etymology of the Unix "dd" command
Michael Meissner
(David J. MacKenzie) writes:
| Does anyone know how the Unix "dd" command got its name?
| I have a theory, based purely on reading the manual page, which in
| SunOS4 begins (I think it was similar in the V7 manual):
|
| NAME
| dd - convert and copy files with various data formats
|
| My guess is that the original author wanted to call it "cc", for
| "convert and copy", but since the name "cc" was already taken, chose
| the next letter in the alphabet and called it "dd". Does anyone have
| a more authoritative answer?
Dd comes from IBM JCL (it's used to create datasets for the job), and
in fact the argument syntax is reminsent of what little I remember of
the JCL syntax. Since one of the primary tasks that the original UNIX
was put to was remote job entry for IBM systems, I imagine some Bell
Labs type thought s/he was being clever.
--
Michael Meissner email: meis...@osf.org phone: 617-621-8861
Open Software Foundation, 11 Cambridge Center, Cambridge, MA
Catproof is an oxymoron, Childproof is nearly so
David J. MacKenzie
I have a theory, based purely on reading the manual page, which in
SunOS4 begins (I think it was similar in the V7 manual):
NAME
dd - convert and copy files with various data formats
My guess is that the original author wanted to call it "cc", for
"convert and copy", but since the name "cc" was already taken, chose
the next letter in the alphabet and called it "dd". Does anyone have
a more authoritative answer?
--
David J. MacKenzie <d...@eng.umd.edu> <d...@ai.mit.edu>
Henry Throop
I read (I think in UnixWorld, or some other magazine) that it stood
for DarneD if I know, but I don't know whether to believe this or
not. (No, I'm not making think up, btw.)
Henry
---
Henry Throop
Internet: thr...@jacobs.cs.orst.edu
jim frost
>Dd comes from IBM JCL (it's used to create datasets for the job), and
>in fact the argument syntax is reminsent of what little I remember of
>the JCL syntax. Since one of the primary tasks that the original UNIX
>was put to was remote job entry for IBM systems, I imagine some Bell
>Labs type thought s/he was being clever.
I got the impression that UNIX was used primarily for text processing
originally (that was the whole idea, wasn't it?) and not for IBM
remote job entry at all. Certainly there seems to be no leftover
cruft (either messages or /dev entries) to signify that it was ever
used for doing IBM job entry, unlike the CAT phototypesetter.
Perhaps I'm misunderstanding "original UNIX."
Happy hacking,
jim frost
saber software
ji...@saber.com
iccad
>
>"convert and copy", but since the name "cc" was already taken, chose
>the next letter in the alphabet and called it "dd". Does anyone have
>a more authoritative answer?
Not a more authoritative answer but my guess is simply:
"dd" = Dump data
--
==============================================================================
Terry Lalonde (613)592-2122 Revenge is a confession of pain. (Seneca)
Internet: mitel!lalonde@uunet Usenet: ...!uunet!mitel!lalonde
==============================================================================
Douglas W. Jones,201H MLH,3193350740,3193382879
by ma...@world.std.com (jim frost):
>
> meis...@osf.org (Michael Meissner) writes:
>
>>Dd comes from IBM JCL (it's used to create datasets for the job)...
>>... Since one of the primary tasks that the original UNIX
>>was put to was remote job entry for IBM systems, I imagine some Bell
>>Labs type thought s/he was being clever.
>
> I got the impression that UNIX was used primarily for text processing
> originally (that was the whole idea, wasn't it?) and not for IBM
> remote job entry at all. Certainly there seems to be no leftover
> cruft (either messages or /dev entries) to signify that it was ever
> used for doing IBM job entry, unlike the CAT phototypesetter.
>
Hill Computing Center and one floor down from the original UNIX group).
We knew of UNIX as a timesharing system project that was supported for
technical text processing applications, but we also knew that the best
games were on UNIX.
The UNIX editors and text fomatters (ed and troff) were almost fully
compatible with the editors and formatters on the MH-TSS system (the
Murray Hill Timesharing System, based on a H 6000 mainframe running GCOS,
where the editor was qed and the formatter was roff). The compatibility
wasn't surprising; my GCOS manuals list the authors of the editors and
formatters on MH-TSS as Ritchie, Thompson, Kernighan, and McIlroy, the
same people who wrote the corresponding UNIX software.
