See also: IRC log
<Jeffrey> I can only join IRC, jianhui from CAS, china
<JacekK> jianhui, I'll probably count this as regrets
<Jeffrey> ok
<scribe> scribe: holger
RESOLUTION: minutes approved
<caribou> ACTION: [ONGOING] BNS, Rama to start the Examples document [recorded in http://www.w3.org/2002/ws/sawsdl/minutes/20060502#action01]
<caribou> ACTION: Amit to send their Master's Theses usecases [ONGOING] [recorded in http://www.w3.org/2002/ws/sawsdl/minutes/20060502#action02]
<cgi-irc> use case page from Amit and John
<caribou> ACTION: ericP to decide our namespace and send it to editors [DONE] [recorded in http://www.w3.org/2002/ws/sawsdl/minutes/20060502#action03]
JacekK: f2f confirmed for June: 20-21
... September Last Call, Parking Examples doc as WG note
... optimistic shedule March 2007 final recommendation
... Saying important things only on IRC is bad
... Editors should probably keep todo list
<caribou> editors can put editor's notes into the documents
<caribou> even in published drafts
JacekK: Jacekk will add extra field if issue is addressed in editors draft or not
<JacekK> ACTION: JacekK, editors to amend the issues list and the spec with an editor todo list (resolution implemented flag) [recorded in http://www.w3.org/2002/ws/sawsdl/minutes/20060502#action04]
Jacekk: are there examples?
Rama: sending raw data is not usefull, use case context would be missing
... we need some template
JacekK: I was hoping that people send whatever they have, in a first place
... we can work on a template after
Rama: Fine, I'm going to send files
JacekK: issue about lsdis namespaces
... the usual URIs for examples are example.org
EricP: the W3C policy is that namespaces are checked at publication time
<ericP> also, *.example
Joel: mention of action concept will be removed
holger: hard to make big business example self contained
<ericP> i think it's wise to separate examples from use cases
Joel: in document toy example might be used, however business context important to get relevance
Rama: agrees example must have some relevance
<cgi-irc> SUMO Finance ontology: http://lsdis.cs.uga.edu/projects/meteor-s/wsdl-s/ontologies/LSDIS_Finance.owl
Jacekk: combine discussion with issue 9
Joel: example currently in apendix, is this in line with w3c spec practice?
EricP: no w3c policy, up to editors to decide
Joel: big example in apendix, however simpler example could be used for introduction
<ericP> big +1 to example up front
Rama: example up front will help readers very much
Joel: need then simpler example (currently RosetaNet).
Rama: can be layered, first simple, later more complex
JacekK: summarizes, that 2 examples are required (one simple for the beginning one complex for apendix)
RESOLUTION: issue 9 closed: two examples expected, one simple in introduction, one large and complete in appendix
Jacekk: 3 options available as in agenda
... tendency for option 1 say nothing, leave to the Web
Rama: possible new issue, do we want modelReference explictly state type of semantic annotation when having multiple annotations (issue 7)
Jacekk: valid question, issue 7 might need to be resolved before
lhenocque: votes for not including language (since might change over lifetime)
john: concern about making spec dependent on languages
Jackek: no concern can be done via URI
Rama: if we go with option 1, not sure if it works with complex types
... should assumptions/discussion be in the spec?
Jacekk: not necessary, issue list and minutes are recording
ericp: w3c usually uses editors comments to issue that reference specific emails to keep track
jacekk: formal process before last call necessary?
ericp: not necessary
RESOLUTION: issue 3 closed: adopt option 1 for now, since no concrete concerns against it yet
jacekk: out of scope defining semantic language
<caribou> true
jacekk: external language not including semantics but only mappings is still not covered by "in scope"
... suggest to postpone this out of scope until we have candidate recommendation
<ericP> +1 to postponing
lhenocque: I may need to think more about this
jacekk: issue will be open + active for one more week
holger: editorial issue - should the semantic language be more specified in doc?
ericp: let editors make proposal
jacekk: editors free to say the semantic language must give us IRIs