Print

Print


"Alexandre E. Kopilovitch" <[log in to unmask]> writes:

> Stephen Leake <[log in to unmask]> wrote:
>
> >... when I am forced to change the design, Ada finds
> >all the inconsistencies for me; C++ just muddles thru, hiding the
> >problems.
>
> It's very true, but ... you may be surprised, but Ada behaviour in
> such a case is not always the desired thing. I really saw (not once)
> the situations where programmers were definitely preferring that
> "muddles thru, hiding the problems" against frustation caused by
> immediate revealing of the problems.

Yes, in the short run, that is what programmers want. Software
engineers, on the other hand, want a solid, long term solution.

> >Ada can do class/object; it can also do other paradigms. If you
> >like class/object for a particular project, go ahead and do that.
> >
> >Ada gives you more choice; C++ restricts your choices.
>
> That's it. C++ explicitly provides *leading* paradigm, you are free
> of burden of choice - note, that this is not a routinely choice -
> you have to choose a paradigm, and that will lead to many
> consequences.

Software engineers prefer choice. The C++ paradigm is good for some
problem domains; Ada is good for more problem domains.

> >> And what is Ada package - which real thing it corresponds?
> >
> >A package with a type and operations is a class.
> >
> >A package with hidden state is an object.
> >
> >etc, etc, etc. Yes, you have to pick a paradigm to use; sometimes that
> >is hard.
>
> And this is the point. Why take such a challenge - and
> responsibility, when you may switch to C++ and rely upon popular and
> respectable single solution?

Because the "leading" paradigm may not be the best for your domain.
And because Software Engineers are paid to think for themselves, not
just follow the herd.

> Well, one can say that Ada in fact has its own leading paradigm -
> the "package". But that paradigm is connected more to the world of
> theories than to the world of observable realities. So, if you deal
> with your problem through theories then you can use the "package"
> paradigm effectively, and all goes fine; but my speech is about a
> situation where you have no theoretical supply and therefore you are
> forced to observe raw reality.

No. What else can I say?

> >> So, if we don't know enough about the problem then all Ada's
> >> major tools become useless.
> >
> >Even if they are "useless", they are not "negative".
>
> Not always. Sometimes useless things become negative - simply
> because of someone's temptation for making use of them.

Fire them!

I'm getting the impression that you are saying "we are trying to let
incompetent software engineers write real programs". If that's true,
you need far more intelligent tools than either C++ or Ada.

> >> Ada facilitates stratification of the "problem space"; that is
> >> strategic advantage when we have enough (perhaps, informal)
> >> knowledge about the problem, but if not - then we are forced to
> >> model/simulate the reality directly, and Ada do not facilitate
> >> that as conveniently as C++.
>
> >Hmm. You have not said why Ada is _worse_ than C++. Well, you did say
> >"Ada doesn't do class/object", but that is not true.
>
> I did not say "Ada doesn't do class/object" - reread please - I did
> say "Ada do not facilitate that as conveniently as C++". Perhaps I'd
> better use word "freely" instead of "conveniently", but for many
> programmers those are the same.

Ok. So do you believe Ada worse than C++, assuming a competent
software engineer?

--
-- Stephe