IRC log of rif on 2008-02-22

Timestamps are in UTC.

13:27:51 [RRSAgent]

RRSAgent has joined #rif

13:27:51 [RRSAgent]

logging to http://www.w3.org/2008/02/22-rif-irc

13:28:06 [Hassan]

Discussing forward compatibility of BLD - what should implementers need to know

13:29:14 [Hassan]

the "traditional" way is to ignore what they need and provide extensions they need

13:30:29 [Hassan]

"features", such as equality and named args, would be specified per variation of a dialect departing from BLD

13:31:26 [Hassan]

XML annotations tags could be used to specified the features in a "a la carte" fashion to create an extension of BLD

13:31:44 [Hassan]

s/annotations/annotation/

13:35:28 [Hassan]

If a feature is not implemented, some transformation is applied and comments/tags included to mention that fact (e.g., if named args are not implemented, they are transformed as positional and comments are generated saying so

13:37:04 [josb_]

josb_ has joined #rif

13:37:13 [Hassan]

s/saying so/saying so)/

13:37:33 [Hassan]

s/tags /tags are /

13:39:37 [Hassan]

Sandro keeps on describing his transformation scheme to adapt missing features into others supported

13:40:58 [Hassan]

This is "poor man" fallback mechanism could be used to specify how to adapt missing features using existing ones

13:41:14 [Hassan]

s/This is/This/

13:41:28 [AxelPolleres]

you could even use xquery... more convenient than xslt

13:43:06 [Hassan]

This mechanism could be a preprocessor or a filter

13:45:19 [Hassan]

ChrisW: will this part of the specs?

13:45:36 [Hassan]

Sandro: non-normative, perhaps...

13:47:45 [Hassan]

CSMA: The specs should define RIF Core so as anything that is not in it be rejected (using Sandro's device)

13:49:09 [Hassan]

Questions: what does 80% sound mean?

13:49:40 [Hassan]

Sandro: this is based on the level of severity (i.e., ignore, warning, error)

13:50:30 [Hassan]

Jos: then this should not be soundness but completeness

13:51:15 [Hassan]

Sandro: the mechanism is there to allow degrees of severity

13:52:00 [DaveReynolds]

One example "partial soundness" might be in terms of error behaviour - well typed data leads to the same results but ill-typed data may be let through.

13:52:41 [Hassan]

Sandro: the intention is to "minimize" some overall severity level by composing them

13:53:58 [Hassan]

Gary: I doubt that BR people would be happy to see their rules changed behind their back - they would rather do this sort of things themselves - not the machine

13:55:59 [Hassan]

Sandro: the spec say what features should or must be there; and some graceful fallback mechanism

13:58:40 [Harold4]

I think this is an interesting start for RIF Interoperability Technology.

13:59:22 [Hassan]

ChrisW: So what do we do?

13:59:59 [Hassan]

Sandro: XTAN (XML Transform As Needed) can be a tool on its own (not just for RIF)

14:00:50 [Hassan]

Sandro: I.e., a specific document should describe it

14:01:34 [ChrisW]

zakim, pick a victim

14:01:34 [Zakim]

Not knowing who is chairing or who scribed recently, I propose ??P0

14:01:40 [ChrisW]

zakim, pick a victim

14:01:40 [Zakim]

Not knowing who is chairing or who scribed recently, I propose ??P0

14:02:00 [AxelPolleres]

26-28 of may

14:02:52 [johnhall]

scribenick: johnhall

14:05:40 [AxelPolleres]

scribenick: axel

14:05:49 [AxelPolleres]

scribenick: AxelPolleres

14:06:10 [AxelPolleres]

Harold: finitelty nested functions can be translated to Datalog.

14:06:56 [AxelPolleres]

... if nesting depth can be limited.

