atom feed29 messages in org.oasis-open.lists.sca-jRE: [sca-j] AW: ISSUE 8: Concurrency ...
FromSent OnAttachments
Michael RowleyFeb 21, 2008 11:45 am 
Barack, RonFeb 21, 2008 2:15 pm 
Mike EdwardsFeb 22, 2008 3:13 am 
Mike EdwardsFeb 22, 2008 6:30 am 
Shih-Chang ChenFeb 22, 2008 7:07 am 
Michael RowleyFeb 22, 2008 9:18 am 
Michael RowleyFeb 22, 2008 9:28 am 
Michael RowleyFeb 22, 2008 9:30 am 
Michael RowleyFeb 22, 2008 9:48 am 
Barack, RonFeb 22, 2008 10:25 am 
Michael RowleyFeb 23, 2008 8:05 am 
Mike EdwardsFeb 24, 2008 5:48 am 
Mike EdwardsFeb 24, 2008 6:10 am 
Mike EdwardsFeb 24, 2008 6:39 am 
Michael RowleyFeb 25, 2008 8:10 am 
Mike EdwardsFeb 25, 2008 10:29 am 
Michael RowleyFeb 25, 2008 12:04 pm 
Mike EdwardsFeb 26, 2008 5:51 am 
Michael RowleyFeb 27, 2008 5:38 am 
Mike EdwardsFeb 27, 2008 2:57 pm 
Simon NashFeb 28, 2008 6:42 am 
Simon NashFeb 28, 2008 10:26 am 
Michael RowleyFeb 28, 2008 5:31 pm 
Simon NashMar 13, 2008 8:03 am 
Michael RowleyMar 13, 2008 11:33 am 
Simon NashMar 13, 2008 11:44 am 
Peshev, PeterMar 13, 2008 12:41 pm 
Mike EdwardsMar 14, 2008 2:09 am 
Peshev, PeterMar 14, 2008 6:32 am 
Subject:RE: [sca-j] AW: ISSUE 8: Concurrency model for Service Reference instances
From:Mike Edwards (mike@uk.ibm.com)
Date:Feb 27, 2008 2:57:57 pm
List:org.oasis-open.lists.sca-j

Michael,

OK, let's declare a truce on the code front ;-)

I agree that this needs no changes in the spec.

I'd still like to ask why we need to keep setConversationID()? Is there really that much useful function associated with it? Yours, Mike.

Strategist - Emerging Technologies, SCA & SDO. Co Chair OASIS SCA Assembly TC. IBM Hursley Park, Mail Point 146, Winchester, SO21 2JN, Great Britain. Phone & FAX: +44-1962-818014 Mobile: +44-7802-467431 Email: mike@uk.ibm.com

"Michael Rowley" To Mike Edwards/UK/IBM@IBMGB, "OASIS Java" <mrow@bea.com> <sca@lists.oasis-open.org> cc 27/02/2008 13:38 Subject RE: [sca-j] AW: ISSUE 8: Concurrency model for Service Reference instances

You are absolutely right about the race condition in my code. And yes, getting code to be thread safe while also allowing some concurrency is far from trivial.

Nonetheless, nothing we are suggesting requires anything new from the spec. If the developer wanted to keep things simple, the entire buyBook routine would just be synchronized, the developer would gain simplicity at the cost of some concurrency. I'm not sure we need to get rid of setConversationID() as a means to improve concurrent programming scenarios.

Michael

Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU