Forgot your password?
Cyhist Jan 21 1997 B
Document Actions
Date: Tue, 21 Jan 1997 13:49:56 -0600
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
From: Peter da Silva <peter@baileynm.com>
Subject: Re: CYHIST: emulation and Microsoft history
In-Reply-To: <199701211924.NAA23715@web.nmti.com> from "Suzanne Johnson" at
Jan 20, 97 10:41:15 pm
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
I believe the following information is accurate, but it's been over 20 years now and I learned it second hand well after the events in question.
>When I started working at Intel in 1978, one of the most popular pieces of software they distributed (both to Universities and commercial organizations) was a suite of emulators which would allow a person to generate and compile code for microprocessors on other computers. Specifically, I recall the emulators for the 8080 which ran on the PDP-10.
Which were used by Gary Kildall to develop CP/M before the 8080 was really solidly out there, giving CP/M a head start in the 8080 world.
>systems. These were the "blue boxes" that were actually a self-contained system with text-editor and development software intended to simplify code development for microporcessors. (Actually, an early "personal computer").
We have several of these boxes here.
The operating system, ISIS, was clearly one of the models for CP/M. CP/M feels a lot like a cross between ISIS and a DEC operating system like RSX or TOPS. Which, given its background, isn't surprising.
One of the interesting things about ISIS is that the command line arguments were provided to the program as the first line of interactive input on the "terminal". I've often though of combining that with the UNIX concept of pipes and filters, and wondering what it would be like to deal with an O/S that was like UNIX but where a third default file descriptor was used for commands. First of all, of course, programs like "xargs" would be unneeded: you would simply pipe "find" into the "stdcmd" pipe. You could do all sorts of things with interactive filters that are currently impractical, where programs do things like reopening /dev/tty or reading from stderr...
I think it would be kinda nifty, ya know...
______________________________________________________________________
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
From: Peter da Silva <peter@baileynm.com>
Subject: Re: CYHIST: emulation and Microsoft history
In-Reply-To: <199701211924.NAA23715@web.nmti.com> from "Suzanne Johnson" at
Jan 20, 97 10:41:15 pm
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
I believe the following information is accurate, but it's been over 20 years now and I learned it second hand well after the events in question.
>When I started working at Intel in 1978, one of the most popular pieces of software they distributed (both to Universities and commercial organizations) was a suite of emulators which would allow a person to generate and compile code for microprocessors on other computers. Specifically, I recall the emulators for the 8080 which ran on the PDP-10.
Which were used by Gary Kildall to develop CP/M before the 8080 was really solidly out there, giving CP/M a head start in the 8080 world.
>systems. These were the "blue boxes" that were actually a self-contained system with text-editor and development software intended to simplify code development for microporcessors. (Actually, an early "personal computer").
We have several of these boxes here.
The operating system, ISIS, was clearly one of the models for CP/M. CP/M feels a lot like a cross between ISIS and a DEC operating system like RSX or TOPS. Which, given its background, isn't surprising.
One of the interesting things about ISIS is that the command line arguments were provided to the program as the first line of interactive input on the "terminal". I've often though of combining that with the UNIX concept of pipes and filters, and wondering what it would be like to deal with an O/S that was like UNIX but where a third default file descriptor was used for commands. First of all, of course, programs like "xargs" would be unneeded: you would simply pipe "find" into the "stdcmd" pipe. You could do all sorts of things with interactive filters that are currently impractical, where programs do things like reopening /dev/tty or reading from stderr...
I think it would be kinda nifty, ya know...
______________________________________________________________________