14:07:26 [AxelPolleres]

(we have a corporate rate of 89 EUR from NUIG at westwood hotel BTW, for f2f10)

14:07:35 [sandro]

Atom + Expr => UniTerm

14:08:39 [AxelPolleres]

jos/michaelK: discussion naming uniterms vs atomic formulae

14:09:27 [AxelPolleres]

I am lost!!!!

14:09:57 [AxelPolleres]

3 strands of discussion going on.

14:10:30 [AxelPolleres]

jos: problems with functions predicates uniterms.

14:10:45 [AxelPolleres]

josb/sandro: make explicit in the syntax

14:11:00 [AxelPolleres]

christian: context not enough to decide?

14:11:55 [sandro]

Sandro: It's much easier to write "I don't support logic function" than "I don't support Uniterms within Uniterms".

14:12:20 [AxelPolleres]

michealK: I find it srange just to make changes just to make it easier to implement a certain tool.

14:12:37 [AxelPolleres]

josb: it is confusing because all is a term.

14:13:07 [AxelPolleres]

chrisw: it can stil look the same but we can call the syntactic term

14:13:21 [AxelPolleres]

... in the syntax tree different

14:13:27 [AxelPolleres]

s/term/construct

14:14:14 [AxelPolleres]

Harold: we have this since years and now and it passed many reviews.

14:14:29 [AxelPolleres]

... will not chang it because one python tool doesn't work.

14:14:59 [AxelPolleres]

... prolog and f-logic tools like Flora have this implemented.

14:15:21 [AxelPolleres]

jos: that's not the point. a pred is a pred a func is a func.

14:15:47 [sandro]

Chris: This is changing the grammar, not the syntax.

14:16:05 [sandro]

Gary: Yes -- call a function a function and a term and term, rather than calling them both Uniterm and getting confused.

14:16:08 [AxelPolleres]

axel: not the same for a prolog programmer who think in terms of functors

14:16:27 [AxelPolleres]

gary, chrisw: doesn't affect the language, just the grammar.

14:17:19 [AxelPolleres]

christian: we can make the move without loosing anything.

14:17:54 [AxelPolleres]

harold: FOL is too restrictive for the SW.

14:19:07 [AxelPolleres]

axel: the uniterm stuff also caused difficulties for the metamodel attempt which I did.

14:19:24 [AxelPolleres]

it shouldn't make a differnce to name them differently.

14:19:32 [sandro]

Jos: There is always a difference. Atomic Formula are mapped to T/F; Terms are mapping to elements in the domain.

14:19:46 [AxelPolleres]

michaelK: it is basically a change of grammar only, I agree.

14:19:57 [AxelPolleres]

harold: you also change the mathematics.

14:20:16 [sandro]

s/mathematic/mathematical english/

14:20:19 [AxelPolleres]

miachaelK: slight chane in the english description and the grammar necessary, not more.

14:24:19 [AxelPolleres]

ChrisW going to paint something on the flipchart.

14:24:58 [AxelPolleres]

AdrianP: in favor of uniterm.

14:25:30 [AxelPolleres]

AdrianP: uniterm more clear than "function"

14:25:46 [AxelPolleres]

ChrisW: there was a change from compaund to uniterm yesterday

14:25:52 [AxelPolleres]

harold: ... as it used to be.

14:26:47 [AxelPolleres]

christian: no, COMPOUND stays there.

14:27:12 [AxelPolleres]

christian: we replace compound by uniterm.

14:27:23 [AxelPolleres]

... in terms

14:27:47 [AxelPolleres]

ChrisW: ho to distinguish funcrtions from predicates.

