Re: How to motivate use of OO?

topmind a écrit :
(snip)

I've studied the patterns of changes and variation re-occurance that I
see over the years.

Great.

I have concluded

Where can we read the demonstration (I mean, a logical, rational, well-backed one) that led to this conclusion ?

that the generation/creation
patterns of variations tend to be a *Cartesian Join* of all possible
features. There may be *some* rough treeness, but not enough to rely on
in the longer run.

Let's take it as an axiom until you answer the previous question.

Thus, say there are 26 total features, and we label them from A to Z.
The variations would then tend to be random subsets of the features:

"tend to" ? "random" ?

instance 1: A,C,R,Q,V,Y
instance 2: D,J,L
instance 3: C,G,R,T,V,Z
instance 4: E,F,K,L,M,P,R
etc.....

So, please tell me how OO deals with such.

Nope. *You* tell us how *you* deal with such.

The procedural/relational approach is generally to put the feature
lists in tables, and then a task will tend to look something like:

Task Foo(...) {
if A {....}
if B {
if K {...
} else {
...
}
}
...
if Y {....}
if Z and not M {....}
}

....

OMG


It ain't pretty,

That's an understatement.

but I have not found something better.

What can I say ? Evidence speaks for itself...

Further, the IF
structure tends not to be the same from task to task.

And ?

Polymorphic
solutions tend to assume

"tend to assume" ?

the same IF "framework" for each different
task,

Care to explain ?
.

Relevant Pages

  • Re: Cross-Browser onmousedown JavaScript
    ... It would only make sense to implement the W3C events model if there is a demand for currently unused event features; and if this were happening, nothing would prevent Microsoft from extending the current model, rather than implementing the W3C's model. ... Given this situation, two solutions are available: either change/correct the scripts, or change/evolve the browser's model. ... The choice of changing the model looks like the less painful; after all, scripts do not need to updated, and the browser ends up with more capabilities, right? ...
    (comp.lang.javascript)
  • Re: VFP 9 on Windows 98 SE
    ... and the only reason that you would'nt recoup your ... My point is that if you are not using the productivity features of VFP, ...
    (microsoft.public.fox.programmer.exchange)
  • Re: How to motivate use of OO?
    ... Thus, say there are 26 total features, and we label them from A to Z. ... The variations would then tend to be random subsets of the features: ... There are design patterns to deal with these combinations. ... programmer helping programmers. ...
    (comp.object)
  • Re: z/OS and OS/VS Cobol
    ... This might be a sign as to why the standards were written as they were; ... >often ignore 'new features that are available'. ... >What is the true benefit of 70's style programming? ...
    (comp.lang.cobol)
  • Re: Great SWT Program
    ... A user using a GUI app may notice something ... and notice tedium using a GUI app to do it. ... all available features, in that one can do so by poking around in ...
    (comp.lang.java.programmer)