OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

business-transaction message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: [business-transaction] Draft minutes BTP Conference call ,30 January 2002


Sorry for the delay.  I've summarised the notes that Alex and I had made
considerably - I hope I've not mis-interpreted or mis-attributed anything
significant.

-------------------------------------------------------------------


Draft Minutes of BTP Conference call , 30 January 2002


Decisions and actions flagged with -->

Attendees:
---------
	Bill Pope
	Doug Bunting
	Geoff Brown
	Sanjay Dalal
	Bill Cox
	Sazi Temel
	Alastair Green
	Peter Furniss
	Tony Fletcher
	Pyounguk Cho
	Alex Ceponkus
	Mark Little
	Jim Webber
	Victor Corrales

Minutes of 16 January meeting
-----------------------------
Draft minutes had been circulated by Bill Cox; completions (attendees list)
were being sent to him.
Approval will be by email

Agenda
------
There were no additions to the draft agenda

Votes on issues
---------------

Issues 6, 8, 17, 18, 52, 74, 77, 78, 82 were on the agenda. Bill Pope asked
if discussion was wanted on any of these, the others would be taken
together.

#78 had been under heavy discussion on the list – it would be discussed, but
decision deferred today.

#52 to also needed discussion

#77 referenced #78, but both the proposed solutions to 78 would fix it, so
it would be voted

--> Issue 91 (submitted by Oracle) – Geoff Brown withdrew this issue. Peter
would update the issues list accordingly.

Vote to approve proposed solutions to 6, 8 17, 18, 74, 77, 82, 47, 49, as
circulated with the agenda

Proposed Bill Cox, seconded Geoff Brown
None opposed.
--> Solutions approved

Discussion of issues:
--------------------

Issue 52 (Forms of prepare)

There had been some discussion on this – Doug wanted to be sure it was all
settled. Peter had stirred it up that discussion with a technical change
that he had now withdrawn. Sazi said this concerned the difference of what
happens to cohesions v atoms and whether confirm-transaction can be issued
to a cohesion including in the confirm-set inferiors that haven’t been
explicitly prepared.  Others said there were lots of ways of choosing
prepare and confirm, some could be error-prone.  Discussion led onto how
prepare_inferiors and confirm_transaction work, but this was beyond issue 52
itself.

Vote to approve proposed solution for Issue 52, as circulated with the
agenda
Proposed  Peter Furniss
None opposed
--> Solution approved

Issue 89 (J2EE interoperablity - JMS short circuit

Geoff Brown requested a decision on issue 89. Oracle wanted to have a btp
definition of the transaction state as the content of the JMS-defined
XMLTYPE, allowing interaction with jms.  This was urgent because the next
J2EE version is closing.

The discussion on this concluded that it would fit an additional binding for
BTP.
--> Geoff Brown will draft a binding.

The xmltype is defined in the new (draft) text for JMS.
--> Geoff would make available to the committee what he can on this (subject
to JCP constraints on their draft documents)

Issue 78:
--------

The email discussion was recapped and continued. Various points made were,
summarised by Doug as:

doug summarizes: schemes like http become ambiguous when you layer things
like soap on top of it.  efficiency & easy book keeping make overlaping
names with locations attractive.  Unfortunatly there are multiple bindings
that use simliar looking locations that makes it difficult.

bill: we lack implementation experience for combining the two together ..
especially if you are looking down the road at something like UDDI.

mark: that's why we are proposing to allow both

doug: issue 78 last three paragraphs should use the term URN, then we can
continue discussion.  Because we are taking about names and not
identification in the global sense.

--> it was agreed to modify the proposal for issue 78: changing the
parenthesis at the end to be "that is, a URN", and using the term URN in the
last three paragraphs.

People had dropped out and the meeting was no longer quorate, so 78 could
not be voted. It was agreed to continue discussion on 78 on spec list and
call an email vote

Future face-to-face
-------------------
--> Bill Pope to solicit the list for possible dates for a march f2f

Oracle offered to host

Next conference call
--------------------

BEA has volunteered for next call and HP is also able to host.






[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Powered by eList eXpress LLC