14:28:33 [AxelPolleres]

jos running faster than gary to write the solution on the whiteboard

14:30:05 [apaschke]

as an example: typical directives and built-ins in the ISO Prolog STANDARD such as assert(p(X)), findall(X,p(X),List), derive(p(X))

14:30:47 [apaschke]

although we don't support them yet in RIF, a Prolog implementer would need them to write meaningful rule programs

14:31:19 [apaschke]

How do we serialize p(X), with which constructs? , clearly p(X) is a term here, but should we really serialize it as a function

14:32:27 [sandro]

On paper: ATOM := Pred '(' TERM* | -> ')'

14:32:48 [sandro]

On paper: Atom := Pred '(' TERM* | -> ')'

14:32:59 [sandro]

On paper: Expr := Fun '(' TERM* | -> ')'

14:33:16 [EtnaRosso]

EtnaRosso has joined #rif

14:33:42 [EtnaRosso]

EtnaRosso has left #rif

14:34:16 [AxelPolleres]

Harold: In the first section of BLD we often talk about Terms, where uniterms are referred.

14:34:24 [AxelPolleres]

... we need to change it.

14:34:54 [AxelPolleres]

josb: not one word needs to change.

14:36:30 [AxelPolleres]

Gary: we didn't get rid of uniterm, it is still there, but with two concrete distinctions.

14:37:05 [sandro]

Jos and others: Assert/Retract are connectives/operators NOT predicates

14:37:18 [sandro]

... so it's okay that their arguments are Atom

14:38:09 [AxelPolleres]

MichaelK: Ideally we should have one gorund for the BNF and one for the XML.

14:38:39 [AxelPolleres]

sandro: one schema for all RIF? and then specializations for BLD, Core, PRD, etc.?

14:39:17 [AxelPolleres]

MichaelK: we need to work it out, don't know if it is possible. but maybe to early for the BNF to change it now.

14:39:39 [josb]

josb has joined #rif

14:39:41 [AxelPolleres]

christian: if we start something that's shaky now, we put WD2 at risk

14:40:34 [AxelPolleres]

sandro: I undserstand it is a lot of work.

14:40:40 [AxelPolleres]

jos: i don't

14:41:25 [AxelPolleres]

chrisw: it is not about that change, but on whether int terms of a more severe change on the framework, it is useful to change the bnf now, which will change anyway later on for the framework.

14:41:59 [AxelPolleres]

michaelk: let's wait for the framework to change the bnf adhoc now.

14:42:01 [Harold4]

Through our Uniterms we can have practically important features such as higher-order syntactic sugar at no extra cost.

14:42:06 [Harold4]

Examples:

14:42:07 [Harold4]

product[R,S](X,Z) :- R(X,Y), S(Y,Z).

14:42:24 [AxelPolleres]

s/to change/instead of changing/

14:43:01 [Harold4]

The above is the product of two relations R and S.

14:43:21 [AxelPolleres]

sandro: framework is not critical path and a new XML syntax will take at least a day in f2f to get it through

14:43:31 [Harold4]

The relation is a term used as a predicate name.

14:43:35 [AxelPolleres]

... puts wd2 at risk.

14:44:00 [AxelPolleres]

josb: there is a clear distinction between atomic and functions in FLD document.

14:44:15 [Zakim]

-DaveReynolds

14:44:18 [sandro]

Ciao DaveReynolds

14:44:22 [AxelPolleres]

... it is very unlikely that we need to reverse anything if we do this change now.

14:44:58 [AxelPolleres]

harold explaining what he just pasted in the IRC.

14:45:35 [sandro]

Jos: product(R,S) is a Predicate Name, not a Function Name. The distinction remains.

14:46:02 [AdrianP]

and this restricted form of high order logic and also in many meta programming approaches the clear distinction between atomic formulas and functions breaks

14:46:04 [JeffP]

bye Dave

14:46:09 [AxelPolleres]

s/[R,S]/(R,S)/

14:46:30 [AdrianP]

a typical example: deafeasible logic

14:46:40 [AxelPolleres]

josb/MichaelK discussion how to specialize BNFs.

14:46:52 [AxelPolleres]

Axel: would be easy with an abstract model instead :-)

14:48:20 [AxelPolleres]

JosB: I can do the BNF change.

14:48:35 [AxelPolleres]

Harold: No, we could do it, but we need to think about it.

