Future Focus - Tomorrow's Insights for Today's Decision Makers 
 
Web Services - The Hype and the Hope - Part 2

July 2002

Aaron Kumove -- Managing Director, Horizon Consulting

Web Services, as we saw last month is a low level specification designed to ensure that applications can call on each other’s services regardless of who wrote them, in which language, running on which operating system.  This has the potential (if one believes the hype,) to usher in an era of unprecedented interoperability that the IT industry has never known and to enable B2B e-Commerce to flourish . . .  Maybe . . .  Let’s remember that Web Services is at an embryonic stage of development.  The standards are still evolving, new ones are being developed, and there are few real world applications of substance today.

There still some big unanswered questions and some frankly unsubstantiated claims that need to be set straight.

If all goes to plan then Web Services could enable any application to interoperate and integrate with any other.  That is of course if the vendors can keep from “extending” the standard for their own purposes, (and of course that has never happened before . . .)

Already there are some political cracks in the Web Services camp that threaten the necessary cooperation that is required to make this the ubiquitous integration glue that it is claimed to be.  
The Web Services Interoperability organisation (WS-I) founded by IBM and Microsoft is denying Sun Microsystems equal status as a founding member.  For this to work the vendors are going to have to set aside petty differences to build a bigger pie that they can all share.  If one or two major vendors break away and form their own standards body the promise of ubiquitous integration and interoperability will be dead before it is even born.

You may have heard that Web Services will enable legacy applications to be transformed into discrete components, which can be accessed over the Web, by other applications.  Well, maybe . . . but this frankly has almost nothing uniquely to do with Web Services.  If a legacy application lends itself to being broken into separate components then yes, one could potentially provide Web Services interfaces to these component parts.  That is however an awfully big “if” . . . There is absolutely nothing about Web Services that will offer any kind of assistance in untangling a mass of legacy code and separating it into discrete components, let alone separating the business logic from the presentation.  Everyone with mainframe “green screen” applications knows what a non-trivial issue this is.  
Sorry, no silver bullet here.  Web Services will only help you in exposing legacy applications to the Internet if the applications are first separated into logical component parts devoid of presentation information.  In some cases this may be more trouble than it’s worth and other solutions may be preferable.  

You may have also heard that Web Services will enable B2B processing and collaboration between organisations over the Web.  Again not quite . . . The Web Services model as it exists so far has no notion of a business process.  It has a notion only of a single request of a service and a single reply from that service.  A larger business process will involve many requests and replies of a service or of a number of services to provide a business outcome.  The Web Services model does not however understand this, so if that is what you need you are going to have to custom build it if Web Services is the only tool in your kit bag.  (There are other specifications that do inherently understand the notion of larger business processes such as ebXML, Rosettanet and others . . . These will be examined in upcoming columns.)

The key point to remember is that Web Services is a low level means for one application to call services of another.  Higher level concepts like processes do not exist for Web Services today.  

Web Services also have no notion of transactions.  That means that the standard behaviours we expect of transaction-aware technology do not exist for Web Services.  (2 Phase commit, rollback, etc . . .)  This means that if you want to conduct transactions with a Web Services application you are going to have to build these capabilities yourself.  

Web Services also have no notion of security or authentication so if this is important to you are going to have to build it yourself.

In fairness, there are some draft proposal being floated by vendors to address these shortcomings of Web Services and in time they are likely to be solved.  If however you want to use Web Services “in anger” today (as very few in the world are,) than be aware that you will be building proprietary mechanisms to address these limitations and the very promise of ubiquitous integration and interoperability that is the heart and soul of Web Services will be denied to you in the process.

 

Aaron Kumove -- Managing Director, Horizon Consulting


Want to comment on this issue?  Perhaps you would like to add to the views expressed here, or even disagree violently!
Either way we welcome your thoughts and would be happy to include your views in upcoming issues.

Read Past Issues


Would you like to SUBSCRIBE or UNSUBSCRIBE?

Future Focus is a regular feature in MIS Magazine.  An additional archive of previous Future Focus articles can be found there.  (Just type 'Kumove' in the search box).

PRIVACY:  We do not provide our mailing list or details of any member to any other party under any circumstances.
Copyright © 2002 HORIZON CONSULTING

Horizon Consulting is a leading provider of successful strategy and management implementation services for knowledge economy organisations.  Our clients are world leaders in obtaining strategic advantage through eBusiness and Information Technology. 

Horizon Consulting PO Box 2252, Wellington, New Zealand; Tel: 64 4 939-9944; 
Email: service@horizonconsulting.co.nz