[cpp-threads] meta-question about Ada

Andrei Alexandrescu andrei at metalanguage.com
Tue Oct 18 22:59:44 BST 2005


Boehm, Hans wrote:
> As someone who has been watching this from the sidelines for a long
> time,
> but was never seriously involved with Ada, my impression is that it has
> a few strikes against it, which are mostly unrelated to the
> actual strengths of the current language:

At the risk of beating a dying horse, I don't think any of these 
arguments hold for the academia.

> 1) In my view, the pre-1995 version of the language had some
> deficiencies
> that rendered it unusable for many of us. (Notably, no function pointers
> or real equivalent.)  Thus many of us gave up on it before the "real"
> version of the language appeared.

If I remember correctly, LISP initially had full dynamic scoping of 
names (by a bug in the implementation), and that didn't prevent 
enthusiastic people from fixing it (and even availing themselves of 
dynamic scoping when they wanted it, via special syntax). That's because 
there was a "chose" about LISP. There was no "chose" about Ada, not 
enough for any frenetic group to add the needed feature to the language.

> 2) The standard seems to go out of its way to use its own unique
> terminology,
> causing probably unnecessary communication problems with the rest of us.

Just look at the C++ standard :o)).

> 3) It doesn't use C syntax.

One word: Haskell :o).

I still am just as biased. Also I don't buy the argument that "many 
people in academia seem to look at the military background of Ada and 
reject the language on that basis." I am in no position to provide 
useful insights or data on the issue, but the military does fund a ton 
of research, and everybody's ok with that. Heck, the Net came from the 
military and everybody loves it :o). My own research is in speech 
recognition of Arabic language. The military interest is obvious, let 
alone that Arabic is an interesting language to build models for.

To add insult to injury, the word around the grad student lounge is that 
Ada "is much like Object Pascal, only more boring". Of course there's a 
lot of ignorant bashing in such a statement, but it comes from pretty 
knowledgeable people, and it sure isn't motivating.


Andrei



More information about the cpp-threads mailing list