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


Help: OASIS Mailing Lists Help | MarkMail Help

ws-rx message

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

Subject: RE: [ws-rx] Proposal for issue 27

Well, in my opinion it is not a good idea to describe what is out of scope of the WS-RX spec. The intention of the proposal is to make it clear that the inOrder delivery assurance is about messages within a Sequence.
If that is still not clear, I can have another shot at it.


-----Original Message-----
From: Patil, Sanjay [mailto:sanjay.patil@sap.com]
Sent: den 8 september 2005 19:56
To: Andreas Bjärlestam (HF/EAB); ws-rx@lists.oasis-open.org
Subject: RE: [ws-rx] Proposal for issue 27

I thought the issue i027 was to clarify that - the problem of ordering of messages across Sequences is out of scope for the WS-RX specifications and belongs to the higher up layers. I am not sure if the proposal below adds the intended clarification to the specs. IMHO, the proposal clarifies further the requirements for ordered delivery within a Sequence but it is not saying much about cross-sequence ordering being in or out of scope for the WS-RX specifications.

I was not present on the conf-call when this issue was discussed but here is the cut-paste of the discussion from the minutes:
PZF: I've seen considerable discussion on the mailing list whether this represents correct use of the protocol as designed - I believe the feeling of the TC is to accept the issue and discuss further when we get on to it.
Stefan: Is this an issue of clarity in spec? Or something else?
Andreas: It is about more clarity in spec
Stefan: So, the spec should say "ordering should be enforced within a sequence"? Or specifically state something with respect to the ordering guarantee?
Andreas: It should state the former. 
PZF: Are there any objections to accepting this as an issue?
Duane: I have issues with this, but won't hold up the discussion... spanning of multiple sequences, IMHO, belongs to application layer.
Andreas: I agree, but didn't find it very clearly stated in the spec.
Duane: Is this then a proposal to clarify in the spec that the spanning of multiple sequences belongs to the application layer?
Andreas: Yes
PZF: This is an example of an issue which should be written more clearly....
PZF: Any objections to the issue (silence). Issue accepted
ACTION: Andreas to write a clear proposal for this issue.


>-----Original Message-----
>From: Andreas Bjärlestam (HF/EAB) 
>Sent: Thursday, Sep 08, 2005 2:48 AM
>To: ws-rx@lists.oasis-open.org
>Subject: [ws-rx] Proposal for issue 27
>In chapter 2 (lines 183-185) replace
>It requires that the sequence observed by the ultimate 
>receiver be non-decreasing.
>It requires that the messages within a Sequence will be 
>delivered to the ultimate receiver in a non-decreasing order.
>It requires that the messages within a Sequence will be 
>delivered in a non-decreasing order.
>We should also consider writing Sequence with capital S at line 176.

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