14:49:07 [Harold4]

In the example product(R,S) is a complex Predicate Name, consisting of a function name, product, applied to to R and S.

14:49:16 [Zakim]

disconnecting the lone participant, ??P0, in Team_(rif)13:25Z

14:49:19 [Zakim]

Team_(rif)13:25Z has ended

14:49:20 [Zakim]

Attendees were DaveReynolds

14:49:23 [AxelPolleres]

MichaelK: I don't want to do it for the WD because there is no rush. Proposals should be made explicit and discused for the next draft.

14:49:49 [AxelPolleres]

Josb: will make a proposal for the BNF for telecon in a week and a half.

14:51:05 [sandro]

Sandro: I am going assume we'll do the uniterm split, and have XTAN not have to follow the grammar. If I hear otherwise, I sure hope it's soon, or XTAN is at risk.

14:51:16 [AxelPolleres]

ACTION: jdebruij2 to create a new BNF for FLD and BLD (distinguishing atoms and preds) by Monday 3rd of March

14:51:16 [trackbot-ng]

Created ACTION-438 - Create a new BNF for FLD and BLD (distinguishing atoms and preds) by Monday 3rd of March [on Jos de Bruijn - due 2008-02-29].

14:52:38 [Harold4]

Such a COMPLEX Predicate Name is not allowed in FOL, where Predicate Names must be simple (PCDATA-like) strings.

14:54:13 [AxelPolleres]

Topic: coffee break!

14:54:25 [AxelPolleres]

continue at 16:15

15:07:54 [josb_]

josb_ has joined #rif

15:14:36 [AxelPolleres]

Topic: TestCases

15:15:02 [sandro]

http://www.w3.org/2007/OWL/wiki/TC4

15:15:10 [JeffP_]

JeffP_ has joined #rif

15:15:36 [AxelPolleres]

scribnick: AdrianP

15:15:46 [AxelPolleres]

scribenick: AdrianP

15:16:00 [AxelPolleres]

scribe: Adrian Paschke

15:16:01 [AdrianP]

Topic: RIF Test Cases

15:17:23 [sandro]

http://www.w3.org/2005/rules/wg/wiki/Arch/Test_Cases

15:18:19 [AdrianP]

Sandro: shows test case on the WIKI

15:18:26 [sandro]

http://www.w3.org/2007/OWL/wiki/TC4a

15:21:37 [apaschke]

Sandro: There are some tools on the WIKI to describe/document test cases

15:22:15 [apaschke]

Jeff: We would need test cases to test our implementations

15:22:28 [apaschke]

Jeff: This is a general requirement for implementers of RIF

15:23:15 [josb_]

josb_ has joined #rif

15:23:48 [apaschke]

Harold: There are some test cases in the appendix of BLD specification

15:24:40 [AxelPolleres]

AxelPolleres has joined #rif

15:24:46 [Harold4]

http://www.jdrew.org/rif/wd2/RIFPosCondExample.rif

15:24:50 [apaschke]

Harold: follow the link to the online directory

15:25:04 [apaschke]

Harold: Only for testing the syntax

15:25:06 [Harold4]

http://www.jdrew.org/rif/wd2/RIFPosHornExample.rif

15:25:16 [csma2]

csma2 has joined #rif

15:25:25 [Harold4]

http://www.jdrew.org/rif/wd2/RIFSlotCondExample.rif

15:26:38 [Harold4]

The above are 3 test cases for XSD validation (with XSV), linked to from http://www.jdrew.org/rif/wd2, linked to from http://www.w3.org/2005/rules/wiki/BLD#Appendix:_Specification

15:26:46 [apaschke]

A use case applying test-driven development to safeguard rule interchange which I had proposed when we started RIF

15:26:48 [apaschke]

http://www.w3.org/2005/rules/wg/wiki/Rule_Interchange_Through_Test-Driven_Verification_and_Validation

15:27:23 [apaschke]

