W3C

- DRAFT -

RIF telecon March 18 2008

18 Mar 2008

Agenda

See also: IRC log

Attendees

Present
Mike_Dean, Mark_Proctor, Harold, Hassan_Ait-Kaci, Sandro, csma, AdrianP, IgorMozetic, Dave_Reynolds, josb, Stella_Mitchell, ChrisW, Gary_Hallmark, Adrian_P, AxelPolleres
Regrets
PaulVincent, LeoraMorgenstern
Chair
Christian de Sainte Marie
Scribe
Igor Mozetic

Contents


 

 

<csma> Scribe: Igor Mozetic

<csma> scribenick: IgorMozetic

Admin

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2008Mar/att-0064/04-Mar-2008-rif-minutes.htm

<csma> PROPOSED: accept minutes of March 4 telecon

RESOLUTION: accepted minutes of March 4 telecom

<csma> PROPOSED: accept minutes of March 11 telecon

<csma> http://lists.w3.org/Archives/Public/public-rif-wg/2008Mar/att-0072/rif-minutes-11-March-2008.html

RESOLUTION: accepted minutes of March 11 telecon

F2F10

no update on F2F10

Action review

Liaison

FLD/BLD grammar

josb: no news on OWL+RIF task force

Harold: discuss UNITERM
... UNITERM is split in functions and predicates
... since nobody supported UNITERMs, they were taken out of docs

<Adrian_P> Expression

josb: expresses concern over the term 'expression'

Harold: Expression is only function expression

<markproctor> how do I get on the speaker queue?

josb: what about predicate expression?

<sandro> type "q+" markproctor

sandro: let's wait for feedback and mark issues under discussion

<Harold> Mark, we did have a neutral word for both predicate and function applications: Uniterms. But no one came to their support.

<Harold> So we again split them into Atoms and Expressions.

<Hassan> Not sure I understand the argument (!) about arities,,,

<Adrian_P> Boolean functions

csma: let's keep separation between functions and predicates

<markproctor> harold: yes it does seem a bit more a mess around the definitions and how different systems seem the differently.

<markproctor> in the PR world it's not something I'm comfortable with

csma: let's keep the naming and mark it as 'under discussion'

<markproctor> and would like to see resolved

<sandro> PROPOSED: Open an issue on terminology around terms/functions/predicates and include an editorial comment in WD2 asking for feedback.

<sandro> PROPOSED: Open an issue on terminology around terms/functions/predicates and include an editorial comment in WD2 asking for feedback, but keep "atom" and "expr" for this draft.

<sandro> +1

<josb> +1

<sandro> RESOLVED: Open an issue on terminology around terms/functions/predicates and include an editorial comment in WD2 asking for feedback, but keep "atom" and "expr" for this draft.

<Harold> I can hear someone typing through their mic.

<sandro> ACTION: cdesaint to open the issue on terminology around terms/functions [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action01]

<trackbot-ng> Created ACTION-447 - Open the issue on terminology around terms/functions [on Christian de Sainte Marie - due 2008-03-25].

<Adrian_P> definitions are e.g. given in the book about logic programming of Lloyd

csma: next issue logic terms vs. external calls

harold: external call cannot be in the head
... atomic can be in the head

csma: next issue: lists

harold: lists are not on the critical path

csma: PR handle lists as objects with specific builtins

<Adrian_P> in production rules you have collections of objects

<Adrian_P> an define patterns on top of the collections

<markproctor> personally I like the pair idea

harold: many want standardized syntactic sugar for lists

<Adrian_P> as far as I remember we discussed lists as not being on critical path in F2F8

<Adrian_P> as well as modules

csma: not having listst in WD2
... next issue: nested Forall

hassan: logical Forall is not the same as in BRs

csma: here we talk about BLD
... nested Forall should not be allowed in BLD, but we need them in FLD

<GaryHallmark> I think Forall( ?X ?Y ...) is the same as Forall( ?X Forall( ?Y ...))

