-
DRAFT -
SV_MEETING_TITLE
5
Sep 2006
See
also: IRC log
Attendees
Present
Allen_Ginsberg, ChrisW, csma, DaveReynolds, Donald_Chapin, FrankMcCabe, GaryHallmark, GiorgosStoilos, Harold, Hassan_Ait-Kaci,
JeffP, johnhall, Jos_de_Bruijn, Jos_De_Roo, LeoraMorgenstern,
MalaMehrotra, MarkusK, MichaelKifer, Mike_Dean, PaulaP, sandro, StellaMitchell
Regrets
FrançoisBry
AxelPolleres
IgorMozetic
MohamedZergaoui DeborahNichols
Chair
Chris Welty (ChrisW)
Scribe
Allen_Ginsberg (Allen)
Contents
Administration
<ChrisW>
http://www.w3.org/2005/rules/wg/wiki/F2F3#minutes
f2f3 minutes now on wiki
too soon to approve
Will
approve next week
<ChrisW> http://lists.w3.org/Archives/Public/public-rif-wg/2006Aug/0041.html
last week telecon minutes posted
Approve?
Does not have regrets or attendance
David will send email to Alex to please
update attendance - regrets
No updates to agenda
Liaison
<Donald_Chapin> nothing new
UCR
<Allen> UC1 is clearly
about exchanging both facts and rules. Maybe Sven's question was about an
earlier version or something. (addressing Sven's
review of UCR)
I haven't finished putting in the action updates from last week.
ACTION: allen
to post the previous work on links between reqs and
[recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action01]
<ChrisW> UC
<rifbot>
Created ACTION-98 - Post the previous work on links between reqs
and [on Allen Ginsberg - due 2006-09-12].
<sandro>
ACTION-98 CLOSED
<sandro>
Looking at: http://www.w3.org/2005/rules/wg/track/products to see overview
of topics
<LeoraMorgenstern> what
document is Chris reading from?
<LeoraMorgenstern> URL?
<sandro>
Chris is probably looking at http://www.w3.org/2005/rules/wg/track/issues/open
<LeoraMorgenstern> Okay, thanks.
<ChrisW> http://www.w3.org/2005/rules/wg/wiki/Use_Cases
<Allen> http://www.w3.org/2005/rules/wg/wiki/UCR
http://www.w3.org/2005/rules/wg/track/issues/3
<ChrisW> "Which CSFs (other than Alignment) does the XML syntax requirement
support? "
CSFs are http://www.w3.org/2005/rules/wg/wiki/UCR/Goals
look at 9-circle diagram
near top of that page
<csma> The idea is
to show that XML is important for more than just W3C.
ie it supports widescale
adoption in other ways, too.
ChrisW: + low
cost of implementation
<ChrisW> + Extensibility
what about Interoperability?
<csma> That's interop
between rules
<FrankMcCabe>
link between XML and extensibility is pretty weal
<sandro>
I don't know if it's weak yet -- I don't know how we're doing extensibility
<DaveReynolds> -1 on XML
having strong support for the extesibility CSF
<csma> I think XML is a CSF
itself -- it makes
<GaryHallmark> xml syntax
supports low cost implementation because you can reuse standard language tools
<sandro>
FrankMcCabe: That is -- "low cost of
adoption"
<sandro>
csma: Implementation is one part of adoption
<sandro>
sandro: low cost of software development (adoption by
vendors); low cost of deployment (adoption by end-users).
<FrankMcCabe> you could interpret
"implementation" more generally
<ChrisW>
ACTION: Frank to draft
solution to Issue # [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action02]
<rifbot>
Sorry, couldn't find user - Frank
<ChrisW>
ACTION: Francis to
draft solution to Issue 3 [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action03]
<rifbot>
Created ACTION-99 - Draft solution to Issue 3 [on Francis McCabe - due
2006-09-12].
http://www.w3.org/2005/rules/wg/track/issues/4
<ChrisW> How about Frank looks at all the issues
around linking Reqs o Goals and CSFS,
issue 3, 4, 13
<ChrisW> http://www.w3.org/2005/rules/wg/wiki/UCR/Document_issues?highlight=%28ucr%29%7C%28issues%29
<sandro>
ammend action 99 to cover issues 4 and 13 as well
<ChrisW>
ACTION: Allen to deal
with ISSUE-5 [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action04]
<sandro>
rifbot not happy about the database being offline
either.
<ChrisW> NOTE that rifbot stopped before ACTION 4
<Allen> I'm scribing again
<PaulaP> Axel sent regrets
RIFRAF
<ChrisW>
Action review for RIFRAF
Hassan's Action?
Hassan and Christian will talk
continued
Alex and Axel not here
End of action review
Issues
Taxonomy/ontology
tools for building RIFRAF artifacts?
<Hassan> Great idea - but we need to agree on terms
Should we use OWL?
< ChrisW > should we use OWL
Or RDF/RDFS
<FrankMcCabe> excellent idea.
