assign ?x value))

17:15:32 [csma]

do(assert(object(attr->Value)))

17:16:46 [Gary]

clarification - assert is only at the slot level

17:17:12 [ChanghaiKe]

what does this mean? do we assert a whole object?

17:17:15 [Gary]

- or you can assert a predicate

17:17:28 [csma]

Assert(?o[attr->value])

17:18:31 [Gary]

all - please look at http://www.w3.org/2005/rules/wiki/PRD#Operational_semantics_of_actions

17:18:58 [csma]

Forall ?x (?x#FooBor) If condition then assert(?x[attr->value)

17:20:10 [PaulVincent]

HAK: this is nonstandard notation

17:20:41 [csma]

Assert(frame|atom)

17:20:51 [csma]

New Var Class

17:22:27 [Hassan]

q+

17:23:12 [PaulVincent]

CSMA: proposed to use new for new objects and assert for updates

17:23:31 [csma]

ack gary

17:24:00 [PaulVincent]

GH: see PRD spec draft http://www.w3.org/2005/rules/wiki/PRD#Operational_semantics_of_actions

17:24:18 [PaulVincent]

CSMA: aside: looking at doing same for conditions

17:26:00 [csma]

q?

17:26:01 [PaulVincent]

GH: trying to combine operational and model theory

17:27:16 [PaulVincent]

HAK: this has not been done before so id of concern

17:28:41 [Hassan]

q-

17:28:43 [PaulVincent]

HAK: overloading of assert is dangerous

17:30:07 [PaulVincent]

HAK: PR engines assert objects not predicates

17:30:33 [PaulVincent]

CSMA: assert is for facts (as is retract) in PR engines...

17:31:40 [csma]

ack changh

17:32:37 [PaulVincent]

CK: difference between propositional logic vs object logic

17:33:46 [PaulVincent]

CK: assert into WM is the whole object - but this could be a bunch of triples in a single transaction

17:34:09 [Hassan]

yes CK - this is what I meant by granularity

17:36:10 [PaulVincent]

GH: we don't have any semantics for object based behavior

17:36:35 [PaulVincent]

GH: ... need to reuse frames etc from core

17:37:18 [Hassan]

q+

17:37:27 [PaulVincent]

Main issue if translating OUT OF PRD ie recombining frames etc into object operations

17:37:45 [PaulVincent]

CSMA: issue is one of serialization to map from objects to frames

17:37:59 [csma]

ack hassan

17:38:40 [PaulVincent]

HAK: issue is we need to be clear: choice is to assert triples that correspond to obj fields

17:40:08 [PaulVincent]

CSMA: Example: JRules asserts an object with field values: serialized into XML means all values?

17:40:30 [PaulVincent]

HAK: ... yes this is a series of asserts

17:41:29 [ChanghaiKe]

JRules asserts an object at once

17:41:36 [PaulVincent]

CSMA: Also need to avoid ambiguity

17:42:58 [PaulVincent]

HAK: why no remove

17:43:10 [PaulVincent]

GH: ... this is retract

17:43:19 [csma]

q?

17:43:21 [PaulVincent]

CSMA: which was an issue for MK

17:45:34 [Hassan]

can you post it?

17:45:36 [PaulVincent]

GH: See example 2.1.1 in http://www.w3.org/2005/rules/wiki/PRD

17:46:16 [PaulVincent]

Action: CK to create an assert and assign object from JRules

17:47:53 [Gary]

s/2.1.1/2.11

17:47:57 [PaulVincent]

CSMA: needs to be atomic or not?

17:48:26 [PaulVincent]

HAK: issue is of event in between the assert operations - atomic means its locked for the transaction...

17:48:38 [PaulVincent]

+1 to HAK's concern

17:48:48 [ChanghaiKe]

+1 to HAK concern

17:49:37 [PaulVincent]

CSMA: Is there a semantic issue here ie can an engine have an interrupt in an retract-assert vs modify

17:50:01 [ChanghaiKe]

I can give an example

17:50:04 [PaulVincent]

GH: does this apply for fine-grain assert/retract

17:50:50 [PaulVincent]

CK: example: truth maintenance system

17:50:55 [ChanghaiKe]

if temp > 120 then maintain alarm on network element

17:51:24 [PaulVincent]

CK: if I remove temp then TMS would deduce larm not to be maintained

17:51:53 [PaulVincent]

Of course this is the engine semantics, not the language ...

17:53:46 [PaulVincent]

CSMA: remove value not object - is this still the same?

17:54:03 [PaulVincent]

CK: ... but I can't remove the sensor (ie a.temp)

17:55:08 [PaulVincent]

GH: retract ([predicate|frame|object])

17:55:39 [csma]

?x: Device if ?x[value > 120] then assert(?x[alarm->True])

17:56:38 [PaulVincent]

Do we need to prefix/postfix assert/retract ops with a marker eg external(object operation)

17:56:44 [Hassan]

+1 with CK

17:56:46 [csma]

?x: Device if ?x[value > 240] and ?x[alarm->True] then Modify(?x[alarm->critical]

17:57:16 [PaulVincent]

Note to CSMA: we are 5mins to end of call...

17:59:09 [PaulVincent]

CK: paper exists that explains differences ?

17:59:29 [PaulVincent]

Action: CK to find appropriate paper reference or example

18:01:03 [ChanghaiKe]

http://blog.athico.com/2008/02/shadow-facts-what-you-always-wanted-to.html

18:02:01 [Zakim]

-hassan

18:02:03 [Zakim]

-Gary

18:02:16 [Zakim]

-ChanghaiKe

18:02:35 [Zakim]

-Paul_Vincent

18:02:36 [Zakim]

-csma

18:02:37 [Zakim]

SW_RIF(PRD)1:00PM has ended

18:02:39 [Zakim]

Attendees were csma, Paul_Vincent, ChanghaiKe, Gary, hassan

18:02:54 [csma]

rrsagent, make log public

18:03:01 [csma]

rrsagent, make minutes

18:03:01 [RRSAgent]

I have made the request to generate http://www.w3.org/2008/10/07-rif-prd-minutes.html csma

18:06:16 [csma]

csma has left #rif-prd