<Harold> Gary, I agree.

<markproctor> in PR forall is a logical element. It means when a pattern is true for all matching objects.

<markproctor> forall to iterate the entire WM, has little value for PR systems.

<GaryHallmark> -1

csma: Forall is needed when quantifiers are dirrefent

<GaryHallmark> ... for nested forall

<Adrian_P> we have no scoping yet. So nested Forall can be flattened and is not really needed

sandro: wonders about roundtripping, why to change the nesting?

<GaryHallmark> nesting is useful when alternating quantifiers in FOL

<sandro> PROPOSED: remove nested Foralls for WD2. Maybe we'll get some use case some day for putting them back in.

PROPOSED: remove nested Forall (nested variable declarations) from BLD

<sandro> PROPOSED: remove nested Foralls for BLD WD2. Maybe we'll get some use case some day for putting them back in.

<josb> +1

no objections

<GaryHallmark> +1

<sandro> RESOLVED: remove nested Foralls for BLD WD2. Maybe we'll get some use case some day for putting them back in.

csma: last issue: Metadata

harold: currently metadata not in BLD, proposal by Thur.
... natural attachment point for Metadata is Forall tag
... will add metadata by the next telecon

<sandro> ACTION: cdesaint put metadata on agenda for next telecon [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action02]

<trackbot-ng> Created ACTION-448 - Put metadata on agenda for next telecon [on Christian de Sainte Marie - due 2008-03-25].

XML syntax

csma: issues: stripping
... slots are not fully stripped

josb: RIF XML syntaxt is horribly verbose

<markproctor> the verbosity applies to the entire RIF

<Harold> You keep the 'object model' in the XML syntax.

<markproctor> you are developing an XML framework that is beyond most engineers

<markproctor> it will not be human readable

<markproctor> and thus only specialists can maintain systems that use the XML you create

<Harold> I responded to [11]: http://lists.w3.org/Archives/Public/public-rif-wg/2008Mar/0080.html

csma: full stripping makes derivation of XML syntax easy

josb: but not the reading of XML

<Harold> We could have OPTIONAL 'strip-skipping' in certain cases,

<Adrian_P> introducing Atom and Expr instead of Uniterm means also making RIF more verbose

<Harold> as Sandro discusses about 5 years ago in a web page.

<GaryHallmark> I would prefer spending effort on readability on the presentation syntax

harold: we could allow both alternatives, eg, arg

<josb> Gary, I agree that we should make the presentation syntax readable, but we need to keep in mind that also people will be working with the XML syntax.

<Adrian_P> from F2F9: RESOLVED: We keep named arguments, explaining in BLD that: A RIF consumer that does not support named arguments can implement them, with relative ease, by treating them as positional arguments (of a different predicate, formed in a stable but implementation-dependent way) in the lexical order of the argument names. (Closing ISSUE-44).

csma: normative syntaxt should be fully stripped, but falback doesn't be

gary: one shouldn't read the XML

<Hassan> +1 with Gary

<AxelPolleres> +1 to Gary, readable pres syntax pleeease

gary: prefers fully stripped syntax

<Harold> The fully striped <Atom><op>p</op><arg>abc</arg><arg>efg</arg></Atom> could be optionally stripe-skipped to <Atom>p abc efg</Atom>.

<csma> 0Adrian, why do you mention that resolution from F2F9?

<Harold> Basically, our current user slot names (keys) are fine;

<Harold> they maintain full striping on the user level, writing it as

<Harold> <slot>mykey myfiller</slot>

<Harold> because

<Harold> <slot name="mykey">myfiller</slot>

<Harold> doesn't XML-technically allow markup such as F-logic's

<Harold> <Var>Someslotname</Var> and <Expr>...method...</Expr>

<Harold> instead of the constant mykey,

<Harold> while

<Harold> <mykey>myfiller</mykey>

<Harold> would be "non-generic" markup (http://www.ai.sri.com/pkarp/xol/xol.html).

csma: since args are already ordered, this applies to slots and values as well

<sandro> LOL :-)

josb: why use abbreviations for tags if we don't care about the size of XML?

