#ObjectOriented
4 messages in this thread
>As far as I am concerned, OO makes programs slower, bigger, more buggy, …
>than plain good (not too structured) programming.
I find this kind of statement w/o any basis. I want to hope I'm
misinterpreting it, but your statement seems clear cut and in context.
Surely there are OO scenario's/designs/languages/etc that can bring this
about, sometimes, but surely this can be this case in non-OO scenario's too.
This tends to reduce down to a common argument about whether C++ is
slower, larger, more complex (you didn't say this), more buggy,
uglier (you didn't say this), etc than C. Although I find this kind
of thing with a concrete answer, I find it terribly misguided.
To me, I'm given a problem to solve. If OO fits the bill, then it does.
I know my real world, my goals and needs, staff considerations and training,
short and long term issues, re-use considerations, etc, and will decide from
there. Moreso, please note that there is seperation from design and
implementation (though the cross-over can become real, it is still
distinct).
Re "plain good (not too structured) programming": what is that?
Surely a roundabout (hacking??) approach is generally not optimal for most
real code. Or did you mean something else?
>>I find this kind of statement w/o any basis. I want to hope I'm
misinterpreting it, but your statement seems clear cut and in context. Surely
there are OO scenario's/designs/languages/etc that can bring this about,
sometimes, but surely this can be this case in non-OO scenario's too.<<
Haha, if you express it this way I have to agree with you. However, I have
never been able to write a program that was smaller and faster with an
OO-language than in another one. MicroS**t claims their programs are OO. If so,
it is nearly reason enough not to try to use it.
>>Re "plain good (not too structured) programming": what is that? Surely a
roundabout (hacking??) approach is generally not optimal for most real code.
Or did you mean something else?<<
I did speak of something else. I am very much against hacking (not for pleasure
of course!). What I meant is the kind of reasoning that let the GOTO
instruction present in PASCAL and the kind of reasoning that made the EXIT LOOP
instruction appear.
Bart Van Bockstaele at 100574.2352@compuserve.com
… On AutoPilot and Amiga all the way from Belgium NOT from Mars!…
>>I find this kind of statement w/o any basis. I want to hope I'm
>>misinterpreting it, but your statement seems clear cut and in context. Surely
>>there are OO scenario's/designs/languages/etc that can bring this about,
>>sometimes, but surely this can be this case in non-OO scenario's too.<<
>
>Haha, if you express it this way I have to agree with you.
Well, what other way is there to express it?
You are saying that OO is always larger/slower/whatever.
It's not the case and only brings up part of the picture.
> However, I have
>never been able to write a program that was smaller and faster with an
>OO-language than in another one.
I've written many. Nonetheless, this is comparing apples to oranges
in the current context.
>MicroS**t claims their programs are OO. If so,
>it is nearly reason enough not to try to use it.
I'd hardly use MS as an example of quality software, no matter what they say
or think.
Again, as well, programming languages and methodologies don't guarantee
anything, nor do they write or design our apps for us. We are still the
vital part, and there is no excuse for failing to make the effort to put
the pieces together correctly. Plugging incorrect values into a mathematical
expression is sure to get the wrong answer, no matter how exact and correct
the expression is. It's not different with OO or anything else for that
matter.
>What I meant is the kind of reasoning that let the GOTO
>instruction present in PASCAL and the kind of reasoning that made the EXIT LOOP
>instruction appear.
I don't follow. Do you just mean that ok is good enough?
Certainly goto and exit loop have their places. As with any feature
the savvy is not the abuse the situation and to do things as appropriate and
applicable.
>>You are saying that OO is always larger/slower/whatever.<<
Not really. What I meant is that using an OO-language you mostly end up with a
program that is larger/slower/whatever. Which can nearly not be untrue.
>>Again, as well, programming languages and methodologies don't guarantee
anything, nor do they write or design our apps for us. We are still the vital
part, and there is no excuse for failing to make the effort to put the pieces
together correctly. Plugging incorrect values into a mathematical expression
is sure to get the wrong answer, no matter how exact and correct the expression
is. It's not different with OO or anything else for that matter.<<
Agreed. Absolutely and totally.
>>Certainly goto and exit loop have their places. As with any feature the
savvy is not the abuse the situation and to do things as appropriate and
applicable.<<
Again, I agree. However, many, many programmers here in Belgium have even been
thaught this exit thing doesn't even exit.
Bart Van Bockstaele at 100574.2352@compuserve.com
… On AutoPilot and Amiga all the way from Belgium NOT from Mars!…