The later development, PWB-UNIX (the Programmer's Workbench) was indeed
heavily used for remote job entry on IBM mainframes, and was loaded with
RJE cruft, and some RJE cruft has survived from Murray Hill all the way
to the version of Berkeley UNIX I'm using right now. Consider the
following quote from the BUGS line in my man page for the finger command:
The encoding of the gcos field is UCB dependent - it knows
^^^^
The name of the gcos field in /etc/passwd file only makes sense if you
know that there were rje links from the original UNIX system to the
MH-TSS system, running the GCOS operating system, at Murray Hill. You
were supposed to put your GCOS user id in the gcos field!
Since GCOS wasn't an IBM operating system, this doesn't answer the
question about the dd command. That doesn't really matter. It is
so obviously patterned after the dd statement in IBM's JCL that to
ascribe any other origin to it would be silly. Just about every
programmer in the world in 1973 had at least a passing familiarity
with IBM JCL, most of us had spent painful hours trying to get our
DD statements right at one time or another, and the parameter syntax
of the UNIX dd command is clearly a parrody of the IBM syntax.
Doug Jones
jo...@herky.cs.uiowa.edu
Michael Meissner
(jim frost) writes:
Well there is the GECOS field in the password file.
Ken Konecki
>Does anyone know how the Unix "dd" command got its name?
I always thought it came from Device to Device copy.
Cheers,
-Ken K
--
Ken Konecki
"Eat well, stay fit, die anyway"
e-mail:ke...@tellabs.com -or- ...!uunet!tellab5!kenk
U.S. Mail: 1271 Portchester Circle, Carol Stream, IL 60188
The Grey Wolf
>>Does anyone know how the Unix "dd" command got its name?
I always thought it came from its nickname -- we used to call it "data
dump" waybackwhen.
--
Brain fried. (Explanation is in file "core".)
Chip Christian
>originally (that was the whole idea, wasn't it?) and not for IBM
>remote job entry at all. Certainly there seems to be no leftover
>cruft (either messages or /dev entries) to signify that it was ever
>used for doing IBM job entry, unlike the CAT phototypesetter.
>
>Perhaps I'm misunderstanding "original UNIX."
>
play Space War.
--
-Chip
Chip Christian
AT&T Bell Laboratories
Murray Hill, NJ
ch...@allegra.research.att.com
"We've got a kinder, gentler machine gun hand..."
AT&T_opinions = ! my_opinions;
Jeremy J. Epstein
[Much stuff deleted]
> [...] Consider the
> following quote from the BUGS line in my man page for the finger command:
>
> The encoding of the gcos field is UCB dependent - it knows
There are other gcos-isms left...the portable C compiler (upon which many
commercial compilers are based) will respond to the acute accent (`)
with the error:
"foo.c", line 1: gcos BCD constant illegal
--
Jeremy Epstein
TRW Systems Division
703-876-4202
j...@virtech.uu.net
Steve Lamont
> //sysin dd lrecl=80,blksiz=800,disp=(new,new,save),
> space=(20,20,rlse)
//SYSIN DD DCB=(LRECL-80,BLKSIZE=800,RECFM=FB),
// DISP=(OLD,KEEP),SPACE=(20,20,,RLSE),
// DSN=C1882P.FOO.BAR.BAZ.WOOF
>(hmm, doesn't look quite right, who cares)
spl (the p stands for
//PUNCH DD SYSOUT=B)
--
Steve Lamont, sciViGuy (919) 248-1120 EMail: s...@ncsc.org
NCSC (The other one), Box 12732, Research Triangle Park, NC 27709
Don't send in no bums. I want deals.
-John Steinbeck, _The Grapes of Wrath_
Barry Shein
>Does anyone know how the Unix "dd" command got its name?
It's a play on the IBM/JCL DD command, DD stood for DATA DEFINITION in
JCL and was how you attached files to a job. All files had to be
pre-allocated and were defined externally to the job, sort of like the
Unix < and > commands only absurdly complicated.
