See also: IRC log
<trackbot> Date: 23 February 2010
8306 added to agenda by agreement
RESOLUTION: minutes of 2010-02-16 approved w/o objection
dug: introduce high level editorial
... move x-path to section 3.xx
... from section 7
... clarify "replace" and "insert" semantics
... define "insert" as always the child of the target
... eliminate "child" vs. "sibling" ambiguity
... a table of samples to show how everything works
ram: need more time
katy: is xsd validation sufficient?
dug: we could add more text
bob: conclude by next week?
ram: not sure
bob: on agenda next week
dug: introduce the changes to mex
... changes in seciton 3, swap section 4,5, section 6, section 11
... section 12 is new for bootstrapping
... and changes to xsd and wsdl
... question about example 8-2
ashok: ok with it
ram: remove line 6 and the corresponding text
dug: need more time until next meeting
<Dug> I think it would be unfair to accept it today - most people haven't had a chance to review it yet
asir: if we allow policy in addressing, we think it's premature
bob: any issue for that assertion?
asir: we need to describe the
semantics associated with the assertion
... an example is premature without behavior
gil: adding text will address your concern?
asir: there is a guideline on what can be asserted
dug: how about extensibility?
<gpilz> can we get a reference on where WS-Policy states that you MUST enumerate the allowable sub-assertions?
asir: you can insert anything, but
have to describe the assertion behavior
... drop it and open a new issue
ashok: this is not a nested policy, it's an outer policy applied to the endpoint
tom: assertion has to come with behavior/semantics
asir: if it's not a nested, then it is ok; remove it and open a new issue
katy: need more time on it
li: need more time to read it
dug: open issue at example 12-1
bob: anothe week is sufficient?
... please don't bashing each other over the head on nonnormative examples
... is this proposal covering all the issues?
ram: it is useful in general,
... suggest shortening it for time sake
<Dug> Gil's comment: http://lists.w3.org/Archives/Public/public-ws-resource-access/2010Feb/0025.html
ram: describing the proposal
gil: three terms is confusing
dug: i'm ok with the proposal if we merge the terms
<Dug> go with "consumer"
RESOLUTION: 8306 resolved w/o objection with the proposal but using the word "consumer"
<Dug> Note, a (subscription|enumeration context) expiring is not considered to be an unexpected termination, therefore a xxxEnd message MUST NOT be sent in this case.
bob: objection to accepting it now
instead of LC?
... can we break moratorium to accept it?
... 8886 is accepted w/o objection
gil: the proposal is correct, as expiry is not unexpected
ram: not sending xxxEnd is going too far
dug: why not say it?
gil: ram's solution allows sending
End when subscription expires
... leaving wiggle room is not good
yves: i'm ok with it and CWNA
katy: propose is clarify the implicity behavior
ram: timer may go wrong and we need to send End in that case
<Dug> new words for the spec: Acts of God MUST NOT occur. :-)
<Yves> how about cosmic radiations flipping bits in memory? MUST NOT occur? :)
ram: if the system detects that
... come up with some text
bob: happier if we close it today
dug: it only says timeout does not trigger End event, not prohibiting anything else triggering it
ram: propose new text...
... if timeout is normal, don't send; but allows other cases
bob: it is a LC issue then because no
... has been reached after live discussion
... it will delay our LC schedule
bob: let's work on the agenda
<Ashok> Bob: We agreed to the agenda. This issue was added at the end ... I would rather focus on substantive issues
<Ashok> ... if this issue is not quick to resolve, let's move on
bob: not permiting changing of
... that has been agreed
... other business?