Garry: I want to test the RIF implementation

15:28:04 [apaschke]

Garry: Want to use a test case to test the received RIF rule sets

15:28:35 [apaschke]

Garry: Maybe you want to evaluate different RIF implementation with test cases

15:28:47 [apaschke]

Christian: I don't understand this

15:29:07 [apaschke]

Garry: I want to know which RIF compatible engines pass the most tests

15:29:18 [apaschke]

Christian: I see the point

15:29:56 [apaschke]

Christian: I had different use cases for test cases in mind

15:30:39 [apaschke]

Chris: Implementers will advertise the passed RIF compliance tests

15:30:53 [apaschke]

Christian: This is out of scope for us

15:31:29 [apaschke]

Garry: e.g. if we use the condition language as query language

15:32:09 [apaschke]

Chris: this still assumes that there will be implementers which use and really implement RIF allowing answers and queries

15:32:34 [apaschke]

Chris: implement query response RIF systems

15:32:54 [apaschke]

Garry: It would be nice to have query facility

15:33:09 [apaschke]

Garry: most engines have a query feature

15:33:52 [apaschke]

Jeff: In the test case you might have a set of rules, and with the condition language you can have some queries

15:34:08 [apaschke]

Jeff: and we need to provide the output as well

15:34:25 [apaschke]

Chris: this is not our job to provide an API

15:34:40 [apaschke]

Garry: I would like to get some output

15:34:45 [Harold4]

Gary, we could do this a la http://esw.w3.org/topic/SparqlEndpointDescription

15:35:13 [apaschke]

Harold: All what we need to do would be a something like a SPARQL endpoint

15:35:25 [apaschke]

Sandro: Why not integrate RIF into SPARQL?

15:35:52 [apaschke]

Axel: you can not query predicates

15:36:25 [AxelPolleres]

what about that?

15:36:25 [apaschke]

Jeff: We are writting some kind of parser

15:36:26 [AxelPolleres]

SELECT ?X

15:36:26 [AxelPolleres]

FROM ruleset

15:36:26 [AxelPolleres]

WHERE { RIF_CONDITION_WITH_FREE variable ?X }

15:36:43 [AxelPolleres]

and SPARQL-result format as output-format?

15:36:51 [apaschke]

Harold; Hassan has implemented 2 years ago

15:37:03 [Harold4]

(We could also use Rule Responder: http://responder.ruleml.org)

15:37:31 [apaschke]

RuleResponder is like a SPARQL endpoint for arbitraryrule inference services

15:38:02 [apaschke]

Axel: We can use the SPARQL query syntax to ask queries and send back answers

15:38:28 [apaschke]

Sandro: Where clause presentation syntax?

15:38:39 [apaschke]

Axel: presentation syntax

15:39:16 [apaschke]

Axel: I don't want to reinvent a new query syntax

15:39:35 [apaschke]

Axel: If we propose something why not close to SPARQL

15:40:11 [apaschke]

Garry: like in Prolog

15:40:26 [apaschke]

Garry: use a head less rule

15:40:47 [apaschke]

Sandro: We don't standardize the API

15:41:10 [apaschke]

Sandro: If you want to write an engine you can do that, but we don't standardize that

15:41:16 [sandro]

Sandro: folks can do presentation syntax and this query syntax in a WG note.

15:42:25 [apaschke]

Garry: right now without a standard we can not safely write queries and get responses, e.g. to apply test cases

15:42:47 [apaschke]

Michael: conditions can be used as queries

15:43:03 [apaschke]

Michael: rules without queries are useless

15:44:48 [apaschke]

Christian: What does it take for us to use the OWL template for test cases

15:44:54 [apaschke]

Sandro: Copy it and start using it

15:46:37 [apaschke]

Adrian: The current test cases are syntactical test cases

15:46:51 [apaschke]

Christian: Test cases for entailment testing

15:47:44 [apaschke]

Sandro: Look at the OWL example test case

15:48:26 [apaschke]

