W3C home > Mailing lists > Public > www-ws-arch@w3.org > June 2002

Re: Late binding

From: Mark Baker <distobj@acm.org>
Date: Fri, 28 Jun 2002 14:14:52 -0400
To: kreger@us.ibm.com
Cc: www-ws-arch@w3.org
Message-ID: <20020628141452.K11776@www.markbaker.ca>

On Fri, Jun 28, 2002 at 12:22:28PM -0400, kreger@us.ibm.com wrote:
> Scenario 3 or 4 will work.
> Scenario 3 works if there is a standard PortType that companies that want
> to form adhoc relationships using web services. In some industries,
> medical? travel?, these sorts of interfaces already exist and it seems
> reasonable to me that some (or all) of it will move to Web services.
> Scenario 4 works w/o a standard PortType, but with a standard 'message'.

They'd have to have standard methods too, right?

> Quite honestly Mark, I'm not sure what you are driving at.

What I'm driving at is that the Web already has some very useful
standardized methods, and that Web services should use them.

> Throwing 'gets' at URLs is no more capable of supporting dynamic contract
> negotiation. It has exactly the same problem. Just because you get
> something doesn't mean you know what to do with it once you have it unless
> its standardized, described to death, or already agreed on. Even if its XML
> and has a nice XML schema.

Right, but at least I can get it with vanilla HTTP.  You can't with
vanilla SOAP.

To restate the issue, there's two problems;

- how to get information
- how to understand it

SOAP helps you with neither, but HTTP helps you with the former.

MB
-- 
Mark Baker, CTO, Idokorro Mobile (formerly Planetfred)
Ottawa, Ontario, CANADA.               distobj@acm.org
http://www.markbaker.ca        http://www.idokorro.com
Received on Friday, 28 June 2002 14:04:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:01 GMT