I know Peter Wright is a smart, independent minded fellow, but in his pro-.Net article in Salon today, he comes across as a paid shill for Microsoft. He presents the best case scenario for .Net as an inevitability, and says lots of crap about Microsoft’s amazing innovation even as he talks about the ideas that Microsoft directly built on. According to Wright’s article, there are two amazing features of .Net, web services and the CIL, which lets you write your applications in any language and use them all together in a happy little soup, let’s look at both of them.
Web services are great. They’re very cool. But they’re not Microsoft’s idea, and they’re not something that initiated as part of .Net. The evolution of web services was well documented as it progressed — Microsoft wasn’t the first to this space. The thing is that there are lots of factors that could limit the growth of Web services. One big problem is that plenty of companies don’t want to offer unlimited access to their data over the Web, or even limited access to it. That’s a huge hurdle. Just as companies don’t have an incentive to work to create interoperable data formats that would enable us to exploit XML to its fullest potential, they don’t have an incentive to create Web services that provide access to their data in an open fashion.
Just as write once, run anywhere has not panned out to be the panacea that was promised with Java, the CIL stuff from Microsoft isn’t going to be the revolution that people like to think it will be, either. Are we really going to see development shops where everyone works in different languages and builds applications that all work together? No way. There are practical reasons why this can’t happen, completely aside from the basic technology issues that have been laid out in plenty of places. This is the sort of thing that gets developers all excited when discussed as a theory, but rarely makes any sort of practical sense.