[JMC] I think you may
have missed the announcement for the new WS-RX TC, not to mention WSS and other
TCs :)
[Hamid]: Let me tell you
the story of WS-RX. We built WS-Reliability spec for two years, and two months before
Microsoft and IBM decided to create a new TC (WS-RX), we knew about that. We
had many discussions and arguments with Microsoft and IBM about the agenda of
WS-RX (especially the references to their specs such as WS-Policy, WS-Trust,
etc… which are not yet Oasis standards). The result of the discussion was
that we could not make any convergence or suitable agreement. Microsoft and IBM
will spend at least a year and half before WS-RX is voted. Our spec
WS-Reliability has been an Oasis standard for some time now, and it is
currently being used in ebMS-3. Microsoft is not in a hurry of voting WS-RX as
quickly as possible because all what it cares about is that as long as the vote
is close enough to the release of Longhorn, that is fine for them. On the other
hand, IBM is in a hurry of completing WS-RX and it will start to get worried
about this. What will happen in the future is that WS-Reliability is more
likely to evolve to the same content as WS-RX but without the dependencies of
the proprietary specs such as WS-Policy, WS-Trust, and so forth that Microsoft
is trying to push with WS-RX.
Anyway, this discussion is out of subject. My point here is
that when I talk west, you talk east. WS-* specs are not SOA. I was not talking
about WS-* specs. I was talking only about SOA. WS-* specs are low level
protocols for SOA which is a higher layer. I said that Microsoft is not
interested in standardizing SOA. For example, have you seen the architecture of
Indigo being pushed by Microsoft as a standard? Of course not.
[JMC] Some
may say that that is too strong a statement, and that the Semantic Web fits
that bill.
[Hamid]: I disagree with you
that semantic web is The instrument of the future of the Internet.