Re: Polysemy in programming langs (was: Why does the meaning of words change?)

Remembers me of my first steps in Assembler on
Apple-II. The whole memory was 48Ko, if I remember
well, and this included the screen memory (8Ko ?) When
you executed an Assembler program, there were no
protected areas, so you could write by error into the
screen memory, for ex (very funny) or even in system
areas (not a good idea). My main concern was to keep
the whole program inside the free area, so I had to
save on every single byte, using machiavelic tricks. I
sometimes spent hours saving 3 bytes.

In fact, at first I didn't even use an Assembler, but
rather typed directly the hexadecimal codes, like:
8C F3 02... The day I decided to buy an Assembler
program (on a 5" floppy disk), it was a giant's step
for... well, maybe not for Mankind, but for me anyway.

The purpose of the software was to allow designing and
using private fonts. I illustrated its possibilities
by making appear automatically some verses from the
Tao-Tö-King, written in Chinese, downwards and
right-to-left, with the French translation underneath
(I don't know Chinese, I flatly copied from a book).
That was the time where I could work 18 hours a day on
a screen + keyboard (no mouse, of course). Anyway, it
was astonishing to see what you could do with a
program of only a few thousands bytes, while Microsoft
Word needs megabytes to do the same (OK, there are
some additional functions, I won't deny it; but Word
doesn't allow you drawing a font: I did).

Ah, those good old times... (it was in the last
century).

--- John Cowan <cowan@...> wrote:
> Ph. D. scripsit:
>
> > Then there was the PDP-8, the first mini-computer
> IIRC,
> > introduced in the 1960s. The only way to store the
> value
> > of the accumulator into memory was via the
> instruction
> > DCA, Deposit and Clear Accumulator. This copied
> the
> > value to memory then set the accumulator to zero.
>
> Yup.  The "accumulator" (AC), for younkers, was the
> main (and on some
> models the only) register, so-called because it was
> used to
> accumulate sums.  The PDP-8's instruction set was
> small enough
> that I still remember all of it:
>
> AND: logically AND a memory location with the AC,
> leave result in the AC
> TAD: add a memory location to the AC, leave the
> result in the AC
> ISZ: increment a memory location; if it becomes
> zero, skip next instruction
> DCA: move AC to a memory location, clear AC
> JMS: move the program counter to a memory location,
> jump to the following
>      memory location (the PDP-8's subroutine-calling
> convention)
> JMP: jump to memory location
> IOT: I/O instruction: pass one of 8 commands to one
> of up to 64 devices
>      (the meaning of each command was
> device-dependent)
> OPR1: depending on various bits, clear or complement
> the AC or the "link"
>      (condition code), or rotate the AC one or two
> bits left or right
> OPR2: skip next instruction depending on the AC's
> value, sign, and/or the link
>
> > (The PDP-8 used 4K of 12-bit words.)
>
> Expandable up to 32K.


=====
Philippe Caquant

"High thoughts must have high language." (Aristophanes, Frogs)


	
		
__________________________________
Do you Yahoo!?
Yahoo! Photos: High-quality 4x6 digital prints for 25¢
http://photos.yahoo.com/ph/print_splash