<Hassan> +1
but does OWL have expressiveness?
<sandro> seems like it could be good enough
<Hassan> agrees, but does everyone else agree?
<ChrisW>even
if something in RIFRAF is beyond OWL, doesn't mean it
doesn't go in RIFRAF
<Zakim> csma,
you wanted to ask about the link between UC and reqs
<DaveReynolds>
what does owl encoding buy us?
<FrankMcCabe> OWL would
enhance precision
<DaveReynolds>: if it is only
for people why do it?
< ChrisW > will need to
include text annotation. Problem is we are beyond capabilities of questionaire tech is doing this.
<DaveReynolds>: this is less
work than fixing questionaire?
<sandro>
looks like it is
<csma> encoding in owl buys us help in
extracting the meaning of the terms
<DaveReynolds>
need to analysis questionaire data
<Hassan> don't
think it will be that much work and will be a time saver....Questionaire
is losing forest for trees.
<Hassan> would like to discuss specifications with sandro
< ChrisW > might want to ask
Axel as well... continue to pursue this idea
<sandro>
general sense: let's go ahead
Technical Design
< ChrisW >
Action Review
Peter's
action done
Any
discussion on Peter's versus pre-existing wiki?
<csma> if we remove negation aren't they the
same?
Chris and Harold agree with that
<csma> could be used for first draft of
technical design
<MichaelKifer> semantics
doesn't make sense for naf
<csma> didn't understand Michael’s email
<MichaelKifer> agreed, but this
was clarified in 2nd email.
naf
changes what is an intended models
naf makes a global difference
<Harold> We could start with http://www.w3.org/2005/rules/wg/wiki/A.1_Basis%3A_Positive_Conditions
and extend those to Horn rules in Phase 1.
<csma> for phase 1 it makes no difference...
<MichaelKifer> phase 2 naf would require more but compatible with earlier
<csma> do phase 1 w/o negation
<JeffP> what about
individual vs data values for variables in P's
proposal
< ChrisW > michael's proposal was different
from P's in that way
<MichaelKifer> also
need to divide predicates along those lines too then
< ChrisW
> does M's semantics include that (possibly)
<MichaelKifer> yes, but it was a
general framework not a semantics per s
< ChrisW
> is breaking the universe up this way required for phase 1
<Harold> for many purposes not necessary, but not a
major problem if we need to do it later
<DaveReynolds> +1 to starting
with union, RDF would want to bind variables to both kinds
could
add these separate domains into later refinements
<Harold> it
should not contain this core is union; extensions could be
refinements
<Hassan> My phone is dead ... I'call
<JeffP> these
issues can be related in some way, might want to think
more carefully
<Hassan> I'm back - sorry! design
choices could impact the way we extend to negation
<MichaelKifer> not
related. data types uses sorts. negation
proceeds same with or without sorts
<Hassan> agree with
Michael
< ChrisW > question is do we
need it in the core?
<Hassan> this semantics is agnostic to the universe of
discourse
<Harold> We could have a uniform 'universe' of
constants initially, then split into "i"
and "d" constants, then split "d" according to XML Part 2 (Datatypes). Maybe we don't want new kinds of variables for
all of these refinements of constants, instead permit certain kinds of
'generic' variables (for some of these distinctions).
< ChrisW
> only one has been proposed so far and that one makes the distinction. Can
we simply that?
<Hassan> it is silly to have that.
< ChrisW
> not silly. should it be in core or in
extension.
<JeffP> another way
is to distinguish predicates. eg,
if you have builtin than its variables have type info
<FrankMcCabe> puzzled by
peter's intention.
<MichaelKifer>
typing variables is problematic. Don't make the distinction in the core.
<csma> support that. first
version should be simple. reaction to core will tell
us what to do
<Harold> I agree with Chris and Michael and Christian:
Easier not make distinction in first core.
< ChrisW > next step - put
peter's work in wiki page and connect it to technical design but delete
negation and revise interpretation to get rid of distinction
<MichaelKifer> start new wiki page
with original?
< ChrisW
> no.
<Harold> I can help
<Harold> can Michael’ss
proposal be the starting point (made compatible with peter's)
<ChrisW>
ACTION: harold to start new wiki page on
the core language to include Peter's proposal modulo changes we discussed
[recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action05]
Meeting adjourned
Summary of Action Items
[NEW] ACTION: Allen to deal with ISSUE-5
[recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action04]
[NEW] ACTION: allen
to post the previous work on links between reqs and
[recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action01]
[NEW] ACTION: Francis to draft solution to Issue
3 [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action03]
[NEW] ACTION: Frank to draft solution to
Issue # [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action02]
[NEW] ACTION: harold
to start new wiki page on the core language to include Peter's proposal modulo
changes we discussed [recorded in http://www.w3.org/2006/09/05-rif-minutes.html#action05]
[End of minutes]
$Date: 2006/09/05 16:30:39 $