<Hassan> You mean size doesn't matter??? ;-)

csma: refers to resolution on naming conventions, Sep.11 2007

<csma> RESOLVED: we will have naming conventions; people edit the page to propose theirs (with explanation and reasons for any differences from what's already on page)

<csma> (from telecon setp 11 07)

<GaryHallmark> we could drop integers and just use 0, S0, SS0, etc

<Adrian_P> We had a discussion about camel case and long construct names

josb: open an issue regarding implementation of the above resolution

<sandro> ACTION: cdesaint open an issue on naming of XML element [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action03]

<trackbot-ng> Created ACTION-449 - Open an issue on naming of XML element [on Christian de Sainte Marie - due 2008-03-25].

<sandro> (elementS)

csma: issue: stripping

<sandro> csma: In NAU the key could be a constant, but in frames, they keys can be variables, etc.

<sandro> STRAWPOLL: Normative XML syntnax for RIF dialects will be full-striped

<sandro> +1 with a sigh

<GaryHallmark> we could spend a lot of time debating what stripes to remove...

<Hassan> 0 (to early to decide)

<josb> -0.5

+0.5

<AxelPolleres> 0

<DaveReynolds> +0.5

<GaryHallmark> +.839

<StellaMitchell> +1

<Harold> -0 (not as the only option: see optional stripe-skipping above)

<Adrian_P> 0 ( to early to decide, there are exception sometimes useful)

<mdean> +0.5

<markproctor> yeah still thinking about it :)

<markproctor> I won't know really

<markproctor> until I see how you handle types

<markproctor> personally I like the pair idea

sandro: change the XML for WD2 to be fully striped

<josb> The term "slot" is used nowhere in the BLD document except for the XML syntax. Instead of "slot", "argument" should be used, as in the BLD syntax description.

<markproctor> I would say this would be subject to agreement on how typing is handled.

<sandro> Harold: we'd need to be more precise about what fully-striped means.

<Adrian_P> agree with Mark, example name - value pairs

<markproctor> slots or named arguments - prefer slots.

<markproctor> as it's not always an argument

<sandro> csma: I think we could pass this resulution at a future telecon

csma: issue: XML syntaxt for BLD

<markproctor> 0 ( to early to decide, need to see how typing is handled)

<sandro> PROPOSED: XML Syntax of BLD in WD2 before fully striped -- modify frames and named argument uniterms.

<Harold> -1

<sandro> PROPOSED: XML Syntax of BLD in WD2 be fully striped -- modify frames and named argument uniterms.

<sandro> Harold: In my definition it already is.

harold: my definition of fully striped is different

<sandro> We can see these examples all spelled out in my e-mail: http://lists.w3.org/Archives/Public/public-rif-wg/2008Mar/0077.html

<csma> <Frame><slotKey>TERM</slotKey><slotValue>TERM</slotValue></Frame>

<markproctor> ok change of mind, thought about it more.

<markproctor> -1

<sandro> Ciao, DaveReynolds

<Adrian_P> to add a new role tag for a slot seems to be superficial

csma: move that discussion to future

<sandro> this is a VERY superficial issue, Adrian_P

<csma> <Frame> <!-- Basic form -->

<csma> <object> TERM</object>

<csma> <slotKey> Const </slotKey>

<csma> <slotValue> TERM </slotValue>

<csma> </Frame>

<markproctor> just feel that there re too many unknowns

<markproctor> yeah

<markproctor> I'd like to see more use cases

<Harold> You mean <slot>k f</slot> should be expanded to <slot><key>k</key><filler>f</filler></slot>?

<sandro> Harold, Christian wants my option 2. You're suggesting my option 3.

csma: XML schema in BLD is not alignd with BNF
... several updates in WD2 are needed
... are there any constraints on Var?

