Comments on: Finally, Web Services for the REST of Us? http://www.fastforwardblog.com/2007/06/12/finally-web-services-for-the-rest-of-us/ Wed, 07 Jan 2009 04:43:46 +0000 http://wordpress.org/?v=2.7 hourly 1 By: Joe McKendrick http://www.fastforwardblog.com/2007/06/12/finally-web-services-for-the-rest-of-us/comment-page-1/#comment-15909 Joe McKendrick Wed, 13 Jun 2007 19:47:48 +0000 http://fastforwardblog.com/2007/06/12/finally-web-services-for-the-rest-of-us/#comment-15909 Great question, Ryan. As I understand it, WSDL 2.0 (Web Services Description Language) supports HTTP binding, which means it can be applied to RESTful services as well as SOAP/WS-* services. Dave Orchard, BEA's Web services guru, puts it this way: "WSDL 2.0 has tried very hard to model RESTful Web services.," and notes that "WSDL 2.0 can describe a variety of REST services." http://www.pacificspirit.com/blog/2005/05/16/witw_wsdl_20_http_binding Dave also reports that he has actually linked to a REST service with WSDL 2.0: http://www.pacificspirit.com/blog/2005/03/02/yahoo_search_web_service_in_wsdl_20 Hugo Haas of the World Wide Web Consortium (W3C) also talks about this capability in WSDL 2.0: http://www.w3.org/2005/Talks/1115-hh-k-ecows/#(30) Have any readers out there worked with WSDL to describe REST services? Great question, Ryan. As I understand it, WSDL 2.0 (Web Services Description Language) supports HTTP binding, which means it can be applied to RESTful services as well as SOAP/WS-* services.

Dave Orchard, BEA’s Web services guru, puts it this way: “WSDL 2.0 has tried very hard to model RESTful Web services.,” and notes that “WSDL 2.0 can describe a variety of REST services.”
http://www.pacificspirit.com/blog/2005/05/16/witw_wsdl_20_http_binding

Dave also reports that he has actually linked to a REST service with WSDL 2.0:
http://www.pacificspirit.com/blog/2005/03/02/yahoo_search_web_service_in_wsdl_20

Hugo Haas of the World Wide Web Consortium (W3C) also talks about this capability in WSDL 2.0:
http://www.w3.org/2005/Talks/1115-hh-k-ecows/#(30)

Have any readers out there worked with WSDL to describe REST services?

]]> By: Ryan Gahl http://www.fastforwardblog.com/2007/06/12/finally-web-services-for-the-rest-of-us/comment-page-1/#comment-15852 Ryan Gahl Wed, 13 Jun 2007 14:04:32 +0000 http://fastforwardblog.com/2007/06/12/finally-web-services-for-the-rest-of-us/#comment-15852 While I believe REST has much value, I also feel the same way about WS-*. The power of the WSDL is really it for me though. In the REST world, I'm probably just uneducated... but is there an equivalent RESTful standard that allows a program to discover a service's interface, like WSDL allows? Without something like that, it means hand coding against a set of methods, and if those methods are ever changed it means refactoring my service consumers, again by hand. With WSDL, I know the methods automatically, and if the service contract changes, my proxies (or service adapters) can auto generate, or adapt, on the fly with only minimal babysitting/regression testing needed. Again, am I missing something, or is this kind of power available in the RESTful paradigm already. Or... is that part of the "the tools are not built yet" comment in this article? While I believe REST has much value, I also feel the same way about WS-*. The power of the WSDL is really it for me though. In the REST world, I’m probably just uneducated… but is there an equivalent RESTful standard that allows a program to discover a service’s interface, like WSDL allows? Without something like that, it means hand coding against a set of methods, and if those methods are ever changed it means refactoring my service consumers, again by hand. With WSDL, I know the methods automatically, and if the service contract changes, my proxies (or service adapters) can auto generate, or adapt, on the fly with only minimal babysitting/regression testing needed. Again, am I missing something, or is this kind of power available in the RESTful paradigm already. Or… is that part of the “the tools are not built yet” comment in this article?

]]>