Strong type checking ... (not really) 
Author Message
 Strong type checking ... (not really)

This really have nothing to do with Eiffel either, maybe we should move the
thread to alt.religion.computers


 > Well as usual the C thought police from AT&T come out.
 >
 > C people do share this trait with COBOL people, that
 > having learnt one language they don't want to move onto other
 >  and better things.
 >
 > Your above sentence also reads arrogantly (did you intend it thus).
 >
 > The assumption being that languages apart from C are futile.
 >
 > Surely, you are talking tongue in cheek, so that we can all sit back
 > and have a laugh.

 > Wow, the thought police really do keep track of you.
 >
 > the things Dennis Ritchie had said.  He did say as I remember that
 >  C was flawed and quirky.
 >
 >  Of course I suspect that
 > some of your mud will stick, and keep the "true believers" happy in
 > their faith that anyone who might be critical is not to be trusted.

Sigh!

On getting Brian's paper

Quote:
>I have tried this already, and got a message back with some kind of error.
>But I am persevering, I can assure you.

If Ian Joyner is having trouble with the research address then try the

the research one.

Quote:
>Wow, the thought police really do keep track of you.  I can't remember
>what I put in the HOPL2 post now.

====> sent out july 25, 93


Quote:
>I would interpret Ritchies paper to suggest that C and C based languages
>really are past their prime in terms of usefullness. This is really
>old technology we are talking about now. Of C's success, Ritchie observes:
>"C is quirky, flawed, and an enormous success.

I think we should have the full quote:

        "C is quirky, flawed, and an enormous success. While accidents
        of history surely helped, it evidently satisfied a need for a
        system implementation language efficient enough to displace
        assembly language, yet sufficiently abstract and fluent to
        describe algorithms and interactions in a wide variety of
        environments."

It is a matter of interpretation whether C/C++ is past its prime. I
have a hard time finding support for your interpretation in Dennis'
paper.

====




Tue, 12 Mar 1996 00:30:10 GMT  
 Strong type checking ... (not really)

Quote:

>If Ian Joyner is having trouble with the research address then try the

>the research one.

Thank you I now have it. netlib did refuse my first request for some
strange reason. I got it on the second attempt. An interesting paper,
much of which I agree with. However, it focusses on many specific
aspects of stock standard Pascal. I had to remember back to my student
days 15 years ago, and its implementation on a CDC {*filter*}. In brief
many of Kernigan's reasons were why I switched to ALGOL and Simula. I
too found Pascal restrictive when I got back to it. However, the Pascals
I have used since are Apples Object Pascal, and Unisys A Series Pascal.
Obviously they provide some non standard relief to standard Pascal's
flaws, and Kernigans well founded concerns.

Kernigan's concerns however do not apply to Pascal (ALGOL) like languages
as a whole. They in no way justify the criticisms of people like
Mark Terribile, who trot out this paper against everything else,
including Eiffel. In fact I find that Eiffel directly addresses many
of the flaws that Kernigan points out in standard Pascal. There were
only a few things where Eiffel's philosophy is obviously different, but
I think these can be defended. I obviously can't do justice to
Mr Kernigan's article here. I will save that for another time. But the
conclusion of this thread that Mr Terribile started is that the comments
in Kernigan's article in no way apply to anything outside of stock
standard Pascal, so please stop trying to discredit anything other
than C by its use. In fact he is doing Kernigan and many others a
disservice by using the paper as he does. He needs to read it more
thoroughly. It is a good yardstick to judge other languages, but
Mr Terribile will find that his judgement is seriously impaired.

I will throw in a quote though, even though I am in danger of being
accused of misquotation in many months time. Kernigan concludes:

"There is no significant way in which I found Pascal superior to C,
but there are several places where it is a clear improvement over Ratfor."

Now I could read that as Pascal is superior to C, but in no significant
way. But I will just take it as, he did not conclude that C was
superior to Pascal either. It does not take much to remove these
bad points from Pascal, as many implementation extensions have
shown, and languages like Eiffel (and ALGOL) prove. However, the flaws in C
and C++ are far more serious. But as usual certain C people (and as
I keep pointing out, hopefully the minority) are focussed on the
speck in their brothers eye, while ignoring the log in their own.
Such is the effect of monocular vision, and it seems myopic monocular
vision at best.

Quote:
>====> sent out july 25, 93

>>I would interpret Ritchies paper to suggest that C and C based languages
>>really are past their prime in terms of usefullness. This is really
>>old technology we are talking about now. Of C's success, Ritchie observes:
>>"C is quirky, flawed, and an enormous success.
>I think we should have the full quote:
>    "C is quirky, flawed, and an enormous success. While accidents
>    of history surely helped, it evidently satisfied a need for a
>    system implementation language efficient enough to displace
>    assembly language, yet sufficiently abstract and fluent to
>    describe algorithms and interactions in a wide variety of
>    environments."
>It is a matter of interpretation whether C/C++ is past its prime. I
>have a hard time finding support for your interpretation in Dennis'
>paper.

Thank you for finally posting the entire section.  Notice I preface my
remarks carefully, as "I would interpret", not Ritchies interpretation.
Also it shows, I have not misquoted Ritchie at all.  I took a
deliberately provocative section.  Taking the quote out of context, does
not change Ritchie's observation that C is flawed and quirky.  (Or is
there something I am missing here, like a dangling pointer that only C
people seem intelligent enough to pick up.) It merely focusses on it.
You can only make accusations of "taking things out of context", when
the original sense is changed.  Therefore I find your intimation that I
had deliberately misquoted Richie in another post quite unjustified.

Again I apologise for this use of comp.lang.eiffel, as the defence is
necessarily longer than the accusation.



Wed, 13 Mar 1996 17:36:45 GMT  
 
 [ 2 post ] 

 Relevant Pages 

1. Strong type checking...

2. Strong type checking + Heterogenous con

3. Strong type checking is good, but...

4. Strong static typing does not work

5. Flying With Python (Strong versus Weak Typing) [actually, Static vs Dynamic typing]

6. naive type checking II (not so naive?)

7. Strong typed variables

8. Strong typing header needed

9. Strong Typing

10. Strong/Weak typing

11. Escape from strong typing (was: swap(x,y))

12. Strong vs. weak-typing in OO design/languagges

 

 
Powered by phpBB® Forum Software