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

RE: WSA Properties (was RE: WSA constraints)

From: Ugo Corda <UCorda@SeeBeyond.com>
Date: Thu, 19 Dec 2002 11:31:59 -0800
Message-ID: <A60C40997573F04C8D778D1B5D799C3B1E4B64@mail2002.stc.com>
To: "Champion, Mike" <Mike.Champion@softwareag-usa.com>, <www-ws-arch@w3.org>

> It would be nice to have a rigorous argument as to which properties would
> have different value in a Web of humans who know how to interpret the
> semantic nuances of the content, and a Web of machines that needs to have
> the semantic nuances rigorously mapped from the data to the software. 

Another interesting way of comparing the publishing-oriented Web (traditional Web) with the transaction-oriented Web (Web services) is to focus on the client machine aspects. 

The publishing-oriented Web is based on client machines (primarily Web browsers) that handle the semantics of incoming documents based on a very limited set of assumptions/semantics. The incoming document's semantics are either presentation-related (e.g. HTML, SVG), with the interpretation of these semantics hardwired in the client itself. Or they are based on application invocation (e.g. the browser sees a Word document and it invokes a Word program that knows how to handle it), with invocation patterns wired in (if the browser does not recognize the document type, it asks the user whether the file should just be downloaded or a particular user-specified application should be invoked). Or the semantics is related to locally installing the incoming document (e.g. a Java applet), at which point the installed program will act upon its own hardwired semantics.

In the case of the transaction-oriented Web, the set of semantics brought to a client machine by incoming documents is much more open ended, and it is not so much based on the MIME type of the incoming document but on its namespace(s) (assuming we concentrate on document-oriented Web services), which in turn are handles pointing to a large variety of possible semantics identification mechanisms. The latest Web Architecture draft being developed by the TAG (see [1]) contains an initial brain storming on the subject (see 3.3.2 and 3.3.3.1). 

While I am waiting with interest to see how the TAG will further develop this subject in its Architecture document, I would like to point out here that this is new territory for the traditional Web (so much so that this subject is still actively debated within the TAG and other architecture/standard fora). I am also pretty sure that the type of experimental knowledge brought forward by REST proponents is primarily based on instances of the publishing-oriented Web, and that nobody really knows today what is the practical large scale impact brought by the transaction-oriented Web upon the REST model.

Ugo


[1] http://www.w3.org/2001/tag/2002/webarch-20021206
Received on Thursday, 19 December 2002 14:32:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:11 GMT