

read: benchmarks.
It could be me, but they don't make much sense.
Anyway, i'v learnt no to trust benchmarks over time, simply because they can say whatever you want them to.
on a side note: why the "linux" to "Open-Source" translation on OSnews? is it more politically correct?
ummmm, you didn't read it did you. They were testing open source apps like Pyhon,PHP, Apache, which can run on just about any platform in the world, and even in their own tests here they too them for a spin on windows.
Too bad they didn't do a benchmark with Zope, the Python based application server.
Edited 2006-07-11 07:45
They call this a "benchmark"????
They tested DIFFERENT code/applications/databases on each platform.
What a bunch of tards...could they not fine ONE guy in the whole publication that could write some quick routines that do the same thing in all the platforms?
Man, the things that pass for "journalism" these days is just unbelievable.
As they said in the article (did you even read it?) they had two choices.
1. Use common code/apps/dbs for each platform to see what the general performance is like in the real world. Negative: They would be testing the app instead of the platform.
2. Write their own code/apps/dbs for each platform. Negative: They would be testing how good they were at porting each app rather than the platform.
Either way they would have had people complaining.
Edited 2006-07-11 00:31
Interesting, but one thing that bothered my pedantic ass is "Active Server Pages." They talked about .NET and they talked about ASP, so I can only assume what they tested was ASP.NET and not ASP. The two technologies (ASP and ASP.NET) are so profoundly different that I dislike it when people confuse the two, particularly in a technical article.