[
Lists Home |
Date Index |
Thread Index
]
On Sun, 19 Jan 2003 22:00:56 +0100, Gerben Rampaart <VisualBasic@Home.nl>
wrote:
>
> http://zdnet.com.com/2100-1104-981059.html
>
> One of the quotes:
>
> "There's this division of labor that's emerging between those who can
> develop (Web) services and those that can put them together to make an
> application"
I'm not exactly sure what that sentence means. Basically, the problem is
that Web services are well-defined right now only for very simple
interactions, such as the proverbial checking of a stock price or sending
an order. Most "real" interactions that people want to automate with Web
services involve a number of interactions, with the exact sequence
conditional on the results of previous ones. For example, scheduling a
meeting with a group of people or booking travel reservations entails an
understanding of the "choreography" of Web services invocations.
The W3C waded into this because there are a number of partially-competing,
partially-contradictory proposed specs and proprietary implementations of
this kind of thing out there already. One, known as WSCI (Web Services
Coordination Interface?) was published as a W3C Note in June or so; that
was followed by a proposal known as BPEL4WS that was *not* submitted to the
W3C, apparently because (at least one of) the authors could not live with
the W3C Intellectual Property Rights policy. See the Charter of the new WS
Choreography working group at http://www.w3.org/2003/01/wscwg-charter for
more background on the technical problems to be solved.
The gist of the article is to note the split in the Web standards world
between those companies and specs which operate under the RAND license
model and those which operate under the royalty-free license model for
intellectual property. See the article itself and similar recent articles
(such as those on the WS-Reliability spec that a number of companies
preferring the royalty-free model proposed) for more on the politics and
companies involved.
|