Harold: We have different kind of results

15:48:33 [sandro]

Premise: p(a->1 b->2)

15:48:35 [sandro]

Conclusion

15:48:45 [sandro]

Conclusion: p(b->2 a->1)

15:49:25 [apaschke]

We would need a query

15:49:32 [apaschke]

Take a look at this use case:

15:49:33 [apaschke]

<apaschke> A use case applying test-driven development to safeguard rule interchange which I had proposed when we started RIF

15:49:49 [apaschke]

<apaschke> http://www.w3.org/2005/rules/wg/wiki/Rule_Interchange_Through_Test-Driven_Verification_and_Validation

15:50:13 [apaschke]

Christian: You ask your rule engine if the test is true

15:50:22 [apaschke]

Christian: this is out of scope

15:50:32 [apaschke]

Christian: of our working group

15:52:16 [apaschke]

Action; AdrianPaschke: add some test cases to the WIKI page

15:52:17 [csma2]

action: jeffp to post test cases for RIF

15:52:17 [trackbot-ng]

Sorry, couldn't find user - jeffp

15:52:43 [csma2]

action: jpan2 to post test cases fir RIF

15:52:43 [trackbot-ng]

Created ACTION-439 - Post test cases fir RIF [on Jeff Pan - due 2008-02-29].

15:52:55 [apaschke]

action: AdrianPaschke add some test cases to the WIKI page

15:52:55 [trackbot-ng]

Sorry, couldn't find user - AdrianPaschke

15:53:21 [Harold4]

There are five Examples in http://www.w3.org/2005/rules/wiki/BLD#XML_Serialization_for_the_Interchange_of_RIF-BLD and three Test Cases in http://www.jdrew.org/rif/wd2.

15:53:28 [apaschke]

action: adrian paschke add some test cases to the WIKI page

15:53:28 [trackbot-ng]

Sorry, amibiguous username (more than one match) - adrian

15:53:28 [trackbot-ng]

Try using a different identifier, such as family name or username (eg. agiurca, apaschke)

15:53:52 [apaschke]

action: apaschke add some test cases to the WIKI page about test cases

15:53:52 [trackbot-ng]

Created ACTION-440 - Add some test cases to the WIKI page about test cases [on Adrian Paschke - due 2008-02-29].

15:54:21 [apaschke]

Chris: topic "Metadata"

15:56:42 [apaschke]

Jos: explains extensions of the syntax for Rulesets

15:57:52 [apaschke]

Jos: controversial where Directives

15:59:33 [apaschke]

Chris: the valid directives are fixed and need to be implemented by a compliant engine

15:59:53 [apaschke]

Christian: We would need to add a lot of elements

16:00:05 [apaschke]

Christian: because we can not forsee all use cases

16:00:22 [sandro]

csma: Normally I would support the idea that the directive would be part of the syntax, since it cannot be ignored. But if they can be included in lots of places in the syntax, then it's better to have it as meta syntax.

16:01:43 [apaschke]

Christian: you want to use different fact bases

16:02:19 [apaschke]

Christian: either we can refer to external dataset from anywhere or it is a meta information at any level

16:03:03 [apaschke]

Axel: you want to have contextual rules

16:03:55 [apaschke]

Christian: I want ?X to bind to DB1 and ?Y to bind to DB2

16:04:42 [apaschke]

Christian: If directives are at the meta level, they are allowed everywhere

16:04:59 [apaschke]

Christian: or we can say import is a built-in used -

16:07:25 [apaschke]

Chris: The issue is do we need this directive construct

16:08:00 [apaschke]

Jos: second controversal thing is MetadataStatement

16:08:24 [AxelPolleres]

I see there are two issues, basically repflecting rule/dataset and named rule/dataset imports in christian's proposal.

16:08:39 [AxelPolleres]

... this is similar to FROM and FROM NAMED in SAPRQL.

