Conference turris::languages

The base topic title has been changed, at the request of
a reader and with the permission of the original author.

Nevertheless, I believe the original title raises an
important, interesting point:  Based on this data, Ada is more 
popular than C & C++ (combined) as an intro programming
language, by a factor of about 1.6 to 1.  While I realize that
Universities don't necessarily view their role as pragmatic 
training of future software engineers, it still says something
about their view of the two languages.

   Gary

    iam taking a course in software engineering right now at northeastern
    university, and one of the first things the teacher said, is that
    he thinks ADA is dead now.
    
    he wants us to write the 2 software projects we have to do in C and
    then in C++.
    
    i have no idea on what basis he said the above.
    
    \nasser
    

    Curious.  Is he a full-time instructor or a continuing ed one?  I know
    at Daniel Webster where I teach, there was quite a variation of
    opinion between those two groups, with the full time educators being
    more pro-Ada.

    re .2
    
    Yes but you aren't taking your first course in programming I would
    assume, Nasser? Noone would argue that C is a good language in which to
    teach people to program, or let him step forward who would. Assuming
    that you're doing diploma or post-grad or even just later year under-grad
    stuff your university may take the view that you are adult enough to play
    with loaded fire-arms.
    
    P.S. How many days to go? (-:

        .4

    actually it is NOT a programming course, it is a software engineering
    course . (big difference :)

    i had a choice of an elective, so i choose an easy course because
    iam tired, and i picked this one.

    as for the teacher, yes, he is a FULL TIME teacher.

    later on, he said some students from the industry called him and
    complained to him about saying this on ADA, he then told us he
    did not mean that ADA is dead, he meant it is not doing as well
    as C or C++ or something like that.

    >you are adult enough to play with loaded fire-arms.

    i dont like guns. thank you. guns are bad for you.

    next project will be in C++, so that will be fun, i have not coded
    much in C++, so it is a chance to dig in it more. 

    i agree that ADA is a better langauge to teach programing with. 
    
    i like ADA myself, and it is much more readable thasn C or C++.
    
    
    \bye
    \nasser

    Re: Ada/C/C++ and a sw. eng. course:
    
       Maybe it's because if you use Ada the language issues are too
    trivial from an engineering point of view?  I mean, you don't have the
    challenges in managing bugs and schedule slippage like you do with C
    and C++... :-) :-) :-)
    
    					-mjg

        .6

    good point \-mjg :)

    but you must admit, ADA is more work to write with, offcourse the
    payoff is more, but in C, one can like jump in and start cranking out 
    code, with ADA it is a slow start. 

    *just* as an example , in external linkage, in ADA if i have a function 
    written in language X where X is not ADA, 
    and i want to use that function and to call it from the ADA program , 
    in ADA you have to write a not so simple interface to it, (ok we can
    argue about this point) , especially if the foreign routine passes 
    parameters differently than the ADA allows (say, a function that 
    modifies one of its arguments).

    but in C, it is a piece of cake to call that foreign function. you
    just write a prototype  and link to it, you dont even have to 
    write the prototype if you are lazy ;)
    

    iam not saying this way is better, i think the ADA way is better
    because it force you to be clear what the interface is, and the
    compiler will type check the interface for you, but for many (iam 
    taking about students ), this might be too much work just to call a 
    function not written in ADA from ADA.

    \nasser

    Re: .7:
    
       Since we're down this rat-hole, the problem is that Xlib and Xt are
    written to be C-only interfaces to those parts of the X11 system.
    
       If someone had written native Ada bindings into Xlib and an
    Ada-native toolkit, I guarantee you that life in Ada would be much
    easier.  (For your example, those "functions" would have been
    procedures.)
    
    					-mjg

    here is a statement from a book "software engineering" by
    sommerville, about coupling in OO systems:

    "inheritance in OO system, however, leads to a different forms of
    coupling. Objects which inherit attributes and operations are coupled
    to their super-class. Changes to the super-class must be made carefully
    as these changes propagate to all of the classes which inherit their
    characteristics."

    \nasser

>       If someone had written native Ada bindings into Xlib and an
>    Ada-native toolkit, I guarantee you that life in Ada would be much
>    easier.  (For your example, those "functions" would have been
>    procedures.)

     A friend of mine who works at Rational has ported some X pieces (I
don't know which ones) to Ada.  He also asked Rational to have them put
in the public domain (I don't know if he was successful).  You can reach
him (Gary Barnes) at [email protected] (please don't deluge the guy - if
somebody gets the information, post it here so others don't have to
follow the pointer).

Re .8:

What does the X Windows system have to do with .7?
				/AHM

    Re: .11:
    
       Nothing -- I was stupid and mistook .7 as a critique of using Xlib
    from Ada instead of native Ada bindings for X.
    
    					-mjg