Transcript and Dot Notation

[Geoff Canyon]

On Feb 25, 2006, at 6:01 PM, Dan Shafer wrote:

> Would you rather have:
>
> (a) No object orientation
> (b) OO with the current syntax with poor performance
> or
> (c) OO with dot notation and acceptable performance
>
> I'm not saying those are the *only* choices but they're the big ones.

Off the top of my head, I'd rather have

(d) real tables
(e) text rotation
(f) multi-channel sound support (I put this one in for Scott Rossi)
(g) co-routines
(h) anonymous functions
(i) inlining
(j) a map function
(k) (optional) variable typing
(l) macros

But that's me.

As far as object-orientation goes, I proposed object frontscripts and  
backscripts about three years ago as a way of achieving many of the  
advantages of OO without in any way changing the syntax of the  
language or the way it works.

<opening wide and putting my foot dangerously close to my mouth>

Heck, I think I could implement a significant chunk of OO in Rev as  
it stands right now, and it wouldn't even take that long. Just insert  
a front script, tag an object with a custom property representing its  
"class," test the class of the target in the front script to call the  
appropriate class routine, and encapsulate values in custom  
properties. Automatic inheritance might be a bit tricky, but I'm sure  
something could be worked out.

That would go a long way toward the principals of OO without having  
to change anything.

I'll shut up now -- I'm tired and probably causing more trouble than  
I'm solving.

gc