is OO any use?

Hi,

This is not a troll, so please forgive me if it comes across as such.

I'm writing because I'm in the process of forking pic18forth (which
generates sluggish but ultra-compact code) to a forth that generates
pure machine code (way faster, but nowhere near as compact).

This fork is producing a forth that is very 'vanilla' and
architecture-neutral, with the possibility af implementing the same
language, API, runtime environment etc for other PICs, and even other
ucontrollers such as the various AVR lines.

Also, this forth will be very usable as a backend for other language
compilers, such as python (subset), C, Java subset.

What I'm asking here is - is there any real practical value to be gained
from supporting OOP in a PIC runtime environment?

I successfully implemented a reasonably efficient OO layer that supports
most of the OO basics like inheritance, polymorphism and dynamic
allocation (but no gc or refcounting), as well as early and late binding.

Some possible applications for OO on PIC might include:
 - controlling multiple analog inputs
 - multiple serial ports, using the USART plus bit-bashed serial I/O
   on other port pins
 - multiple LCD display modules
 - if interacting with multiple other PICs (eg, via serial multidrop),
   could represent those other PICs as objects

Your thoughts?
Is OO on PIC a useless wank, or can it have real practical value?

-- 
Cheers
David