>
>In particular, we argue that HTTP can and will be extended to cover more of
>this space than any of its predecessors. New developments in Web
>infrastructure point the way towards "push" HTTP transmissions, streaming
>data, and integration with distributed object RPC systems.
Presumably by this latter you mean that http can be integrated with the likes
of CORBA IIOP, DCOM and Java RMI: all IP based, all "distributed object
RPC systems".
I've been of the opinion that some or all of these 3 are, in fact, the
"right TP"s
on which to really build a solid infrastructure for a wired future, and that
http is really just one more in the long list you mention of ultimately
unsatisfactory
attempts. In its original conception, it's stateless, it has no amortization of
connections, it's insecure, etc. Sure, it's being spiffed up to deal with such
issues (there's cookies, after all :-). But is it really the 'right'
foundation to
start from, given where we want to end up? (Or maybe we just plan to
end up in different places....)
As I said, it's an opinion, and I can be swayed by good argumentation.
I know that you and Adam
are hardly oblivious to the merits of DOs, so you must have good reason
for rejecting their TPs in favour of http as a starting point. Care to share?
While I'm on this theme, how do clueful folks justify a future for html
in a componentized world?
Ron.
"In other words: there's an Orb-like thingie in just about everything,
supporting a queryable BO that can do meaningful things ?"
--Sandor Spruit