Microsoft's C# (Sharp) & .NET -- A Heads Up

[Alex Martelli]

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:39815602.18CCB9A0 at hursley.ibm.com...
> Andrew Kuchling wrote:
> >
> > Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> > > This IL stuff is not a big deal intellectually so why doesn't M$ just
> > > release the language definition and sources and let the big & small
guys
> > > play?
> >
> > It would certainly be interesting to see a spec for MSIL and CLR, in
> > order to gauge the complexity of writing an independent
> > implementation.  It might provide a decent chance of unifying the
> > hopelessly fragmented state of scripting languages on Unix.
>
> This is exactly what Jean-Claude Wippler has been trying to do for ages. I
> don't know if you remember Minotaur but it is conceptually the same as
.NET
> except that it used Forth as the lowest layer (the IL).

A key difference appears to be that IL is not designed for easy
interpretation,
but rather specifically for compilation down to machine-code (possibly JIT,
possibly at assembly-installation time, that could be an administrative
choice).

Forth, of course, is an excellent form for interpretation, not particularly
suited
for machine-code generation.  JVM's bytecode seems to be somewhat in
between.  Of course it will be possible to interpret IL, if trying hard
enough
(just as, say, any machine-language can be interpreted, aka emulated), but,
from
what I've read so far, that definitely does not seem to be its emphasis, nor
what
the .NET/NWGS runtime is mostly going to do.

If so, then it would appear to be far closer to OSF's old and beautiful
project,
the ANDF (Architecture-Neutral Distribution-Format), which never really got
anywhere in practical terms -- maybe IL is sufficiently less ambitious, more
focused, and/or better-funded than ANDF... who knows, one can hope!-)


> I have had a look at .NET at it seems really neat, if only it was not
> proprietary and worked across multiple platforms. If that was so then it

It's early to tell, but I suspect that, just as for COM, the specs for IL,
CLS,
CRL, .NET will be available for clean-room reimplementation on other
platforms, if people want to try.  Of course, this is just speculation (I
have
absolutely NO access to any privileged information in the matter), and it's
far from clear that, even if so, then such a reimplementation will indeed
happen (Mozilla's XPCOM could be seen as a cross-platform clone of
COM, but it's still not in a released-stable state, and how many years it's
been since COM specs have been out...?  IL+CLS+CRL+.NET/NWGS
seems to be order of magnitude more work to reimplement, than COM's
far more modest architecture & infrastructure, too...).

> would probably replace JVMs and Java would be just another language you
> could choose to use, rather than one you have to use because of the hype.

I think there's far too much Java deployed out there for JVM's to be so
easily "replaced".  IL, as far as I understand it, does not really emulate
the
typical JVM's functionality; retargeting Java compilers for it, while quite
feasible, would then not be a snap, nor is all of that huge mass of Java
code
going to be recompiled all that easily, I suspect...


Alex