Could Cloud Computing be compatible with Microsoft’s strategy?

I don’t write that much about the overhyped “web 2.0” name, but this article from TechCrunch got me thinking. Could there be a new and an old Microsoft? Could there be the Microsoft we all know (Windows, MS Office, OOXML, monopoly) and another Microsoft (young, web 2.0-ish, cloud-oriented) and blissfully unaware of the methods of the past? I think there is some limited truth in that. It is not the first time that I hear such things, but unfortunately, this “new” Microsoft has not seized the power yet. In fact, I could even point out that it has to take an active role to shape the destiny of the company, otherwise Microsoft will die in the long run.

 I don’t think there is such a clear division between these two “generations”. Surely there are differences between people who think of Windows as the platform for everything else, and the ones who are releasing web services such as Sky Drive and products such as Popfly thinking of the Web as the platform.Now the problem I see is that these two are a bit too integrated within each other than it first looks.

Take the whole buzz about Silverlight 2.0. Silverlight is anything but an open technology, will probably be standardized by the Ecma, and has deep ties within other Microsoft technologies, such as WP/F, WC/F (the core technologies of Vista and .Net). If Silverlight’s use grows significantly, we will be at risk of having a non-interoperable Internet. In other words, the Internet as we know it will cease to exist and people will face the problems they had when Netscape was dead and Internet Explorer 5.5 was ruling the web sites. You will notice that the .Net platform has become yet another tool for maintaining Microsoft’s monopoly through the well-established practice of the “embrace and extend”, unclear legal terms, and the refusal to interoperate at the core of its design. Silverlight has been selectively developed for the Windows and Mac platform; Novell is desesperately trying to do something with their Moonlight project and calling that interoperability.

But let’s take a look deeper in the .Net platform. At its core, the main and probably the most ambitious layer of .Net is the CLR (Common Language Runtime). Touted as a revolution when it surfaced around the release of Windows XP, this CLR theoretically allows for pretty much any language, even those outside of Microsoft realm, to work on the .Net platform. To the best of my knowledge, this stuff actually works most of the time. The problem is, CLR lies deep down the .Net architecture so you pretty much have to trust Microsoft on what they plan to do with their products and platforms and hope they will keep CLR able to churn your own code. It is a major strategic issue and once again, this is not an interoperable technology, and it works at such a fundamental level of the infrastructure (be it the Internet or a traditional applicative stack) that you find yourself locked in with it very quickly. Needless to say, CLR is not open source. You can have some detailed documentation on .Net, even get it to compile in debugging mode, but you cannot modify it.

So here we have SilverLight resting on a WP/F (Windows Presentation Foundation, aka XAML, aka Avallon) and WC/F (Windows Communications Foundation, aka Indigo) and being positionned as the competitor to Adobe’s Flash and the contender for the whole multimedia/interactive layer of websites. That’s where I have some trouble believing about the two Microsofts. To be sure, there is some innovative initiatives at the level of Live services. Technologies such as FeedSync are very impressive and do bring value.

 However, there are still some big issues with the search engine performance, and services such as Office Live are simply not compelling enough until you bind them with SharePoint, another great example of customer’s lock-in and control of data by one vendor.

Last but not least, I’d like to mention here what I see as the paradox of Cloud Computing for Microsoft. I think there should not be one player out there who could ideally be better suited to offer Cloud Computing services (massive storage, grid computing, data ubiquity and web apps) than Microsoft. Its installed base and its control over the productivity market would be their greatest assets in this endeavour.

But I don’t think Microsoft can stop thinking outside of the customer’s lock-in mode, and Cloud Computing is just the opposite. If you want to offer compelling “cloud” solutions, then it’s all about easiness, data control by the customer, data portability, and estimating the value of applications at the level of mere commodities. That is, unfortunately, not what Microsoft’s business model is about. So if there are two Microsofts in the same company, I’m afraid they are just like their formats: not compatible.