See also: IRC log
<Marsh> http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/att-0023/20051013-ws-desc-minutes.html
<Marsh> Resolution: Minutes approved
<jjm> Last week minutes approved
<pauld> my SPARQL Protocol review: http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0034.html
? 2005-07-21: pauld to write a proposal for a working group
report for requirements for schema evolution
following closure of LC124
DONE [.6] 2005-09-26: Arthur to figure out how to treat built-in schema
types. (LC315), due 2005-10-06.
? 2005-09-26: DaveO to draft a response and send to the WG.
(LC335), due 2005-10-06.
? 2005-09-26: Arthur to draft above as a proposal to be able to
close this issue (LC344#5), due 2005-10-06.
? 2005-09-26: Arthur to look for simplification options for
comment 12 of 344. (LC344#12), due 2005-10-06.
? 2005-09-26: Jonathan to point this out when it gets
Implemented (LC344#13), due 2005-10-06.
? 2005-09-26: Sanjiva and Roberto to investigate defaulting with
interfaceless bindings (LC333), due 2005-10,06
? 2005-10-06: Marsh to investigate LC301 re .NET scenarios,
due 2005-10-13.
? 2005-10-06: Charlton to augment Hugo's proposal with
parameters for all serializations, and syntax
for suppressing parameters, due 2005-10-13.
DONE [.3] 2005-10-13: Marsh to add RDF links to home page,
due 2005-10-20.
DONE [.7] 2005-10-13: Paul and Glen to review SPARQL draft,
due 2005-10-20.
DONE 2005-10-13: Tony to review I18N draft, due 2005-10-20.
DONE [.4] 2005-10-13: Sanjiva to write up his style-based generic
mapping in to media type (LC304), due 2005-10-20.
DONE [.5] 2005-10-13: Hugo to write up URI's describing architected
serialization format extensibility point
(LC304), due 2005-10-20.
Current Editorial Action Items
? 2005-07-21: Arthur to add stable identifiers for each
assertion, due 2005-09-26.
? 2005-09-26: editors to fix the first paragraph of section 4
... does not make sense at all right now.
(LC344#5), due 2005-10-06.
? 2005-09-26: Editors to add a sentence saying {address} is
optional because it could be defined by other
means, such as an WS-A endpoint reference or maybe
the scenario does not require an address.
(LC344#13), due 2005-10-06.
? 2005-09-26: Editors fix "Case Elements NOT cited" in 6.8.1.2
header to be "Case of elements NOT cited" (LC345),
due 2005-10-06.
Note: Editorial AIs associated with LC issues recorded at [.2].
[.1] http://www.w3.org/2002/ws/desc/#actions
[.2] http://www.w3.org/2002/ws/desc/5/lc-issues/actions_owner.html
[.3] http://www.w3.org/2002/ws/desc/
[.4] http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0022.html
[.5] http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0028.html
[.6] http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0027.html
[.7] http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0034.html
Table not really progressing
Publish what we have
Jacek: Ready to help
Marsh: Will look at status etc. with Hugo
Hugo: explain in status section that we are looking for feedback
Marsh: not sure if Semantic Web
WG will move to recommendation
... need to know who the customers of this note are
Paul: query language for querying
RDF triples
... built a protocol, described in WSDL
... raised 3 LC comments, mainly on HTTP binding
... have in-out query, bound to HTTP in SOAP; also HTTP version
with POST and GET
... bunch of examples
... they will have to republish their document
Marsh: is it a standalone document, we could validate with Arthur's validator?
<bijan> http://www.w3.org/TR/2005/WD-rdf-sparql-protocol-20050914/sparql-protocol-query.wsdl
Paul: yes, schema and WSDL
Arthur: could be contributed to our test suite?
Paul: would be neat
Marsh: do we need permission?
Arthur: is it a final document?
