RE: [business-transaction] Regarding issue #89

["WEBBER,JIM (HP-UnitedKingdom,ex1)" <jim_webber@hp.com>]

Title: Message

Hello TC,

 

 e) I disagree with the idea that this will not have any impact on protocol as it stands now. How would one externalize the state without involving the BTP actors?  We cannot just say it will not have impact on BTP. Surely some actors/roles will involve in externalizing the state and its consistency, perhaps even the participants will involve... If you can do it without involving the BTP actors and without changing any aspects of the protocol, then it is clear that the externalizing state is out of scope of BTP work - there is nothing to do other than agreeing on a form that we think that a BTP system will externalize its state. Even if so, I do not see its urgency at this version of the spec.  If there is no impact, and no changes needed, perhaps I am also missing something here... 
 

I think Sazi has hit the nail on the head here. If we are to allow externalisable state, the there will be one more role in BTP multiplied by however many actors can play that role. Would this not therefore necessitate considerable changes?

 

However, I could still happily be convinced of the viability of this with a suitable use-case.

 

Also, I am not using the "royal we" here since there has been far too much of that on these lists lately ;-)

 

Jim