Article:		Whither Web Services?
Subject:		The Exagerated Demise of Web Services
Date:		2002-10-28 08:22:19
From:		Mark Baker
Response to: The Exagerated Demise of Web Services
If I understand you and Edd correctly, you're suggesting that redefining "Web services" will somehow save them? I don't buy that. Most users have a pretty clear idea in their head what they are. And most of the Web services "gurus" have a pretty similar model in their heads that they're trying to write down in the Web Services Architecture WG. But that model is utterly broken, as the recent mess of unnecessary choreography/orchestration specs has demonstrated. Web services will most definitely fail; the backlash has already begun, and there's not enough good associated technology in place to turn that around. Plus, whatever chance they might have had, they lost when the major promoters failed to educate people how to build them properly - though they probably didn't even know themselves; those people who did have a clue in those companies, had their voices drowned out by the vast majority who were clue-challenged.

Previous Message

Up

Next Message

• The Exagerated Demise of Web Services
  2002-10-29 01:40:20 Edd Dumbill [Reply]
  
  No, I don't think you understand what I was saying. I have long admired your ardour against the tottering tower of web services specs, and think it justified, but it may have colored your perception of what I wrote.

  
  What I was saying is that the real excitement of web services is basically the same as that from exchanging XML. Whatever protocols rule the roost it's the interoperability and potential network effects that are interesting.
  
  

  • The Exagerated Demise of Web Services
    2002-12-04 17:56:42 Ian Hollingworth [Reply]
    
    Edd

    
    I agree with you that the it is the "exchanging XML" part of the web services that offers the real value today.

    
    Using XML to exchange information becomes useful when it allows a community with a common interest to orchestrate (or choreograph) standardized end-to-end business processes involving multiple organizations and disparate information systems.

    
    Once the community has (with some rigour) specified how the standardized end-to-end business process will operate (a huge challenge in my experience if the business process has any degree of complexity) the community then needs to agree on the structure and content of the XML business documents used to exchange information (i.e. agree on an XML schema and XML tag usage).

    
    Finally, there must be agreement on how the documents are going to be transmitted (messaging) - this is rarely a major issue.
    
    The hype suggests that web services technologies can make multi-enterprise business process integration a simple "plug 'n play" exercise. The theoretical potential is there, but I think this is many years away. In the mean time, SOAP and WSDL have a place as useful tools (among other viable alternatives) that a community may deploy to facilitate "through the firewall" application integration. SOAP and WSDL do not address the most significant challenge however: obtaining agreement up front within a community on a standardized end-to-end business process.