> I don't want to start a flame war by arguing static vs. dynamic
> typing. Personally, I think there are good arguments for both.
Yeah, let's leave the flame wars to nerds and all others psychologically
less developed individuals ;-)
I think what you said about first developing for swf8 compiler and when
the project grows moving to swf10 compiler is a very good idea. That way
you can get the best of both worlds.
Ideally you could tell the compiler if you want it to enforce strict
typing or if you want it to take it easy.
Actually this is already true (in a slightly roundabout way) with laszlo.
> So does the idea of having two different "null" values.
Oh yeah and the differences between String object and String literal and
Boolean object and boolean literal...phew
I think the biggest blunder of all times concerning types was the
omission of boolean datatype (and here we have a list of fairly
respectable languages like C, Lisp and SQL).
Most of the SQL databases still do not have a Boolean datatype!
It really is amazing since the boolean logic (truth tables, predicate
logic etc.) seems to be in the center of computing.
- rami
P T Withington wrote:
On 2009-11-20, at 13:18, Rami Ojares / AMG Oy wrote:
Thanks for the links!
I promise to read them (but that does not mean I am going to understand them).
the collective (40+ year) experience in the O-O community is that
whenever you think you need to say super.super, or directly call an
inherited method in a superclass, you will end up regretting it if you > do,
inevitably at a later time or in some descendant class when the
little "work around" will have been long forgotten.
Well...I think there is even larger consensus in the community about the
dangers of loosely typed languages (like laszlo).
All the odd and implicit rules of type casting keep me on my toes all the time.
Not to speak of all the odd behaviours echoed in the remotest parts of my code
because some variable had an unexpected type causing all sorts of secret black
magic :-)
The implicit casts of Javascript keep everyone on their toes. So does the idea of having
two different "null" values. Brendan will be the first to admit many decisions
that were made in expedience that he wishes he could take back.
I don't want to start a flame war by arguing static vs. dynamic typing. Personally, I
think there are good arguments for both. As I have said before, I like dynamic typing
for prototyping and (what I call) "evolutionary" programming (when you are
quickly turning around prototypes on user feedback). I like static typing for
eliminating the possibility of runtime errors in a deployed product.
If you want to add to your reading stack, here are two really interesting books:
"The Design and Evolution of C++", Stroustrup
"Advanced C++ Programming Styles and Idioms", Coplien
The first provides some really interesting insights as to why C++ (and its
descendants, like Java) are the way they are. Stroustrup is very balanced in
describing decisions that were made for truth and beauty, vs. those that were
made for expediency.
The second is, as I see it, the story of a developer who wishes he could be
developing in Lisp and the agony he was put through trying to emulate the more
powerful features in C++ because his management was sure C++ was the right tool
for the project.
