Re: Low-level vs. High-level programming (was My First Computer)

["M. Pender" <mpender@hotmail.com>]

Michael J. Mahon <mjmahon@aol.com> wrote in message
20030725020918.24310.00000421@mb-m16.aol.com">news:20030725020918.24310.00000421@mb-m16.aol.com...
> M. Pender replied:
>
> I believe you are referring to the three "predeclared files":  OUTPUT,
> INPUT, and KEYBOARD.  Pascal also permits user-declared files
> up to the limits that the system will support.  This implies that the
> limitation you describe does not exist.

Yes, Apple Pascal permitted *one* additional filestream in addition to stdin
and stdout.  IIRC input and keyboard were just different references for the
same stream, which a person could confirm by comparing the system-assigned
file-ids for the streams.

> Perhaps some stalwart Apple Pascal user can comment?

I think a few already have. ;-)

> >That is the standard I/O implementation of ForTran 77 -- which leads me
to
> >believe (but does not prove) that Apple Pascal was written in Fortran,
> >whether or not Apple had a commercial ForTran product in the box at that
> >point in time.
>
> I can assure you that it is not written in FORTRAN--in fact quite the
> opposite:  Apple FORTRAN is written in Pascal.
>
> Apple Pascal was bootstrapped through the ETH Zurich Pascal
> implementation, originally running on the CDC 6400.  When UCSD
> Pascal was completed, FORTRAN was implemented on top of the
> same P-code interpreter, and the FORTRAN compiler also runs in
> the Pascal P-code enviroment.
>
> FORTRAN implementations which I used permitted many additional
> files to be defined, though few FORTRAN programs ever used them.

Could the 3-stream limitation be an aspect of the UCSD p-code interpreter?
Those are the only two languages I know of that are based on the UCSD p-code
system that were implemented for the Apple II.

> >I stand by my previous statement that Apple Pascal was a half-baked
> >implementation.
>
> You are, of course, entitled to your opinion.  ;-)
>
> But another possibility is that it was fully-baked, but to a different
> specification than you would prefer.
>
> -michael

Fair enough; as long as I still get to call those undocumented features
"bugs." :-)

Seriously though, if the UCSD p-code interpreter defined a minimum number of
filestreams that could be open at once, then the similarity I attributed to
Fortran could just be a limitation of the p-code machine.

- Mike