A Spread Spectrum language

Georg Wrede georg at nospam.org
Thu Jun 19 14:39:25 PDT 2008


There's recently been some talk about the spectrum of a language, as 
measured by the width of abstraction levels a language can handle.

In my Master's Thesis I included a graph I'd drawn, which represented 
[my own, as stated in the caption] understanding of the applicability of 
several languages (C, C++, Forth, GW-BASIC, ASM, Lisp, and some others 
that I had been using at the time) at different levels of abstraction, 
as required by different tasks.

Of these languages, only C++ and Lisp could reach the high abstractions, 
where C++ could only do this because of templates, and the STL in 
particular. (The subject of the thesis was the C++ STL.) In my graph 
there was no upper limit to the abstraction levels reachable with Lisp.

(Theoretically, with /any/ Turing complete language (i.e. a programming 
language in the normal sense of the word) you can solve /any/ problem 
that is solvable in /any/ of the other Turing complete languages. But 
that's not at all the issue here. Only reasonable usability.)

I've read Ray Kurtzweil's The Age of Spiritual Machines. And today I 
read http://www.physorg.com/news132727834.html which seems to go further 
in the same direction. For what it's worth, I do agree with both.

Any language has its heyday. No window of opportunity stays open 
indefinitely. And the day the above writings come true, we'll have 
problems that are not fixable with (any descendant or update of) the D 
programming language.

But in the meantime, I just want to congratulate Walter: D is becoming 
an awesome language that definitely will make waves for a long time.



More information about the Digitalmars-d mailing list