Re: Choreography and REST

Martin Chapman wrote:
> 
> Firstly lets not argue the usefulness of the factory pattern:-)
> 
> Secondly in your example, I could get a c_obj from a
> trader/uddi/directory, without
> there being any explicit make_c. 

That's a feature, not a bug.

> .... There is  no systematic way to know
> whether a c_obj provider must provide the factory
> or whether that is a private act and the object can be
> advertised/discovered seperatley; 

Technically speaking, it can be advertised/discovered separately. It is
a first-class object. But perhaps it is uninteresting to advertise it
for business reasons. So don't.

> .... nor whether once I create it it can be
> shared or not. 

It can be shared. That's a question that the MPAA and RIAA can answer
authoritatively. After all, a network interface just consists of the
bits required to access it. If those bits can be shared, the interface
can be shared.

> ... Without some syntax that ties them together you really do
> have to look at comments/documentation (heaven forbid).

I don't see any tricky issues above, and even if I did, I don't see how
they related to any capability of WSCI. They sound more like features of
some hypothetical UDDI 4 (heaven forbid!).

> Just my personal opinion but I think design by contract/path experession
> type thingy is a pre-cursor to a
> choreography/orchestration/collaboration solution.

I hope we can be more precise in our requirements definition than that.
;)
-- 
"When I walk on the floor for the final execution, I'll wear a denim 
suit. I'll walk in there like Willie Nelson, John Wayne, Will Smith 
-- Men in Black -- James Brown. Maybe do a Michael Jackson moonwalk."
Congressman James Traficant.