The cards looked like:
//sysin dd *
which meant attach the symbolic input name (sysin) to the rest of this
card deck (*), but more telling are things like:
//sysin dd lrecl=80,blksiz=800,disp=(new,new,save),
space=(20,20,rlse)
(hmm, doesn't look quite right, who cares)
starting to look familiar? There were around 200 options available,
and the defaults were almost always wrong...
The job accessed the symbol SYSIN (or whatever, you just made it up,
but a lot of packages used SYSIN, SYSOUT, SYSPRINT and SYSPUNCH as a
standard) as if it were a file and the file was attached externally
through a DSNAME (data set name) card who's format was described in
the DD card.
Let's put it this way, it took a reasonably seasoned programmer to
just make a copy of a file...
The fun part was that JCL errors where I worked years ago cost about
$1.50 each due to minimum costs of running one job through and having
it bomb out. I went through over $1,000 one evening, in real money,
trying to get one lousy JCL bug out which turned out to be a change
someone in systems had made in a standard JCL utility program...
The point being, DD cards were loathsome things so making a Unix
command with that name was black humor.
--
-Barry Shein
Software Tool & Die | {xylogics,uunet}!world!bzs | b...@world.std.com
Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD
David Boyes
>In article <1990Apr22.1...@world.std.com> b...@world.std.com (Barry Shein) writes:
>> //sysin dd lrecl=80,blksiz=800,disp=(new,new,save),
>> space=(20,20,rlse)
>
> //SYSIN DD DCB=(LRECL-80,BLKSIZE=800,RECFM=FB),
> // DISP=(OLD,KEEP),SPACE=(20,20,,RLSE),
> // DSN=C1882P.FOO.BAR.BAZ.WOOF
>
>>(hmm, doesn't look quite right, who cares)
Nope, still wouldn't work. You left out device type and volser,
unless you're assuming it was a cataloged dataset in either the
JOBLIB or STEPLIB cards.
//SYSIN DD DSN=C1882P.FOO.BAR.BAZ.WOOF,
// UNIT=DISK,
// VOL=SER=VSPK01,
// DCB=(LRECL=80,RECFM=FB,BLKSIZE=800),
// DISP=(OLD,KEEP,KEEP),
// SPACE=(20,20,5,RLSE)
>Steve Lamont, sciViGuy (919) 248-1120 EMail: s...@ncsc.org
Anybody else remember *why* it was a good idea to put all your JCL
parms on separate lines?
--
David Boyes | "Where's the ka-boom? There's supposed to be an
dbo...@rice.edu | Earth-shattering ka-boom!...Heavens! Someone has
| stolen the Illudium Q-38 Explosive Space Modulator!
"Delays, delays!" | The Earth creature has *stolen* the Space Modulator!"
The Polymath
}//SYSIN DD DSN=C1882P.FOO.BAR.BAZ.WOOF,
}// UNIT=DISK,
}// VOL=SER=VSPK01,
}// DCB=(LRECL=80,RECFM=FB,BLKSIZE=800),
}// DISP=(OLD,KEEP,KEEP),
}// SPACE=(20,20,5,RLSE)
}Anybody else remember *why* it was a good idea to put all your JCL
}parms on separate lines?
Because they were punched on cards, of course. Putting them on separate
lines let you change one without retyping several others.
I apprenticed in a shop that had a huge file full of standard cards. I
got yelled at once because someone had slipped in a job card with a space
between the final parameters, causing misc. grief. Fortunately, I was
able to go to the appropriate card file and prove that _all_ the cards had
that space, so it wasn't entirely my fault.
Does that make me an "old timer?" It was only 1981 ...
--
The Polymath (aka: Jerry Hollombe, M.A., CDP, aka: holl...@ttidca.tti.com)
Citicorp(+)TTI Illegitimis non
3100 Ocean Park Blvd. (213) 450-9111, x2483 Carborundum
Santa Monica, CA 90405 {csun | philabs | psivax}!ttidca!hollombe
d.m.drangula
>
>Anybody else remember *why* it was a good idea to put all your JCL
>parms on separate lines?
>
I used to do it because it was easier to edit when the duplicator
wasn't working on the keypunch. Anybody else remember changing a
ribbon on a model 34(?). Funny thing, in 1976 or so the only place
I could even order the ribbon from was Big Blue.
