Re: "Size" of Ada vs. C++ is there ))

[minyard@acm.org]

Florian Weimer <fw@deneb.enyo.de> writes:

> minyard@acm.org writes:
> 
> >> > ..and compared to C++, Ada is fairly small. 
> >> 
> >> I'm not sure if this is true if you look at the language core only
> >> (and don't take into account any non-essential run-time library
> >> components, such as containers or valarrays).
> >
> > It's true, especially if you include the language core.
> 
> >  Things like:
> >
> >  * Number of operators
> >  * Number of reserved words
> >  * Size of the specification (Note: This particular item is from vague
> >    memory, but I think it's right).
> 
> 250 vs. 300 pages is not a huge margin.
> 
> What about:
> 
>    * Number of different forms of type declarations
>    * Number of representation control mechanisms
>    * Elaboration issues
>    * Tasking semantics
> 
> These language features are much more complex in Ada than in C++
> (simply because C++ hasn't got some of these features).
> 
> In any case, I doubt that the "size" of the language tells us much
> about the language itself.  Perhaps there are additional factors
> that indicate that the language is too big (for example, if no
> complete implementations exist), but size is not a problem per se.
> For example, I think most if not all Ada programmers appreciate the
> elaboration semantics.  They are rather complex, but without them,
> very annoying problems would arise.

I'm not arguing that point, the "size" of a language is quite
meaningless, otherwise we would all be programming Turing machines :-).
Someone was saying that Ada might not be smaller than C++, and I
was giving some fuel for the debate.

But, it's quite amazing to me that Ada can include all the tasking
semantics, a full set of I/O libaries, all the fancy numeric types,
annexes for real-time, systems programming, information systems,
distributed systems, and safety and security, and still weigh in with
a smaller and more usable specification than C++ and a simpler syntax.

-Corey