Horst Herb wrote:
> On Tuesday 16 May 2006 07:41, Tim Churches wrote:
>> Python implemented on top of the .NET CLR, runs as fast as and in many
>> cases a lot faster than CPython (Python implemented in C).
> 
> PS: IronPython is *not* Python. It is a language very closely resembling 
> Python, that will run  a great many real Python code snippets, but this is 
> where the buck stops. It can be faster because many of the most elegant 
> concepts of Python have not (and can not) be implemented in IronPython - we 
> are talking about 90% Python compatibility, no more - similar as Jython, the 
> Python implementation using the Java VM.

The official Python Web site is not nearly as purist as you are, Horst,
about the Python language - IronPython is listed amongst several others
as a Python implementation - see http://www.python.org/dev/implementations/

> Looks like it is going the way of the Dodo anyway (for practical non-MS 
> purposes), same as all other experiments to build a "faster" Python VM - 
> sacrificing elegance, readability, and functionality for some often only 
> marginal performance gains doesn't strike a chord with  most Python 
> programmers.

Speak for yourself, Horst. We find Psyco, the Python just-in-time
compiler for x86 architectures, and Pyrex, which extends the Python
syntax to allow efficient C-code generation for numeric manipulation,
both to be incredibly useful, and I'm looking forward to using PyPy,
which is Python written in Python, which will eventually allow all sorts
of compiler re-targetting tricks.

> For all I know, Jim Hugunin became a Microsoft employee then, IronPython's 
> licensing was turned into a proprietary "shared source" license, and the 
> Python community lost all interest in it. The plan was originally to have it 
> running on both .NET and Mono, but there is no Mono port now to my knowledge, 
> and the current IronPython incarnation as released by Microsoft is already 
> encumbered by patents and other nonsense.

Well yes, because it is built on .NET, which is encumbered by patents
and other nonsense, which is exactly what I have been saying. At least
we agree on that. Anyway, enough arguments over what is and isn't really
Python for now - at least from me. Over to you for the last word, Horst...

Tim C

_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to