Re: Completeness

At 05:33 PM 9/28/2002 -0700, Jeff Mischkinsky wrote:
>. . .
>The way I interpret David [Orchard]'s argument is that screen scraping 
>html pages in order to fill in forms, while theoretically achievable by a 
>computationally complete system is "too hard" to do today. We do have 
>natural language understanding systems that can be made to do useful 
>things in constrained worlds, but I've spent the last 20 years being in 
>the state of having to wait another 3-5 yrs. . . . .

This is a totally different issue that is independent of whether you take a 
REST approach to Web Services or not.  This is the issue of semantics (or 
lack thereof).  WSDL says NOTHING about the semantics of a Web Service; a 
REST approach doesn't either.  Your client application will not know how to 
fill values into an XML message whose syntax and datatypes are specified by 
WSDL, just as it wouldn't know how how to fill values into an HTML form it 
retrieves via a GET.  In BOTH cases the semantics are absent and a human is 
required to say what to do.

>Please note: I am not saying that there is nothing useful to learned from 
>REST. Quite the contrary. I am in complete agreement with David [Orchard], 
>and all the others who point out that there are many useful lessons to be 
>learned from it. And it would be quite profitable to incorporate the ones 
>which apply in a web services context. . . . .

This sounds like a very constructive idea.  Can you enumerate them?

>. . . I'm just tired of the dogma.

I understand the frustration, but I don't believe it is fair or accurate to 
characterize REST advocacy as dogma.  It is clear that there is a 
significant gap of understanding between REST and non-REST people.  Please 
indicate what would be helpful to you in bridging this gap.

Personally, I think more concrete examples would be helpful.  It is hard to 
draw concrete conclusions from abstract arguments.  For example, I found it 
very helpful a while back when David Orchard mentioned a Web Service 
example and Paul Prescod explained, point-wise, how it could benefit by 
adding a REST discipline to it:
http://lists.w3.org/Archives/Public/www-ws-arch/2002Jul/0270.html .

Some seem to be saying that REST is not always the best choice, even if 
everything can be done in a RESTful way.  Others seem to argue that you can 
do it all with REST and have greater benefits.  I would find it helpful if: 
(a) David Orchard or someone would show a simple, specific Web Service 
application that you think is NOT well suited to REST, and explain why; and 
then (b) Mark Baker or someone else would explain specifically how and why 
they think it could benefit by being implemented in a RESTful way.
.


-- 
David Booth
W3C Fellow / Hewlett-Packard
Telephone: +1.617.253.1273

Received on Monday, 30 September 2002 17:58:53 UTC