See also: IRC log
Present: Abbie Barbir, Assaf Arkin, Colleen Evans, Dave Hollander, Dave Orchard, Doug Bunting, Frank McCabe, Gerald Edgar, Hao He, Heather Kreger, Hugo Haas, Igor Sedukhin, Martin Chapman, Mark Jones, Mario Jeckle, Mike Mahan, Roger Cutler, Sinisa Zimek, Tom Carroll, Ugo Corda, YinLeng Husband
Regrets: Chris Ferris, David Booth, Geoff Arnold, Katia Sycara, Paul Denning,Sandeep Kumar, Shishir Garg
Chair: Mike Champion
Scribe: Igor
------------------------------------------------------
1. [15:35] Confirm
scribe. The current list is at
http://www.w3.org/2002/ws/arch/2/04/scribes-list.html
Igor Sedukhin is the scribe
------------------------------------------------------
2. [15:37] Approve
last week's minutes http://www.w3.org/2002/ws/arch/3/04/2003-04-17-ws-arch.htm
Minutes were approved.
Other administrative issues? Additional agenda items?
ACTION: chairs rise an issue to CG to track the F2F details
------------------------------------------------------
3. [15:40] Action
Item Review
ACTION: Daniel to
check Mark J's text following Scottsdale F2F re
MEP
ACTION: Daniel, and
the usual suspects will work on defining the relationship between
messaging and description/choreog
Both still PENDING
ACTION: Editors
will reconcile terminology used with WSDL
ACTION: Eric will
close issue 21
?
ACTION: Hugo to
send pointer to task force mailing list to WSA (WSD
task
force)
DONE
ACTION: Geoff and
Eric to propose text about protocol independance by Thursday 24
April
ACTION: Chris to
summarize WS-Policy spec for the list
ACTION: Geoff, Chris
to report back to the group once agreement is reached on sync vs.
async next week
?
NEW ACTION: All
check the issues that they agreed to close
NEW ACTION: to the
editors to crossrefernce that (remindier of the
status)
Other
updates:
Abbie has sent
security section. Give another week to review.
ACTION: All to
review security section text
mmahan sent some
stuff privately to eric. will resend to the group, public
list.
the text was
related to privacy P3P
heather: MTF has
agreed to post 3.2 as a note, no further work @
W3C
heather: can't
commit to do the integration of the text in time
frank: what is the
real deadline for publishing
hugo: 14th of
may
<hugo> http://lists.w3.org/Archives/Member/chairs/2003AprJun/0015.html
<hugo> 17
May
frank: agrees to
propose integration of MTF stuff
<hugo> which
means that we have 1 extra week
<scribe>
ACTION: Frank to work with Heather on integration
---------------------------------------------------------
4. [16:00] General
plan for moving forward
heather: hard to
understand the doc
...discussion
arround integration of David's diagrams...
heather: is there a
better way to organize core concepts sesction, not just
alphabetical list?
heather: where is
the UML diagram?
martin: will look
into redrawing it as UML
martin:
timeframe?
<hugo> New
Deadline for Publication requests: 15 May 2003
<hugo> this
time, it's the real date
mchampion&hugo:
can't be tweaking the UML @ the F2F
heather: formatting
has to be improved
mchampion: propose
something to improve it
<Roger> The
short line format looks almost like some sort of
poetry.
<Roger>
Architectural Haiku?
<Heather>
Heather has to leave early, talk to you next week
mchampion: the
France F2F will have more formal votes in good standing to W3C
guidelines
<hugo>
http://www.w3.org/Consortium/Process-20010719/process.html#good-standing
hugo&chairs
will check what those gudelines are
mchampon: member
participation will be looked at in compliance with those
rules
mchampion: revise
the requirements doc and make sure all reqs have
owners
<hugo>
[[
<hugo>
January 2004
<hugo> Working Group
ends.
