In message <F539F53ECAF5D111B73D00805F65BAE6854CF8 at WDIGL003>
          "Raab, Andreas" <Andreas.Raab at disney.com> wrote:


> Two questions arise:
> 
> - (for Tim) Do you know of any other places where an indexed primitive with
> multiple use was split into two (or more) named primitives, besides the MIDI
> parameterGetOrSet and Socket listenOn stuff?!
I would consider primitives that look at the number of arguments to
decide what to do as being akin to spitting in public, i.e. a major
crime warranting hanging, quartering and drawing plus harsh words in
egregious cases. It used to be possible to make lame arguments about
'saving precious primitive numbers' but that just doesn't hold up
anymore. Name'em!
I can't remember any other places where this evil was perpetrated upon
an unsuspecting vm. If I do find others, I'll be back. You can run but
you can't hide. Make my day, punk. Do you feel lucky?
> 
> - (for Andrew and/or Tim or anyone knowledgable) Shouldn't the
> TestInterpreterPlugin primitives [man we *really* need a better name for
> this] automatically check for the number of arguments and fail if the
> argument count is wrong?!
Could do, but runtime cost would be an irritant. Fetch the arg-count
from the vm, compare, fail etc. Why, it could make us even more
miserably slow than that paragon of modern computing virtue known as
java.
It could certainly be a useful debug-like option that would be useful
for development time checking

tim

-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Kiss your keyboard goodbye!