Microsoft’s Middleware Strategy Explained… Almost
December 3, 2008
Computer Weekly’s “Microsoft’s Middleware Strategy Puts Developers First” by Danny Bradbury is an interesting run down of what Microsoft’s enterprise software approach encompasses. You can read the story here. As I worked through the information in the write up, I was struck that Microsoft is creating tools for others to use. I keep forgetting that Microsoft does not deliver products that are ready to use by people other than developers. The article also reminded me why I have to build functions that other vendors provide in their applications. Mr. Bradbury said:
Now, the firm is planning .net 4.0, which includes some significant enhancements. It will make building REST-based services (resources manipulated via URIs) simpler, and makes it easier to use its own markup language, Extensible Application Markup Language, to manipulate the Workflow Foundation and Communications Foundation services. It is promising a tenfold performance increase in the Workflow Foundation, along with enhancements including persistence control and an enhanced visual interface. But the most important overhaul lies in the Windows Server itself, which will include an application platform called Dublin. Initially released as an add-in pack for Windows Server, Dublin will eventually be folded directly into the server platform. It is a set of extensions to Internet Information Server that make it easier to host Windows Communication and Workflow Foundation services on the platform. It is essentially designed to make the assembly and running of composite applications easier, thereby bolstering the firm’s position in the SOA market. Dublin will interoperate with Oslo, which is Microsoft’s next-generation modeling platform. This will include a new visual design tool called Quadrant, a modeling language code named “M”, and an SQL Server database that will be used to store model schemas.
I have a difficult time keeping the shipping products, the service packs, and the proposed products and their betas straight. I haven’t figured out where Dynamics X++ language fits in, and I am clueless when it comes to the M language.
The scale of Microsoft’s effort is somewhat interesting. The problem for me is the timeline that Mr. Bradbury includes in his article. Scanning its entries, I concluded that Microsoft has lacked a clear vision for the products and services on the timeline. As a result, when I read an article that, in effect, says, “We’re almost there”, I don’t have much confidence in this assertion.
In Europe, there is more interest in using open source software to achieve the benefits that Microsoft asserts its middleware delivers. Open source is far from perfect, but it lacks the proprietary taste of the Microsoft approach. Open source doesn’t market very well. When I read about open source software, the information has a more concrete like feel. The job lock in for a Microsoft developer who presumably is conversant with Microsoft’s time line knows that there are consulting fees available for someone who can make Microsoft systems work.
Microsoft’s middleware approach has a component not addressed in Mr. Bradbury’s article–costs. Figuring out how much to budget for a software system is tough. Microsoft makes the implied integration of its middleware approach operate like a cost reducing diet for information technology. The potential problem is the same as for weight loss schemes. There’s some savings, but then the person on the diet gets fat again. Cost control under this type of control plan becomes difficult, if not impossible.
Stephen Arnold, December 2, 2008
Stephen Arnold
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
