RE: [xml-dev] Speed in Languages and Browser Architectures

["Michael Kay" <mike@saxonica.com>]

> 
> > If you really need raw speed C or C++ will be always faster 
> than Java.
> 
> No, it won't. It has been proved false by counterexample that 
> C/C++ will
> *always* be faster than Java.  On some problems, Java already 
> is faster than C or C++. 

And even if it were true in theory, it wouldn't be true in practice. The
XSLT experience was that it took three or four times as long to produce a
working processor in C as in Java, and while the C developers were debugging
their memory management, the Java developers were devising* new
optimizations.

Michael Kay
http://www.saxonica.com/

(devizing? doesn't look right. American spelling confuses me.)