Re: REST, uniformity and semantics

On Friday, May 16, 2003, at 01:03  PM, noah_mendelsohn@us.ibm.com wrote:
> FWIW:  note that I touched only the first paragraph.  The WSDL 
> paragraph
> was lifted unedited from Ian's draft, and was included only for
> completeness.  When you refer to the last sentence, I'm not sure 
> whether
> you're referring to the 1st Para (which I editied) or the 2nd (didn't).
>

I felt that the entire section should be taken into account as we finish
off the current WSA draft.

The final sentence to which I was referring was:

> However, to represent safety in a more straightforward manner,
> it should be a property of operations themselves, not just a feature of
> bindings.

I take this to mean that one should not infer safety from the type
of binding, because if you do you'll be up the creek sans paddle
if you want to use a different binding (over, say a JMS-compliant MOM)
which doesn't have any implied semantics about safety.

Of course, this view leads rapidly to the position that HTTP under SOAP
is just a piece of minimalist plumbing, and one should not rely
on those semantics of HTTP which do not have natural analogues
in all of the alternative transports (MOM, BXXP, SMTP, etc.). But this
introduces what we might characterize as an "architectural tension"
with respect to what Bryan Thompson wrote earlier:

> With reference to the working draft of the WSA requirements document, 
> AG003
> provides that the WSA MUST be consistent with the current and future
> evolution of the WWWW and AC011 reads "is consistent with the 
> architectural
> principles and design goals of the existing web."
>
> With respect to my understanding of the history of the web, REST is
> an architectural analysis that is integral to those architectural
> principles and design goals.  The TAG also states "Understand REST" as 
> a
> "Good Practice" in http://www.w3.org/2001/tag/webarch/.

Geoff

Received on Friday, 16 May 2003 15:07:16 UTC