16:09:14 [apaschke]

in this context we should also consider constructive views used for scoping

16:09:58 [apaschke]

Jos: with the metadatastatement you can something about the predicates you are using

16:10:01 [AxelPolleres]

We need something like modules (in rulesystems like e.g. FLORA typically referred to by '@' after (atomic) formula.

16:10:24 [AxelPolleres]

and in SPARQL you get something similar with the GRAPH keyword...

16:10:41 [apaschke]

Christian: Why is this not metadata

16:10:52 [apaschke]

Sandro: because it is a triple

16:11:16 [apaschke]

Chris: all metadata is binary

16:11:35 [apaschke]

Chris: a statement about rulesets or rules

16:12:07 [apaschke]

Chris: where the ruleset or rule would be the subject

16:13:09 [apaschke]

Axel: why not specifiy medata statements in an external RDF document

16:13:47 [apaschke]

Axel: MetadataStatement can be defined outside of a RIF document and refered

16:14:43 [apaschke]

Harold: you could link to the external document from the place where the property/predicate is used

16:14:50 [apaschke]

Harold: you would centralize it

16:16:33 [apaschke]

Axel: MetadataStatement duplicates RDF

16:16:51 [apaschke]

Christian: MetadataStatement is meta data and should not be used

16:17:37 [apaschke]

Christian: If we put that into RIF it means you need to implement RDF

16:18:40 [apaschke]

Garry: I'm fine with metadata, but directives are interpreted differntly by different engines

16:18:55 [sandro]

Sandro: How about having rif:Document can have rif:Ruleset and rif:AuxiaryRuleset, which are linked by URI (xml:id)

16:18:57 [apaschke]

Garry: that needs to be clearly thought off

16:19:09 [sandro]

Sandro: (one man's metadata is another man's data.)

16:19:30 [apaschke]

Garry: needs to be clearly specified (maybe even as part of the semantics)

16:20:14 [Zakim]

Zakim has left #rif

16:21:00 [AxelPolleres]

the mapping looks very familiar to me... look RDF <-> XML mapping in http://www.w3.org/2005/rules/wg/wiki/AbstractModel?highlight=%28abstract%29

16:22:23 [AxelPolleres]

If we write RIF in RDF, we can indeed put metadata anywhere! :-)

16:22:36 [apaschke]

Christian: RIF also needs to provide a way to specifiy metadata

16:22:45 [sandro]

something like html's rel=meta

16:23:12 [apaschke]

Chris: It would be strange to seperate metadata from the rules in two documents

16:24:52 [apaschke]

Sandro: need an import for bind to external data

16:25:11 [sandro]

Sandro: ... but we can do that later.

16:26:15 [apaschke]

Michael: better to have metadata in our RIF language

16:26:38 [apaschke]

Jos: don't want to give a semantics to metadata

16:26:49 [apaschke]

Michael: they could be facts

16:27:02 [apaschke]

Michael: mayb a different namespace

16:27:18 [apaschke]

Michael: we only need to come up with a convention

16:27:28 [apaschke]

Harold: we wanted a scope notion

16:28:09 [apaschke]

Chris: I would object to have the metadata in the domain

16:28:19 [apaschke]

Michael: then you can not process it

16:28:29 [apaschke]

Sandro: have a flag on import

16:31:10 [AxelPolleres]

would the IRI default to the base IRI? Can we define something like a base IRI?

16:31:56 [apaschke]

Chris: objections to IRIs?

16:32:34 [apaschke]

Axel: Can we have a base IRI

16:33:53 [apaschke]

action: hboley add IRIs to presentation syntax

16:33:53 [trackbot-ng]

Created ACTION-441 - Add IRIs to presentation syntax [on Harold Boley - due 2008-02-29].

16:34:37 [AxelPolleres]

Seeing Jos XML example, I confirm the claim that we need a way to define the BASE IRI in the XML syntax.