<hugo>
]]
roger: can't agree
for this working grp to go forever
mchampion:
agrees
mchampion:
eliminate the reqs we can't address in time
doug: limit the
number of reqs by the depth of description of the WSA and what it
covers
mchampion: yes,
doug is just being precise
roger: limiting
scope by restricting what WSs are is wrong
frank: focus on
things that must be there for sure and work on the
text
daveH: solidify
what has consensus, and prioritize things that don't
have
consensus
and address that later
dougb: scribe, not
sure how to capture this in minutes but my point was more that we
have
3 axes along
which scope reduction is possible: requirements, what we describe
(i.e. what
is a web
service) and depth of description (eg. focusing on
"infrastructure)
mchampion: if you
don't have time to spend with the editors, just take the
topic
of interest
and own it
frank: read the
output of the group.
daveH: make points
related to the document
frank: order of the
presentation: put the stakeholders view upfront?
frank: ...
describes the current doc ...
mchampion: text on
SOA + DaveO's text still needs to be worked in
frank: 1.7 WS
technologies: belongs to stakeholder's view section. important
for
consolodated
view on the various piece technologies
mchampion: what is
being asked?
frank: someone take
that section and own it...
mchampion: send an
e-mail to ask for owner
frank: it is
scattered, this section needs to be gathered and put
together
consistently
mchampion: editors
have to send an e-mail outlining where they need
help
ACTION: Frank to
send e-mail inviting for help on tech.consolidated view
section
and may
start writing that section
------------------------------------------------------
5. [16:15 ] Stack
Diagrams
Eric has
proposed
http://lists.w3.org/Archives/Public/www-ws-arch/2003Apr/0254.html
based on a long
conversation among the editors. Can we live with this, or
propose "friendly amendments" and move on? See
http://dev.w3.org/cvsweb/~checkout~/2002/ws/arch/wsa/candidate_diagrams.htm
for a collection of all known previous suggestions that were considered in coming up with this one.
We also need someone to take responsibility for writing the text and editing the properties/constraints to link the diagram(s) with the rest of the document.
<mchampion>
http://lists.w3.org/Archives/Public/www-ws-arch/2003Apr/0254.html
mchampion: can we
put it in the doc
mario: can live
with it. wants a straw poll
<mario>
Suggestion to adopt the current version of Eric's propsed diagram
as is (please
type "y" to
vote for adoption)
<Roger>
y
<mchampion>
y
<Mark_J>
y
<mario> The
diagram ist listed at:
http://lists.w3.org/Archives/Public/www-ws-arch/2003Apr/0254.html
<hugo>
y
<mario>
y
<ugo>
y
<DaveO>
n
<frankmccabe>
y
<arkin>
y
...discussion of
what this diagram is for and if it is applicable and detailed
properly...
<mmahan>
y-
daveO: orthogonal
concepts do not make sense on the stack diagram
daveO: 3 points:
messages, descriptions, discovery and security flow
though
(implied?).
wants more accurate representation of that
diagram 10 by
daveO
arrows? what
for?
daveO:
relationships are arrows
roger: does not see
stack diagrams making anything clearar in the
WSA.
roger: daveB's
diagrams, on the other hand made a lot of discussions
clearer
mchampion: may be
we're closer to consensus than we think we are
frank/roger: stack
just puts the names of related classifications in one
place
mark: agrees that
it is more marketecture than and architecture
roger: suggests an
outline instead of stack
mark: the point of
the diagram is non-linear relationships
<TomCarroll>
Then how about three related diagrams
<scribe>
...this grp needs to create the common marketecture stack diagram,
instead of
enumerating
all stacks that all the members could produce for
that
ACTION: mchampion
to produce text that renders into --A-- stack
diagram
------------------------------------------------------
6. [16:30 ]
Fundamental scoping constraints on WSA
It's obviously not productive to discuss "what does 'web service' really mean". The Chairs proposed a "Plan B" to refocus on the issue of the fundamental constraints on what is in-scope for the architecture that we're defining. See http://lists.w3.org/Archives/Public/www-ws-arch/2003Apr/0203.html
http://lists.w3.org/Archives/Public/www-ws-arch/2003Apr/att-0231/wsa-constraints.xhtml
mchampion:
summarize URI discussion
daveO: URI of what
exactly? no consensus arround that. WS enpoint has URI,
service
element in
WSDl may have a URI in 1.2. An agent implementing WS may be
identified. What
has to be
indentified to correspond to WS?
daveO: on the Web
URI identifies a resource. What that resource is for a
WS?
mchampion: all of
those things are Web resources and have URIs
martin: not all of
those are web resources
daveO: WSDL says a
WS is a collection of endpoints. The thing behind the
endpoints
is a WS
implicitly.
daveO: reiterates
his discussion with WSDL group about naming of those
things
daveO: all we can
do it interact with an endpoint and therefore --IT-- it a
WS
martin: WSDL
confuses the whole issue
frank: there is
something that is a WS and it has a name. name should be
the
identifier.
daveO: there is no
place in WSDl for the URI for the WS
matin: this is
fundametal for WSA
ACTION: group: review, develop opionon and sync up with
WSD