From Ed Foster’s “Gripe Log“:
Now that Steve Ballmer and company have given you all the facts you need to compare Windows and Linux, allow me to add just one little tidbit.
A few days ago, Ballmer published an “executive letter” at http://www.microsoft.com/mscorp/execmail/ in which he invited IT professionals to look at all the propaganda … oops, I mean facts Microsoft has assembled comparing open source and Windows platforms. And while he didn’t point to it specifically, one couldn’t help but notice the white paper at the very top of Microsoft’s “Get the Facts” page (http://www.microsoft.com/getthefacts). “Comparing Microsoft .NET to IBM Websphere/J2EE” is a study commissioned by Microsoft and conducted by The Middleware Company analyzing productivity, performance, reliability and manageability of the two platforms.
Now, there are some interesting facts in that study and there are also some things that could be questioned, but I’m not going to go into detail on its findings or methodology. You can read it for yourself if you’re interested and make your own judgments. Of course, since Microsoft commissioned the study, it will come as no surprise to you that Windows wins. When you pay for the testing, you get to write the test plan.
What interests me most about this particular white paper is that it contains some benchmark results comparing performance of Windows server 2003 and the .Net development framework versus IBM Websphere running on Red Hat Enterprise Linux. Which brings me to that one little fact I wanted to add. The license agreement for Windows Server 2003 states:
“Benchmark Testing. The 32-bit version of the Software contains the Microsoft .NET Framework. Disclosure of the results of any benchmark test of the .NET Framework component of the Software to any third party without Microsoft’s prior written approval is prohibited.”
In other words, Microsoft says competitors need their permission to publish results of a study like the one Microsoft commissioned. If IBM, Red Hat, or some other open source advocate wanted to counter Microsoft’s claims with a study of their own, Microsoft’s license would deny them the right to publish their own set of benchmark results.
Since Microsoft’s censorship clause is probably not legally enforceable, it would theoretically be possible for a competitor to do its own study and dare Microsoft to do something about it. In practice, though, a competitor would probably have a great deal of difficulty getting any of the major independent labs to conduct the test without Microsoft’s approval. And, assuming that the test plan was one designed to show open source in a far more favorable light than Microsoft’s did, it’s highly unlikely Microsoft’s approval would be forthcoming.
Steve Ballmer says customers want factual information to help them answer questions about how open source and Windows platforms compare, and that’s certainly true. Microsoft can and should publish any information it thinks will sway customers to their side. What it shouldn’t do is to try to deny anyone else the right to do the same thing. Sometimes it seems like Microsoft has a monopoly of just about everything, but it shouldn’t have one on the facts.