Al Hospers wrote
> perhaps it is, as Colin K says, a matter of semantics. I think I said
> that in about the middle of the thread. and Director's current mixed
> OOP/procedural state makes it difficult to define the boundaries some
> times.
Hi Al,
OOPs and semantics seem to tango a lot :)
I think one of the most difficult aspects of OOP is the process of
abstraction and identifying classes - especially in the context of
implementing in Lingo. Class definition always comes under scrutiny and
re-evaluation (at least for me) once you start thinking about how the
objects are going to communicate and pass information as necessary. At this
point, there seems to be a need to reconcile the requirement of objects
sharing data with the principle that using accessors is indicative of poor
encapsulation.
To help decide whether a data sharing method is supportive of nicely
encapsulated class rather than being a simple accessor, I use the basic rule
- "does the data represent an abstract idea". While the difference between
passing the percentage played of a movie as a a parameter versus providing
an accessor for the movieTime of the sprite may be trivial on one level, the
former approach is more supportive of the concept that the QTObject is a
fully abstracted object with abstract data whereas the second approach leads
towards thinking "well, if I am getting the movietime from the wrapper, why
not just get the movieTime directly from the sprite?"
> this thread has helped me, and hopefully others, think & rethink all
> of this stuff in a better way.
Yeah - OOPs in lingo is always an interesting discussion (partly, I think,
due to the very diverse background of Lingo programmers)
Luke
[To remove yourself from this list, or to change to digest mode, go to
http://www.penworks.com/LUJ/lingo-l.cgi To post messages to the list,
email [EMAIL PROTECTED] (Problems, email [EMAIL PROTECTED])
Lingo-L is for learning and helping with programming Lingo. Thanks!]