Personally, as soon as I figured a way, I got a Wylbur account
at Rutgers and stayed as far away from keypunch as possible.
D. Michael Drangula ( #include standard.disclaimer )
Card decks are for ANSYS and NASTRAN weenies ;-)
Eric Hvozda
>Personally, as soon as I figured a way, I got a Wylbur account
Just out of question, how many peoples sites are *still* using WYLBUR?
I recently Co-oped at Air Products and Chemicals, and they were, but they said
they were in the process out phasing it out. My supervisor said that the
higher-ups have been saying that for years and doubts it will happen in the
near future. :)
-=-=-
Ack!
Jeremy Brest
>Just out of question, how many peoples sites are *still* using WYLBUR?
Forsythe.stanford.edu, a machine that many people in the
administration use, uses WYLBUR. No plans to phase it out, far as I
know.
.. .. .. .. .. .. .. .. .. .. .. .. .. ..
Jeremy Brest, Swarthmore College, Swarthmore, PA 19081
jer...@cs.swarthmore.edu, jer...@swatsun.uucp, jer...@swarthmr.bitnet
The Grey Wolf
# starting to look familiar? There were around 200 options available,
# and the defaults were almost always wrong...
Isn't it *usually* that way? At least one person in any given organisation
will insist that the defaults are wrong. Let's hear it for environment
or file-based defaults...
#
# Let's put it this way, it took a reasonably seasoned programmer to
# just make a copy of a file...
#
Maybe if we re-did the "rm" command like that, people wouldn't have
so many problems. :-)
# The fun part was that JCL errors where I worked years ago cost about
# $1.50 each due to minimum costs of running one job through and having
# it bomb out. I went through over $1,000 one evening, in real money,
# trying to get one lousy JCL bug out which turned out to be a change
# someone in systems had made in a standard JCL utility program...
#
# The point being, DD cards were loathsome things so making a Unix
# command with that name was black humor.
That may be Unix' idea of black humour, but Unix has presented us with
its own nightmare. It's called find(1).
#
# --
# -Barry Shein
#
# Software Tool & Die | {xylogics,uunet}!world!bzs | b...@world.std.com
# Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD
"
Roan Anderson, The System Administrator
UniSoft Corporation
...!{uunet,sun,amdcad,acad,ucbvax,mtxinu}!unisoft!{greywolf,root,postmaster}
Doug Landauer
Someone at UCLA actually wrote a replacement for /bin/sh (for a version
7 Unix, I think) which interpreted JCL. For instance, you had to type
//SYSIN DD *
in order to get a program's stdin to be the terminal.
He (or a co-conspirator) also had written an editor to replace "ed",
called "029". As you typed, the "card" scrolled to the left (just like
a card does in the real 029) and the proper holes appeared in the
appropriate column . I think he had even gotten a DUP function to
work before giving up on the project.
--
Doug Landauer -- land...@eng.sun.com or ...!sun!landauer _
Sun Microsystems, Inc. -- SPD, SET, Languages La no ka 'oi.
Per Lindberg
>Someone at UCLA...
>... had written an editor to replace "ed",
>called "029". As you typed, the "card" scrolled to the left (just like
>a card does in the real 029) and the proper holes appeared in the
This goes back even further. The first (?) graphic Hollerith editor
was written on the LMI Lisp Machine. I saw it on their CADR model,
beautiful!
Talking about graphic hax, has anyone written a Munching Squares
program in C?
--
Per Lindberg (The Mad Programmer) ! __!__
Front Capital Systems ! _____(_)_____ Ceci n'est pas une Piper
Linneg 5, 11447 Stockholm, Sweden ! ! ! !
jim frost
># The point being, DD cards were loathsome things so making a Unix
># command with that name was black humor.
>That may be Unix' idea of black humour, but Unix has presented us with
>its own nightmare. It's called find(1).
But ... you couldn't get around those "dd" statements. There are any
number of ways to avoid using "find" if you are particularly annoyed
with its syntax.
