[graap-wg] proposal: stateful async agreement

[Toshiyuki Nakata]

OK.

I don't want to have both sides shooting/shouting at me :-)
Best Regards
Toshi


Karl Czajkowski wrote:

>On Mar 22, Toshiyuki Nakata loaded a tape reading:
>  
>
>>Sorry to ask many snippets of questions rather than chunk them to gether.
>>
>>If an initiator issed five different createPendingAgreement operations,
>>May I understand that it is the initiator's responsibility to create
>>5 different initiatorAcceptance EPR so that the response might be
>>discriminated properly?
>>
>>    
>>
>
>Yes, that is the idea... just like with the deliverNotification
>pattern in WS-BaseNotification.
>
>There is nothing particularly expensive about creating EPRs, e.g. new
>addressable resources.  The alternative goes right back into the
>explicit correlation ID mess, which has all the same requirements for
>initiators keeping track of IDs and you still have to pass an EPR
>too. It hurts my brain to think about <EPR, ID> pairs for addressing. :-)
>
>Oh, I see your other email about using the Agreement EPR as a
>correlation ID.  I think that is about the only way to upset the WSRF
>and anti-WSRF folks equally... ;-) I think EPRs should only be passed
>around when there is an expectation that a recipient will eventually
>invoke operations on it.
>
>
>karl
>
>  
>

-- 
We have moved to a new Office!!
Toshiyuki Nakata ?????
Internet System Laboratories NEC 
t-nakata at cw.jp.nec.com
1753, Shimonumabe, Nakahara-Ku, 
Kawasaki,Kanagawa 211-8666,Japan 
Tel +81-44-431-7653 (NEC Internal 22-60210)
Fax +81-44-431-7681 (NEC Internal 22-60219)