Paul: Last Call document
<scribe> ACTION: Bijan to contact WG to ask for contribution to test suite [recorded in http://www.w3.org/2005/10/20-ws-desc-minutes.html#action01]
<Marsh> ACTION: Marsh to forward nits to SPARQL [recorded in http://www.w3.org/2005/10/20-ws-desc-minutes.html#action02]
Marsh: any other comments?
Arthur: have they given up to their objection in parantheses?
Bijan: pat was still unhappy, but realized something has to be done anyway
<pauld> one last SPARQL nit: the document relies upon the reading of several external documents - a WSDL and a pair of schemas, the links of which are buried in the document. In the case of the schemas you have to hop through their namespace documents. Might be nice to have a separate table calling out these links.
<Marsh> http://www.w3.org/TR/2005/WD-ws-i18n-20050914/
Tony: very involved in WSDL,
mentionned on line 3, however no bibliographical
reference
... also references WS-Routing
... nothing but a few typos, pretty clean
Marsh: relation to WSDL: defining some headers?
Tony: yes
... if only one locale, should be in french
... they produced something we will be able to use
<Arthur> gotta go, bye
Marsh: can't start NCName with $ sign (example 6)
Tony: section 3.2, the locale element, missing element in enumeration
Marsh: should ask for
clarification re. $ sign
... should be interesting to see the type pointed to
... section 3.2 show a value of user: confusing
... referencing to WS-Routing wrong, missing WSDL
Tony: also example 4 are in fact
two different examples
... example5, should not be de_de but de-de
<scribe> ACTION: Tony to write up issues on I18N draft and email to WSD mailing-list [recorded in http://www.w3.org/2005/10/20-ws-desc-minutes.html#action03]
Skipping over for now
Not checked, skip over
Marsh: proposal from Hugo, worth some discussion
<Marsh> http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0028.html
Hugo: last week, discussed
ambiguities in way serialization format could be extended
... in consequence, 2 action items
... looked at how to use URIs for extensibility points
... instead of using mediatype, use URIs for the 3
serialization format that we define
... also, in light of resolution for LC337, change type of
in/...fault serialization mediatypes to URIs as well
... this will fix the extensibility ambiguity
... can now say: return XML/RDF or XML/whatever using a URI
defined for that purpose
... will lose mediatype parameters, however their use was
unclear
Marsh: my understanding not just
what mediatype is, but how you go from XML structure defined in
WSDL/Schema to that mediatype
... need some mechanism to go from XML to MIME package
... with RDF, several possibility for serialization
Sanjiva: our we going to specify how to go, e.g. base64 to image/jpeg, and do so for every serialization we define?
Marsh: do you think we could have a generic way of mapping from XML to mediatype?
Sanjiva: yes
Marsh: e.g. RDF constrained with Schema
Sanjiva: put schema in serialization
Marsh: what if what to put mediatype as well?
Sanjiva: should then introduce httpMediatype
DaveO: separate issue: idea of
having URIs for mediatypes is fabulous, TAG (or whoever)
precisely asked IANA to use such a mechanism
... IANA did produce a few, but not much
... W3C could well define some and provide mapping to
mediatypes
<Zakim> hugo, you wanted to talk about RDF Schema, SPARQL-based micro-syntax
Marsh: a little more subtle
Hugo: naming serialization format
with mediatype is problematic, because on the wire may be
similar, although different mediatypes
... there could be several serialization formats sending
application/RDF+XML
... changing the name of serialization format to URI is a
simple change
... changing completely serialization formats is bigger
change
Amy: has similar problem to
sanjiva
... unlike DaveO, we're not getting much here
... if 1-to-1 mapping from mediatype to URI, then what are we
getting unless incompatibility?
Hugo: not proposing to use URI
just in place of mediatype token
... believe one could come up with several application/RDF+xml
serialization format
... would all use that as their content type
... however could use different rules
Amy: would there be a schema for
the messages in RDF?