16:35:18 [AxelPolleres]

As we do need to be able to define prefixes and base iris in the presentation syntax.

16:38:35 [apaschke]

Jos: metadata can be only defined within a Rule tag

16:38:48 [apaschke]

Harold: the other way would be to use Forall

16:39:36 [AxelPolleres]

BTW, what if I give 2 rules the same IRI (and different metadata?)

16:39:45 [AxelPolleres]

... weird.

16:41:04 [apaschke]

Garry: what about a set of facts. I don't want to wrap them all in rule tags

16:41:51 [apaschke]

Jos: you have to wrap them in a rule to allow metadata

16:44:34 [JeffP_]

JeffP_ has joined #rif

16:46:33 [Harold4]

Axel, maybe there is a solution for your issue: global IRIs for Ruleset local IRIs (#-QNames) for rules etc.

16:46:35 [apaschke]

Christian: leave directives and MetadataStatements out of the next working draft

16:47:35 [Harold4]

Action: Harold2 Make proposal for adding metadata into BLD syntax.

16:47:35 [trackbot-ng]

Sorry, couldn't find user - Harold2

16:47:41 [AxelPolleres]

You mean IDs?

16:47:47 [Harold4]

Action: Harold4 Make proposal for adding metadata into BLD syntax.

16:47:47 [trackbot-ng]

Sorry, couldn't find user - Harold4

16:47:49 [AxelPolleres]

(document unique)

16:47:49 [apaschke]

Chris: topic "publication plan"

16:47:57 [Harold4]

Action: Harold Make proposal for adding metadata into BLD syntax.

16:47:57 [trackbot-ng]

Created ACTION-442 - Make proposal for adding metadata into BLD syntax. [on Harold Boley - due 2008-02-29].

16:48:24 [apaschke]

Chris: review actions wrt publication

16:56:17 [apaschke]

publication date is March 21st

16:57:33 [apaschke]

March 21st publication date for BLD

16:57:56 [csma2]

FLD-BLD freeze date is March 11th

16:58:13 [csma2]

FLD-BLD review by March 18th -> decision at telecon

16:59:37 [csma2]

FLD-BLD publication date is XXX

16:59:46 [csma2]

SWC freeze March 14th

17:00:08 [csma2]

SWC reviews by March 18th

17:00:24 [csma2]

SWC publication date is March 21st

17:00:40 [csma2]

s/XXX/March 21st/

17:01:08 [csma2]

DTB (Data Types and Builtins) freeze March 11th

17:01:34 [csma2]

DTB reviews by March 25th

17:07:19 [csma2]

UCR freeze March 14th

17:07:40 [csma2]

s/14/11/

17:09:28 [csma2]

UCR reviews by 18th

17:09:43 [csma2]

UCR publication is March 21st

17:11:22 [csma2]

PRD next internal draft March 7th

17:11:50 [csma2]

PRD telecon to be set up the week after

17:12:46 [sandro]

csma: PRD decisions in my mind:

17:13:00 [sandro]

... next draft will built on Core syntax, by reference

17:13:07 [sandro]

... no aggregated, no binding pattersn

17:13:16 [sandro]

.... semantics rewritten based on states and transitions.

17:13:50 [sandro]

... actions limitted to ASSERT and RETRACT.

17:15:24 [sandro]

... will not include agenda/priorites in this draft --- but they will be in FPWD.

17:16:22 [sandro]

RRSAgent, pointer?

17:16:22 [RRSAgent]

See http://www.w3.org/2008/02/22-rif-irc#T17-16-22

17:22:11 [csma2]

RRSAgent, make minutes

17:22:11 [RRSAgent]

I have made the request to generate http://www.w3.org/2008/02/22-rif-minutes.html csma2

17:37:25 [csma2]

csma2 has left #rif

18:11:39 [josb_]

josb_ has joined #rif

22:52:59 [sandro]

sandro has joined #rif

23:11:22 [JeffP_]

JeffP_ has joined #rif