I enjoyed reading your article, but I think it fails to justify its conclusion. The "placeOrder" may be implicit in your second (simpler) example, but the second example has a very different semantic meaning that the one with the "placeOrder". One uses a verb and the other doesn't. There is no way of knowing what is intended to be "done" with the purchase Order in the second example without an "out-of-band" communication which would tell the person to send this message if they want to place an order. What if they want to know if they the user knows anything about the purchase order? This is a separate agreement. Thus I think you have set up a "straw man" argument in comparing two different meanings or intentions of sending a message. It may illustrate your point you want to make but verbs are important in determining what to do with "nouns". If we only spoke with nouns, and no verbs we wouldn't be able to communicate. I think the same thing applies to communicating on the Internet. But RPC's do work on the Internet. It is what happens with DNS, LDAP and other services that work very well and are the ubiquitous backbone of the Internet. I agree that simplifying communications is a good goal that WebServices fails to achieve. But your example doesn't prove much, in my opinion. The semantics of a request must be accomplished regardless of the protocol to send the information. The protocol isn't that essential, but the semantics of the request is.
Dave Forslund Mark Baker wrote: > On Thu, Sep 15, 2005 at 09:30:18PM -0400, Anil John wrote: > > Agreed RE: the Web. But do the same concepts apply, in the current > > state of technology, to the web service world? > > Ah, I see that you're not familiar with my position. 8-) > > My claim has always been that the Web is a system which can be used to > meet the big picture objectives of SOA; that it's had a "service model", > usable for machine-to-machine communications and integration, since day > one, when TimBL wrote the first Web server. It's just a different model > than that offered by the likes of CORBA or DCOM - a more loosely > coupled, document oriented one more akin to MOM - but just as capable. > > It's also packaged in a manner that makes it very difficult for Web > services proponents to recognize, and to relate it to what they know. > IMO, this is primarily because a key enabling piece of the architecture > happens to share the same name (despite being a very different beast) > with a largely insignificant piece of the CORBA/DCOM-like architectures; > the protocol. > > You might enjoy an article I wrote a few weeks ago on this subject; > > http://www.coactus.com/blog/2005/07/towards-truly-document-oriented-web-services/ > > Mark. > -- > Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca > Coactus; Web-inspired integration strategies http://www.coactus.com > > > > ------------------------------------------------------------------------ > ------------------------ Yahoo! Groups Sponsor --------------------~--> Fair play? Video games influencing politics. Click and talk back! http://us.click.yahoo.com/T8sf5C/tzNLAA/TtwFAA/NhFolB/TM --------------------------------------------------------------------~-> Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/service-orientated-architecture/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
