Re: Emacs history, and "Is Emacs difficult to learn?"

[Emanuel Berg <embe8573@student.uu.se>]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> But yes, jumping in the middle of an instruction can be useful,
> if the "tail" of the byte-sequence encoding that instruction
> "happens" to be a valid instruction.  E.g. back in the 68000, to
> skip a 4B instruction you could either use a jump, or (more
> efficient) place that 4B instruction in the "immediate constant"
> part of a CMP (compare) instruction whose result you just don't
> use (i.e. this CMP instruction was just used as a 6B no-op whose
> last 4B was free for you to choose as you please).

So, jump = compare(... jump ...), in that case?

Interesting. For sure, that is *far* from FP! What would you gain
with the compare method, um, "compared" to just jumping?

By the way, I think *this* (and the C pointers) is the monk
stuff. And a bit of the point with C is that you have memory
addressing, the binary arithmetic, etc., *if* you need it. You can
get as close as you wish. Not the least for programming for a
specific, peculiar platform... (embedded systems, etc.) Yes, Lisp
(and FP in general) is much more gentle. But let's not get carried
away: if a decent programmer (to begin with) does C for say, one
year, just because it is not FP does not mean the code is without
discipline, not modular, etc. And there are no overflows due to
pointers! That guy will be a very self-confident juggler.

> self-modify the code

?

> Or to use some of the instruction bytes as data-constants (in
> case they happen to contain the same bit-pattern).

This is what I said (almost)! I guess there were no assigned
bitstring to append as a prefix to "quote" such constants? (Not
enough room?)

-- 
Emanuel Berg - programmer (hire me! CV below)
computer projects: http://user.it.uu.se/~embe8573
internet activity: http://home.student.uu.se/embe8573