I recall taking some flak from other tech journalists some time back because I still had some faith in the potential of the Web services model. “”It’s hype and we’re sick of it,”” seemed to be the prevailing sentiment.
Since I rather pride myself on being the most cynical person in four out
of any five given rooms; I do not take suggestions to the contrary sitting down. I took great pains to point out to the most vocal critic that:
- While I had written that Web services offered a great opportunity to ISVs, I had also said that would only be the case if competing interests would bury the hatchet and adopt open standards, and
- His mother wore army boots. (You really must travel with journalists some time. We reach depths of pettiness others can only imagine.)
Microsoft’s recent announcement that it will seek a patent covering much of the functionality of its .Net initiative could darken the horizon for Web services development.
The concept of a Web services architecture is a great leveler among developers. Because applications are network-aware, because they can identify and seek out each other by functionality, because they can exchange standardized data freely, smaller ISVs can focus on component technologies without worrying about developing an over-arching architecture to host them. They can stick to the small corner of the transaction picture that they know best and let the network worry about framing it.
This assumes that open standards – XML, UDDI, and SOAP – are applied to the transaction. And while it’s by no means certain, Microsoft’s patent efforts could derail that train.
The patent would cover .Net APIs, potentially controlling who can link what to .Net-based applications. If my trusted-third-party transaction-time-stamping component can XML all it wants and still not interoperate with .Net-based applications, then that’s a substantial chunk of the market that it’s no good to.
Standards in the Web services space have to be open, or they’re worse than useless. And “”open”” means published and accessible. Otherwise, Web services becomes just another internal integration technology instead of a fundamental change to the way software is created.