Re: What ada 83 compiler is *best*

[Gautier.DeMontmollin@maths.unine.ch]

[about McDo etc.]

> Ada and C++/Java operate in the same domain.  They're all used for large
> system development, real-time processing, applications development, etc. 

They may have domains in common.

> Which you choose depends on the development environment you want.  If you
> want to get close to the machine and control the memory usage, us C++
> because it provides incredible memory allocation capabilities.

Incredible but not portable (including with the same compiler on the
same machine if you change memory model). It's a general problem with
macro-assemblers.

> If you
> want simplicity in memory allocation and want enourmous flexibility in
> portability and UI development, use Java.

Or Ada. Recent experience: a port from Ada83 on a platform to Ada95 on another
of a commercial n*100_000 lines program - tens of packages - needed to
change 5 lines (usage of Ada95 math libs). Can you so easily
port Java to another compiler and platform (real question! I'm interested
in a serious answer) ?

> If you really don't care about
> performance or vendor support and want to comply with obscure and obsolete
> government standards, use Ada.  Your call.

Did you try to compare an Ada project compiled by GNAT with cross-package
inlining on, suppress_all, -O2 options, with its C++ counterpart where
*.o files are just linked together ? Again a serious comparision would be
welcome!

[about government, DoD, bureaucrats, other baddies]

Old story!
Are the C++ or Java specifications less bureaucratic ?
Was Ada invented by DoD bureaucrats ?

>> - the most bugs are found at compile time in Ada (a fraction of a second) and
>>   during debugging sessions in Fortran, C, C++ (it may take hours);

> Can you possibly be implying that C++/C compilers don't find bugs, and the
> Ada somehow produces code without runtime errors by virtue of superior
> compiler technology?

I don't. But a good usage of strong typing and subtypes I catch
range errors even at compile time, which, accepted by other languages would
sum up to _months_ of debugging (in my current project) - not to speak about
languages which don't differentiate integers and pointers...

> To make the statement that Ada compilers - by definition and/or
> technological superiority - make Ada a virtual bug-free language is simply
> ludicrous.

virtual bug-free: no; real few-bugs: yes.

>> - an Ada source is easy to read.

> Again, puh-lease.  Some of the worst code I've ever seen is Ada code.

Was it yours ;-) ?
At least it indicates it was readable enough to allow you to _see_
it was bad code !
There are competitions to determine what some C expressions could _mean_ !

(.......)
> Ada is just a programming language.

Really ? Won't it program for you ?

> Languages are NOT
> at the heart of the software engineering crisis.  They are peripheral
> co-conspirators at best.

Of course the main problem is human.
But I'm afraid you're underestimating the co-conspirators!

(.......)
Why don't you want to accept that people find they might be
more efficient by programming in some language instead of another ?
Or even enjoy programming in that language !

Happy Christmas meal at Le Papillon and don't get so nervous about these
programming languages wars!

-- 
Gautier