Re:

[Harry Binswanger hb@xxxxxxxx]

Robert wrote:



And keep your head straight about

which is which, and know what the behavior of the underlying

native|programmed command will be when you hit the one or

the other.  Which is the exact opposite of the "don't
need

to know nuthin'" transparency that Harry was enthusing

about.


Yes, yes (he said, enthusiastically)! I like a situation with both the
*possibility* of knowing and the absence of a *requirement* to know. If
U2 and Xy were transparently integrated, all the user would *need* to
know is that some of the commands launch a plug-in which is written in
user-modifiable XPL.



But Carl says you've tried various things like this over the years, and I
agree that experience is the best teacher. I love induction. But ...
still ... Well, no point in putting forth other points of view. I'm free
to try it.



 It's a steep set of
requirements -- especially if

you didn't write the programming.  Non-obvious glitches

*will* arise, guaranteed -- and they'll come home to roost

... here.  And they'll be artifacts, chimera. 
Ugh.


Even with the brute force method of listing the allowed native commands?








Harry Binswanger
hb@xxxxxxxx