<markproctor> btw the problem on verbosity. the problem with key/value pair represetation is the least of your worries. compared to the rest of the verbosity.

Publication plan

extend the meeting for 5 min

<AxelPolleres> +1 ok for me ta add 5min

<sandro> +1 extend

<josb> +1

Publication plan: SWC

<sandro> josb: I need BLD syntax+semantics frozen before I can update swc.

csma: cannot freeze SWC today

<AxelPolleres> Can this made dependent on confrimation by email from michael?

josb: when BLD syntax+semantics is frozen, SWC can be frozen in two days

adrian: we can freeze UCR now

<sandro> ACTION: Sandro make frozen version of UCR. [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action04]

<trackbot-ng> Created ACTION-450 - Make frozen version of UCR. [on Sandro Hawke - due 2008-03-25].

two reviewers of UCR:

<sandro> ACTION: GaryHallmark review UCR [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action05]

<trackbot-ng> Sorry, couldn't find user - GaryHallmark

<josb> have to go

<josb> bye

<sandro> ACTION: Gary review UCR [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action06]

<trackbot-ng> Created ACTION-451 - Review UCR [on Gary Hallmark - due 2008-03-25].

Gary: agrees to review UCR by next telecon

<sandro> heh

<sandro> ACTION: Igor review UCR -- due 2008-0401 [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action07]

<trackbot-ng> Created ACTION-452 - Review UCR -- due 2008-0401 [on Igor Mozetic - due 2008-03-25].

<sandro> thanks

<Adrian_P> Gary and Igor please send comments on UCR to David and me

<Adrian_P> thanks

<Adrian_P> bye

Summary of Action Items

[NEW] ACTION: cdesaint open an issue on naming of XML element [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action03]
[NEW] ACTION: cdesaint put metadata on agenda for next telecon [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action02]
[NEW] ACTION: cdesaint to open the issue on terminology around terms/functions [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action01]
[NEW] ACTION: Gary review UCR [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action06]
[NEW] ACTION: GaryHallmark review UCR [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action05]
[NEW] ACTION: Igor review UCR -- due 2008-0401 [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action07]
[NEW] ACTION: Sandro make frozen version of UCR. [recorded in http://www.w3.org/2008/03/18-rif-minutes.html#action04]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/03/18 16:38:27 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/nexted/nested/
Succeeded: s/(Forall/Forall(/g
Succeeded: s/if F/is F/
Succeeded: s/striping/stripping/
Succeeded: s/harold/gary/
Succeeded: s/sice/size/
Succeeded: s/csma/josb/
Found Scribe: Igor Mozetic
Found ScribeNick: IgorMozetic
Default Present: Mike_Dean, Mark_Proctor, Harold, Hassan_Ait-Kaci, +6928aaaa, Sandro, +49.351.463.4.aabb, csma, AdrianP, +43.158.801.1aacc, IgorMozetic, Dave_Reynolds, josb, Stella_Mitchell, +1.914.784.aadd, ChrisW, Gary_Hallmark, +49.351.463.4.aaee, Adrian_P, AxelPolleres
Present: Mike_Dean Mark_Proctor Harold Hassan_Ait-Kaci +6928aaaa Sandro +49.351.463.4.aabb csma AdrianP +43.158.801.1aacc IgorMozetic Dave_Reynolds josb Stella_Mitchell +1.914.784.aadd ChrisW Gary_Hallmark +49.351.463.4.aaee Adrian_P AxelPolleres
Regrets: PaulVincent LeoraMorgenstern
Agenda: http://lists.w3.org/Archives/Public/public-rif-wg/2008Mar/0091.html
Got date from IRC log name: 18 Mar 2008
Guessing minutes URL: http://www.w3.org/2008/03/18-rif-minutes.html
People with action items: cdesaint gary garyhallmark igor sandro

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]