#ObjectOriented
6 messages in this thread
>>It is not a natural outcome of OO as you suggest.<<
>
>From what I wrote I think you will understand I didn't want to suggest that OO
>CAUSES trouble. I affirm however that logic clearly dictates that OO-based
>languages ARE more buggy than other, much simpler, languages.
I don't understand. You are suggestion one, or both, of two things:
(1) Programs written in an OOL is more buggy than programs written in
a non-OOL.
(2) The OOL itself is more buggy than a non-OOL itself (that is, the
language specification, and their compiler implementations).
Although in specific, I can find definite cases to support these arguments,
I can find definite (and more) cases to negate them.
As such, I find both very far from the truth in general.
But I'd rather just say I think it does not have any weight.
Furthermore, if the OOs are buggy, that implies trouble to me. ????
>>(1) Programs written in an OOL is more buggy than programs written in
a non-OOL. (2) The OOL itself is more buggy than a non-OOL itself (that is,
the
language specification, and their compiler implementations).
Although in specific, I can find definite cases to support these arguments,<<
What I meant is that OO languages are much bigger than the old classical ones
and that there are thus more opportunities to have bugs in them.
>>Furthermore, if the OOs are buggy, that implies trouble to me. <<
That's my point!
Bart Van Bockstaele at 100574.2352@compuserve.com
… On AutoPilot and Amiga all the way from Belgium NOT from Mars!…
>What I meant is that OO languages are much bigger than the old classical ones
>and that there are thus more opportunities to have bugs in them.
This doesn't follow and is unfounded.
Seriously note that intuition is often wrong, dead wrong.
This is an important insight that folks, including myself,
tend to fight, but it MUST be accepted.
This is a big whopper in say the C++ vs C world. For instance,
since C++ is more complicated, how can it possibly ever result in
a faster or smaller or less complicated program? The answer is that
this question is misguided, even silly. It can and it does.
The thing is not to fight the logic, but to investigate why.
>>Furthermore, if the OOs are buggy, that implies trouble to me. <<
>
>That's my point!
I'd thought you'd said the opposite, but in any event, since they are not
any more buggy, they are not any more trouble! 😉
>>This doesn't follow and is unfounded.<<
Is it? At least in the case of C, it is a fact that C is now a subset of C++
and not vice versa. Logic clearly dictates that if a program contains more
instructions there are more chances for errors and, statistically at least
there will.
Bart Van Bockstaele at 100574.2352@compuserve.com
… On AutoPilot and Amiga all the way from Belgium NOT from Mars!…
>>This doesn't follow and is unfounded.<<
>
>Is it? At least in the case of C, it is a fact that C is now a subset of C++
>and not vice versa. Logic clearly dictates that if a program contains more
>instructions there are more chances for errors and, statistically at least
>there will.
Please note my previous message about intuition ("it is often wrong").
Logic is wrong here. In fact, the logic you present is illogical and
hence the effect is not the case. I'm doing my talk for SD 95 East
right now, and so probably won't be able to get to your responses to me
for a few more days, but, again, this doesn't follow and is unfounded.
Please expound on "if a program contains more instructions", both what you
mean by it, and what you believe the side effects of that are (not
error-wise, just what it does) and I think we'll have a smoother thread.
Greg,
Is it really so hard to understand that a text of say twenty pages is more
prone to contain errors than one of two pages?
If not, why are people taught to split a big project into small overseeable
ones?
Furthermore, isn't it a normal fact of life that it takes longer to read a
twenty page text than a two page one?
As for reachability, I am quite difficult to reach right now. I took over the
responsibility for the department I am working for and this will give me an
overload of work for the next few months 🙁
Bart Van Bockstaele at 100574.2352@compuserve.com
… On AutoPilot and Amiga all the way from Belgium NOT from Mars!…