[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: New Language to write HurdNG

From: Ivan Shmakov
Subject: Re: New Language to write HurdNG
Date: Fri, 23 Mar 2012 16:04:18 +0700
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/23.2 (gnu/linux)

>>>>> arnuld uttre <arnuld.mizong@gmail.com> writes:


 > I don't do anything because I don't want to use C or C++ to write
 > microkernel.  I have my reasons for this:

 > After 40 years we are still using a language that was designed on
 > some ancient piece of equipment to take benefit of the instruction
 > set of that hardware.  We have machines which are 1000 times powerful
 > than those equipments but we still use the same language, earlier we
 > didn’t have any higher level language that competes with the
 > efficiency of C but today we have choices.  Hardware has grown and
 > technically blasted beyond the levels of imagination but system
 > programming languages have remained there.

        The hardware has gone beyond any imagination of virtually any
        XIX century's person, but the English language that we're
        currently using is still pretty much the same as it was over a
        century ago.

        Is that a problem?  Perhaps.  Should we switch to, say,
        Esperanto instead?  Perhaps, but I find that unlikely to happen.


 > I have these questions:

 > (1) What do you guys think of this ?  agreement, disagreement, hate
 > this crazy idea, ? or look at this guy named arnuld, he is completely
 > insane, thinking of rewriting 20 years of work into some new language
 > in next 2 years.

        The only issue with this idea is that writing an modern OS
        kernel has to be a collaborative task.  When it comes to the
        gaining of acceptance, the “innovativeness” of the design is
        secondary to the number of hands that're working on its
        implementation.  Thus, the first thing to switch to a different
        programming language is to ensure that there're in fact those
        hands (and minds.)  And this means that you're to read lectures,
        write textbooks, etc., first.

        And that isn't like a short path to success.


 > If I go crazy then I may even rewrite entired Hurd as HurdNG in new
 > language.

        This would likely to be a completely insane amount of work, even
        if only the pure typing is considered.

 > NOT RELATED: I was surprised to see Intel Fortran was above gcc in
 > performance everytime. I thought C is the fastest language on Earth:
 > http://bit.ly/GHSFZO

        … And what about Intel C vs. GCC?

        FWIW, I have to admit that those Intel folks know their
        processors quite well, and their compilers are fast.

        Fortunately, I'm not a performance freak, so I may focus on
        freedom instead.  And in that respect, GCC and Clang are still

 > Excerpts taklen from my Blog:
 > http://lispmachine.wordpress.com/2010/03/10/a-simple-concept/

FSF associate member #7257

reply via email to

[Prev in Thread] Current Thread [Next in Thread]