Re: "C" vrs ADA

[agnew@trwrc.UUCP (R.A. Agnew)]

In article <578@sugar.UUCP> peter@sugar.UUCP (Peter da Silva) writes:

>What's wrong with Modula-2? As near as I can tell it has all the advantages
>of ADA, and few of the disadvantages. That's my main objection to the language:

Well for one its not on the DOD approved Language list, but my problems with it
are that it doesn't allow operator and procedure overloading and you can't return
a structured result from a function.( no flames about pointers please! )
These features prevent application oriented (problem-oriented) abstraction
(e.g. package Complex_Matrices which uses Complex_Vectors which uses Complex_Numbers )
These abstractions, and more sophisticated ones, allow me to generate code an order
of magnitude faster than I can in Modula II.
I know, because I have done the same problems in Ada, Pascal (Green Hills, Turbo,
and Sun/Berkeley), Modula II (3 different PC compilers), and C (Green Hills, PCC,
and others).  Of course I'm talking about applications (like adaptive beamforming)
and not system software. Please keep in mind that Ada was intended to write 
applications software ( not the washing machine variety that Ada authors describe )
rather than system software. Of course that doesn't mean that it's not good for
writing OS's and compilers; in fact I've had very good results with Ada in those areas)
	
	The treatment of generics in Modula II leaves a lot to be desired. I miss
case sensitivity in both. Modula II's sets are wimpy, but then that's why I
designed PGL1.