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

 


Help: OASIS Mailing Lists Help | MarkMail Help

bt-spec message

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


Subject: RE: [bt-spec] Context


Sazi,
 
I assume the "not" in the first sentence is a typo,or it should say "If" instead of "Unless"  - the sentence doesn't seem to make sense otherwise.
 
 
CONTEXT is *not* used between Superior and Inferior.
 
It is used in relation to application messages - broadly from Initiator to Service, but the particulars are only as understood by the application (which in this sense includes application-supporting infrastructure like interceptors, containers etc.) and not immediately visible to BTP.
 
CONTEXT is also used on the Initiator:Factory exchanges - related to BEGUN as the way the CONTEXT is delivered to the application from the Factory that just made it, and related to BEGIN to indicate that a sub-coordinator/sub-composer is to be created and enrolled with the Superior identified by the CONTEXT.
 
I don't see how Jim's proposal affects transaction coordination as such. It seems to affect only the representation of the information on the Initiator:Factory exchange.  It would be an incompatible change though, precisely because it affects the representation
 
Peter.
-----Original Message-----
From: Sazi Temel [mailto:sazi.temel@bea.com]
Sent: 03 February 2002 19:23
To: Peter Furniss
Cc: WEBBER,JIM (HP-UnitedKingdom,ex1); BT - spec
Subject: RE: [bt-spec] Context


Unless we do not have good reason to send/receive Context between roles other than Initiator and Service, I think I would support Jim in this issue.. Peter, can you summarize why you think Context should be used in Superior:Inferior communications... I have not looked at the issue in detail but I think if this became an issue it should be resolved in version 1.0 since it has an impact on transaction coordination.

--Sazi

At 04:42 PM 2/3/02 +0000, Peter Furniss wrote:
> > as issue list maintainer:  I won't make this a new issue
> > unless you ask, but will if you do.
>
> Hmm. I'm in two minds. Is it possible to make it an issue for the 1.1
> version of the spec :-)

That's almost more of a PR question than anything else. I think if it gets
put on the current list, it's a candidate for 1.0, but a potential
resolution is "deferred" to future version.

> I definitely don't want to get rid of the context, but I am starting to
> think it should only appear in application messages, and not in the
> transaction-coordination messages. I think it could well be a useful
> factorisation.
>
> For example, someone building a BTP-aware service will only have to
> understand what a Context is, whereas people building transaction managers
> don't. The people in the middle (i.e. application developers or
> more likely
> those that develop infrastructure for applications) must understand both,
> but then they are in the right position to do so since they know about
> services and transactions.

I would hope that people building transaction managers would understand how
to use transactions, at least in theory :-)

> How about if anyone else apart from me thinks it's an issue then
> let's talk
> about it, otherwise let's put it asside for now. Anyone else have anything
> to say?

Fine.  Does Jim have a seconder ?

Peter


----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>

Sazi Temel
'                               
bea Systems Inc.        
140 Allen Road  Liberty Corner, NJ 07938
        
sazi.temel@bea.com
sazi.temel@ieee.org
(1) 908 580 3123        
http://www.bea.com




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


Powered by eList eXpress LLC