... and thus need mapping
... multiple mappings to mediatype; but where would WSDL
mediate that interaction?
Marsh: spec now has 3 sections/3 serialization, could define URIs "representing" these sections
<alewis> so the issues is that when we define serialization for a particular media type, we "use up" that media type and preclude other mappings?
Marsh: for example, XML serialized into an HTTP GET. Now we have a second serialization format, but still the same mediatype
Asir: are we asking for 1 or 2 properties?
Hugo: proposes a simple fix:
using URIs
... possibly better solutions, but too late in game
... URI would provide an identifier
... either you know the serialization format and can make sense
of the data, or you don't and can't
Sanjiva: don't want to delay
spec
... but if should have 2 properties, then let's have them
<Zakim> Marsh, you wanted to ask how well this solution solves the SPARQL issue.
Sanjiva: WSDL URIs not a solution for me
Marsh: would this solve SPARQL issues?
Hugo: email with kendal clark,
pretty happy
... not convinced that spec as it stands + LC337 is enough
Marsh: could we say this is the content type that gets over the wire?
Hugo: either keep current
situation (but need clarifications, e.g. what does it mean if
foo/bar as contenttype but not defined in WSDL spec),
... ... or use URIs as proposed, or use a solution similar to
Sanjiva's.
... status quo will not make it because spec is currently
unclear
Marsh: worried to put options forward at this stage, still some confusion
Asir: RDF issue only?
Hugo: no, more general, although indeed problem was raised from RDF
Asir: previously had encodingType in SOAP binding; isn't this similar?
Hugo: maybe people will be happy
to use only the 3 serialization formats we defined; but some
people are already trying to use others
... suppose want to use RELAXNG for the format of messages,
none of our formats (except application/xml) can be used
(because rely on XML Schema)
... so, here is a very concrete example
... will need to introduce a new property in the binding, like
rngHTTPSerialization format
<hugo> http://lists.w3.org/Archives/Public/www-ws-desc/2005Oct/0038.html
Marsh: can we agree on some
principles?
... do we want RDF+XML to be usable out of the box?
Jacek: not quite, but if constrain our bindings too much, then there is an issue
Marsh: do we need more than
Hugo's solution?
... maybe discuss options by email? but then only one option
really standing, and discussion for several weeks now
Jacek: in favour of Hugo's proposal
<alewis> i'm not really happy with it.
Marsh: anyone unhappy and willing to see something different?
Youenn: sanjiva's proposal with 2 properties may be better
Marsh: sanjiva could you put your proposal on the list?
<alewis> hugo just suggested pretty much what i was going to say, only a little more clearly. existing property in IANA media-type format, additional property that points at a mapping, using a URI.
Hugo: will draft something along the lines of sanjiva's
<scribe> ACTION: Hugo to send email describing sanjiva's 2 properties proposal [recorded in http://www.w3.org/2005/10/20-ws-desc-minutes.html#action04]
Marsh: comment from I18N WG
Hugo: pretty easy
Marsh: oh ye?
... should mention that xsi:anyURI, although supporting IRIs in
principal, has no mapping currently in such form (???)
... anyone with a better clue?
... what does this mapping mean?
... 1) adopt as is; or 2) go back and ask for
clarification?
<scribe> ACTION: Marsh to go send email asking for clarification [recorded in http://www.w3.org/2005/10/20-ws-desc-minutes.html#action05]
Marsh: inconsistency of where
faults are placed in bindings
... move tranferCoding
Hugo: sounds good
Resolution: accepted as is
... move tranferCoding from infault and outfault to binding
fault
Marsh: typo: missing "r"
Resolution: resolved
Marsh: section 2.10.3, table has error, should say "interface fault component"
Resolution: resolved
Marsh: sections 2.1.2, 2.2.1,
contradiction between SHOULD and MUST
... believe it was indeed our intention
Resolution: resolved
section 2.4, example need better formatting
Resolution: resolved
Marsh: adjourn