Outlook on final optimizations

[Dan Cook]

(Taken right from the post I made back at HLz80, I thought I'd just
repost here for reference. 5 people here the next day! You guys are
quick!)

I realize that demand & usability is more important than underlying
efficiency, if this is going to get off the ground. Therefore, this
seems to be the order of importance of optimizations that need to be
made:

- Integrate && || operators and their "and" "or" counterparts (MUST)
- Integrate * / % operators (insert default routines for them) (MUST)
- Allow usage of string & array literals (MUST)
- Make values work with recursion automatically via the stack (MUST)
- Store variables within instructions (saves more space than share
addresses) (REALLY SHOULD, efficiency)
- Assembly output optimizations (REALLY SHOULD, efficiency)
- Keyword to keep unused functions rather than removing them
automatically (SHOULD, technicality)
- Integrate signed values (DEMANDED, but NOT NEEDED)
- Integrate ++ -- operators (NOT NEEDED)
- Smarter Macros (stricter formatting, requiring specific data types)
(Would make easier usage).

In all honesty, I doubt that I will ever bother with the last 3 simply
because of necessity and the DIFFICULTY of working with the tangled
web of code I call SquirrelBox without messing something else up.
When I consider al the things that I could do better, I realize that
it would almost be easier to start over and recode it all. That is
why I intend to make SquirrelBox DECENT ENOUGH, and then make a better
language altogether. Some of the features I would like to add just do
not fit into the fundamental design of ADSMBLG as it stands. If it
follows the general syntax and usage though (and I do not see why it
would not) ...ReAntiDisassemblage?

Anyway, I am anxious to get this beast up to par so it still has a
chance!

...

On a side note: What do you see as a necessity from a user
standpoint? If you were introduced to the Antidisassemblage language
yesterday and learned of it's (future or current) limitations, what
would be a deal-breaker? Questions? Comments? What's taking Dan so
long to finish this thing already?