Hello BLS,

> Georg Wrede wrote:
> 
>> (FWIW, I'd sure prefer to fly with ADA or D, than with C.) :-)
>> 
> So you better fly with Boeing !
> 
> "ASTREE is a static analyzer for C programs that proves the absence of
> run-time errors in critical embedded software.
> It has been applied to the flight control software of the Airbus 340
> and
> 380 airplanes. "
> ASTREE is written in OCAML.  // http://www.astree.ens.fr/
> 

If you can *prove* it's correct, I don't care if it's in BF I'll fly on it over something you can't. OTOH I /think/ Ada programs can be written so they won't compile if they can't be proven to be correct, at some level. 

BLS wrote:
> Georg Wrede wrote:
>>
>> It'd be nice if some of them came to this NG.
> 
> will meet one of the leading guys this summer, let's see.

Excellent!!

>> (FWIW, I'd sure prefer to fly with ADA or D, than with C.) :-)
> 
> So you better fly with Boeing !

LOL

> "ASTREE is a static analyzer for C programs that proves the absence of run-time errors in critical embedded software.
> It has been applied to the flight control software of the Airbus 340 and 380 airplanes. "
> 
> ASTREE is written in OCAML.  // http://www.astree.ens.fr/

Yeah. In the old days there was an entire industry selling lint programs for C coders, and some of them cost more than a mid-size car.

Then came C++, and the number of code lines written (in several languages) for the sole purpose of checking that the C++ program even remotely tries to do what it's supposed to, is simply disgusting.

A programming language should make it possible to write code where the immediate purpose is clearly visible at a glance. I believe D is such a language. A professional programmer who /wants/ to code clearly, can do it. But with C++, I really don't think so.

Walter Bright wrote:
> Is this a good idea?

Definitely it is a good idea, as I do not do any D1 development for years. I use D2 exclusively...