On Jan 14, 6:40 pm, Erik Funkenbusch <e...@xxxxxxxxxxxxxxxxxxxxxx>
wrote:
> On Mon, 14 Jan 2008 22:17:46 +0100, Peter Köhlmann wrote:
> > Erik Funkenbusch wrote:
> >>>>>>>>>>> Why would they need Microsoft's permission to publish a Tomcat
> >>>>>>>>>>> benchmark?
> >>>>>>>>>> I suppose if they are running Tomcat on Windows it might fall
> >>>>>>>>>> under the benchmarking clauses of the Windows EULA.
> >>>>>>>>> What clauses might those be? Have you actually read the Windows
> >>>>>>>>> EULA? There is no such clause.
> >>>> Try again.
> >>> /quote Erik F
> >>> What clauses might those be? Have you actually read the Windows EULA?
> >>> There is no such clause.
> >>> /unquote
> >>> Try again
> >> There is no such clause that prevents you from java or tomcat benchmarks
> >> on Windows. Plain and simple. Why is that so hard for you to understand?
Assuming that Java makes no calls to any part of the .NET library,
ever, Erik might be right.
I'm surprised that Microsoft left that loophole open.
Does this mean that we can assume that Linux makes a faster Java
engine than Windows?
Could Microsoft decide that because the Java JVM does have the ability
to call the .NET library, that it is covered by the .NET benchmarking
restrictions?
Is there a possibility that Mono is so much faster than
Microsoft's .NET implementation that Microsoft doesn't want those
benchmarks published?
> Nice try at trying to twist words, peter. But here are my *EXACT* words.
> thad05:
> I suppose if they are running Tomcat on Windows it might fall under the
> benchmarking clauses of the Windows EULA.
> Me:
> What clauses might those be? Have you actually read the Windows EULA?
> There is no such clause.
There are restrictions on benchmarking, but for Windows 2003,
the restriction is on .NET only.
It looks like this particular benchmark cannot be
manipulated by Microsoft.
Still, I'd cache those results somewhere, because I have a funny
feeling that Microsoft will make them "disappear" real soon.
> To date, you have not produced any benchmarking clause which "running
> Tomcat on Windows might fall under".
Let's hope you are correct Erik. It will give us the ability to do
side-by-side benchmarks of Windows and Linux.
> I was entirely correct when I said there was no such clause.
Assuming that Microsoft doesn't find some way to claim that since the
Java JVM calls the library that contains the .NET code, that the
benchmarks still come under the .NET benchmarking restrictions.
Microsoft's legal department has a remarkably good imagination when it
comes to creating clauses that look innocent, but turn out to be much
more comprehensive, but it's only discovered after the violation has
occurred.
> Of course you can't seem to understand simple context (or deliberately
> ignore it), so I doubt you will understand this.
Previous versions of the Microsoft Windows EULA have had restrictions
on benchmarks, and even when the restrictions seemed innocent,
Microsoft has tried to press their case.
When the benchmarking restrictions are taken into court, the Judge
doesn't always agree with Microsoft, but even then, if Microsoft
starts to haul you into court, the cost of lawyers, hearings,
discovery motions, and just getting to a preliminary ruling can be
thousands of dollars, often a few million dollars.
> I will rephrase, just for you though. There are no clauses in the Windows
> EULA that prevent benchmarking of the OS or applications that run under the
> OS that are not .NET based.
Assuming that this isn't another round of Microsoft's "creative
Lawyering", you are right.
> Tomcat and Java have no benchmarking
> restrictions under Windows, as such claims by Rex and others that there are
> such restrictions are wrong.
At this point, I will grant you that point.
Still, keep the benchmarks in cache, because I do suspect that they
will disappear soon. Microsoft does have other ways to make it
disappear.
|
|
