Re: Broken dream of mine :(

[Bas Wijnen]

On Thu, Sep 17, 2009 at 11:56:53AM +0530, arnuld uttre wrote:
> > On Tue, Sep 15, 2009 at 9:56 PM, Bas Wijnen <address@hidden> wrote:
> >> On Mon, Sep 14, 2009 at 03:30:08PM +0530, arnuld uttre wrote:
> 
> > ... SNIP...
> > Anyway, it historically makes sense that a language evolves with
> > hardware.  And C, even if it still has the same name, has evolved a bit
> > as well.  The real evolution is of course in the step to C++, where it
> > really becomes a different language (if you want to use it that way).
> 
> I really don't consider C++ a really technically sane evolution.

Evolution doesn't need to be technically sane. ;-)  I don't consider the
fact that x86 came out as the 'survivor' to be technically sane either,
but it happened nonetheless.

However, I disagree with you about C++.  It is a very nice language IMO,
which only misses a few features.  Python has those features, but misses
other things.  Ah well...

> Right, and I am not genius. are you ?

I don't think so.  That's why I'm saying it's not a good idea to design
a new language before working further on the Hurd.

> I am not influenced by either RMS or Linus's opinions, I like neither
> C nor C++.

Oh I'm sure you are influenced by (almost) any opinion you hear.  You
don't have to agree with it all, but it does change your view a tiny
little bit.

Not liking C and C++ is of course possible and acceptable, regardless of
what was the main influence for that opinion. :-)

> Unlike C, C++ is clumsy and full of monstrous ad-hoc complexity.

Is it?  Not if you use it right, IMO.  But we could of course have a
different view of what is "monstrous".

I think that C++ is in fact much cleaner than C.  The language may not
be so complex, but that means that the constructs you need to get
something done are much more monstrous than what you would need in C++.

> Problem with C is its not very expressive, the point I like very much
> about C is its size. You can do big things with such a small language.

You can, but you need to think of technical details all the time.
That's what I meant when talking about the problem with C that you don't
see while using it: you need to be constantly aware of lots of things,
and that keeps your mind from the actual problems you're trying to
solve.

> I really do want to contribute.

That's good, we need that. :-)

> I have also talked about this with RMS and RMS says he is not
> interested in working on Hurd, that no one should work on Hurd because
> we already have the kernel  and that time should be spent on something
> else (like Guile). Then he complains why people cal it Linux, not GNU
> OS or GNU/Linux OS or Lignux.  I don't agree  GNU needs a kernel (of
> course a microkernel

RMS seems to be very rigid in his opinions (I don't know him personally,
so I can only say "seems").  However, when working on a project that he
isn't personally involved in, it's easy to ignore his opinion if you
disagree with him.  That's good. :-)

> Even after whatever I have said: if there is one place on this Earth
> where I will love to work , that will be one of the FSF's branches :)

On Thu, Sep 17, 2009 at 06:28:23PM +0530, arnuld uttre wrote:
> You wrote a microkernel yourself:
> http://projects.qi-hardware.com/index.php/p/iris/

I know. ;-)

> then I think writing a language will not be as difficult for you (as
> for me who can not even comprehend how Hurd is better than Linux or L4
> , in technical terms)

You are probably right that it is not as difficult, and maybe even
possible.  However, I think it only makes sense to design a new language
if there is a problem for which no existing language is adequate.  For
Coyotos, this was their goal to mathematically prove correctness of the
program.  For the Hurd, I don't see a problem which I cannot solve in
C++ (or even C, with a bit more time).

It's not that writing a language would be impossible; it's just that
it's a large task which we should only start if it is worth it.

Thanks,
Bas

signature.asc
Description: Digital signature