"Stephen Pelc" <stephenXXX@[EMAIL PROTECTED]
> wrote in message
news:47483441.168045546@[EMAIL PROTECTED]
> On Wed, 21 Nov 2007 09:43:55 +1000, "cr88192" <cr88192@[EMAIL PROTECTED]
>
> wrote:
>
>>but, can people readily look at and understand your code?...
>>
>>and, about how large of a project?...
>>50 kloc, 100 kloc, 500 kloc, ?...
>
> I am part of a team involved in maintaining a Forth application
> of 850,000+ lines of source code. The application is handled by
> a remarkably small team. Readability is a management issue more
> than anything else.
>
> "There is not now, nor has there ever been, nor will there ever
> be, any programming language in which it is the least bit difficult
> to write bad code." (Flon's Law)
>
> Stephen
>
yes, interesting.
I guess that this shows that the lang is infact scalable...
well, if people can make use of the lang, that is one thing (probably
useful
I guess).
as for displacing the more "industry standard" languages, that is another.
I have typically implemented 2 different kinds of langs:
smaller special-purpose langs, which have often used either RPN or some
custom syntax.
larger, more involved langs (in recent years, typically involving C like
syntax).
I guess, even if not too terribly much of a fan of s-expressions, in some
ways, I still sort of miss scheme (this was one of the first major
non-C-family languages I ended up learning...).
now, IMO, scheme had a few shortcommings:
s-expressions (good and bad, code is data but the parens are out to get
you...);
non-existence of things like break/continue/return (aka: minor changes in
control flow often requiring not so minor changes in code structure);
....
eventually, I gave up on scheme thinking it was not all that suitible as a
general purpose lang.
my next 2 scripting languages (custom) were partly based on JavaScript
(the
former, sort of like ma****ng together JavaScript and C, the latter,
basically the same language but with a new implementation, a lot of minor
syntactic changes, and some cleanup, but in the process I dropped a few
features, such as full continuations, ...).
eventually, starting earlier this year (about march or so), I went and
wrote
a C compiler framework.
had partly considered going through the pain of implementing a C++
compiler,
but have yet to gather either a good reason or the needed motivation.
I have been minorly tempted though by the idea of re-implementing scheme
(just a lot more "solidly hardcore", aka: direct C integration and
natively
compiled code).
as a result, such a lang might end up being mildly "chimeric", and would
likely require "im****ting" C "modules" (I have already implemented stuff
for
module systems, partly working as a special case of the precompiled-header
system).
this would imply having certain features:
raw structure declarations and the ability to work with pointers;
....
but, as of yet, I don't know...
>
> --
> Stephen Pelc, stephenXXX@[EMAIL PROTECTED]
> MicroProcessor Engineering Ltd - More Real, Less Time
> 133 Hill Lane, Southampton SO15 5AF, England
> tel: +44 (0)23 8063 1441, fax: +44 (0)23 8033 9691
> web: http://www.mpeforth.com
- free VFX Forth downloads
|
