On 6/7/07, John O'Hara <[EMAIL PROTECTED]> wrote:
Bingo. Nicely explained.
I'm glad someone else sees the problem.
We need to keep our software 100% clean; its amazing how much IP law you
need to know to write code and give it away.
Which is why the generic form API for AMQP should be derived from AMQP, not
JMS.
AMQP itself was designed mostly based on IETF concepts which are
unencumbered (like smtp, nntp, nfs). *After* it was designed it was checked
against the JMS specification to see that a JMS API could be implemented in
terms of AMQP primitives.
AMQP is not derived from JMS and has different architecture principles.
Well, I certainly gave some non-trivial input in the early days of the
design of the AMQP protocol to change it to map more closely the
semantics of MOMs like MQSeries, TibCo and in particular to map more
closely to the semantics of JMS & its concepts. Admittedly this was
about 6 years after reading the JMS specification - but if a .Net API
is tainted by its authors having read the JMS specification then maybe
AMQP is tainted too?
--
James
-------
http://macstrac.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
