pymacs! :-)

[François Pinard]

[Paul Winkler]

> (explanation of control and meta codes snipped)

> That's all good information, but is it really common knowledge?  (I'm
> serious; that's not a rhetorical question.)

I thought it was rather common knowledge.  On the other hand, I also
thought it should not be, because it is anachronic (in my opinion) that
people still have a need, nowadays, to understand the internal structure
and properties of charsets.

In particular, I'm more than tired with octal and hexadecimal notations.
They surely have an historical interest, I would not deny this.  There was
a time people were writing in assembler or punching wires in BCD machines,
and these people needed to understand bits structures and electron flows.
But this is all past, so why still burden people with these?  Decimal numbers
are just fine: almost everything is table driven anyway (or should be).

So, I struggle in my things and around to fully avoid such notations, and
convinced many people around me.  Not Richard Stallman, however, who stiffly
wants `C-q' to be followed by octal notations -- to take a simple example.
Sadly enough, I really lost my cause with Unicode people, where nobody was
willing to listen to my arguments.  As things stand by now, we cannot really
work with Unicode without having to suffer hexadecimal all over the place.

Finally, I gave up, and my `recode' program which once was perfectly clean
in that respect, is now full of hexadecimal notations to support Unicode.
I guess we need a dozen more years, or so, before hoping any improvement.
In the meantime, let's continue eating bits with a small spoon :-(.

> >Then this routine -- let's tentatively call it `pymacs.key()' -- and the
> >convention for writing its argument, should be precisely documented.

> Of course.

As Steffen Ries pointed out, we can nicely use `lisp.kbd()' for that.

-- 
François Pinard   http://www.iro.umontreal.ca/~pinard