Robert Stanley
Here is the header from a posting a year ago in comp.sources misc:
=> From: all...@uunet.UU.NET (Brandon S. Allbery - comp.sources.misc)
=> Newsgroups: comp.sources.misc
=> Subject: v06i036: JCL emulator
=> Message-ID: <48...@uunet.UU.NET>
=> Date: 4 Feb 89 03:18:19 GMT
=> Reply-To: lup...@uhccux.uhcc.Hawaii.Edu (Robert Lupton)
=> Approved: all...@uunet.UU.NET (Brandon S. Allbery - comp.sources.misc)
=>
=> Posting-number: Volume 6, Issue 36
=> Submitted-by: lup...@uhccux.uhcc.Hawaii.Edu (Robert Lupton)
=> Archive-name: jcl
=>
=> [You always knew someone would do this...! ;-) ++bsa]
=>
=> I wrote this a while ago, and have just found it on an old tape, so
=> here it is. It emulates our favourite operating system, JCL. To use
=> it, unpack the shar file, run make, then "JCL < deck" as a demo.
=>
=> As it says in the README, I bequeath this code to the net. In particular
=> I don't intend to handle any bugfixes/improvements.
=>
=> Robert Lupton
This thread tickled me into digging it out from my archive. :-)
And here, to reinforce the point, is what must be one of the most
baroque "Hello World" implementations going:
//NAME JOB ROBERT.H.LUPTON,MSGLEVEL=(2,2), comments 78901
// MSGCLASS=A
//COMP EXEC PGM=CC,PARM='-g'
//SYSOUT DD DSN=TEMP,DISP=KEEP
//SYSIN DD *
#include <stdio.h>
main()
{
printf("Hello World\n");
}
/*
//LKED EXEC PGM=LKED,PARM='-lc'
//SYSIN DD DSN=TEMP,DISP=(,KEEP,DELETE)
//SYSOUT DD DSN=TST,DISP=(,KEEP,DELETE)
//GO EXEC PGM=TST
//
Now, if only we can find someone insane enough to co-implement this
with Intercal, we'll be all set for the Summer.
Robert_S
--
Robert Stanley UUCP: uunet!mitel!sce!cognos!roberts
Cognos, Inc. INET: roberts%cogno...@uunet.uu.net
(Research) Voice: (613) 738-1338 x6115
Warner Losh
>Now, if only we can find someone insane enough to co-implement this
>with Intercal, we'll be all set for the Summer.
That is the most evil thought I've seen in a long time. Almost as
evil as reimplementing TECO in TECO (without TECO commments). :-)
--
Warner Losh i...@Solbourne.COM
Eric S. Raymond
> Now, if only we can find someone insane enough to co-implement this
> with Intercal, we'll be all set for the Summer.
Um. I've implemented an INTERCAL compiler in C (except for the butchered-Roman
-numeral output and the storage manager -- but it generates clean and apparently
correct optimized C code). Now what?
Forward into the past! New frontiers in retrocomputing await!
--
Eric S. Raymond = ...!uunet!snark!eric (mad mastermind of TMN-Netnews)
Kenneth Arromdee
>Um. I've implemented an INTERCAL compiler in C (except for the butchered-Roman
>-numeral output and the storage manager -- but it generates clean and apparently
>correct optimized C code). Now what?
Hmm. Maybe if there's enough interest you could post it to alt.sources?
Better yet, forget alt.sources. Why not a comp.lang.intercal?
--
"And they shall be cast out where there is no outlet for their evil doings..."
-- the Book of Ubizmo, on sinful uses of electricity
Kenneth Arromdee (UUCP: ....!jhunix!arromdee; BITNET: arromdee@jhuvm;
INTERNET: arro...@crabcake.cs.jhu.edu)
Matt Squires
missed the description of Intercal. Could some kind soul on the net
please send me the original article, a description of the (joke-like?)
language, etc? In another words, let me in on the joke?
Also, if you too want the information, please don't clutter the net with
"me too" articles, just send me a request and I'll send you the
intercal documents too.
Matt Squires
squ...@eecs.nwu.edu
...{oddjob|gargoyle|chinet}!nuscrl!squires
Tom Ivar Helbekkmo
>-numeral output and the storage manager -- but it generates clean and apparently
>correct optimized C code). Now what?
Post it! Post it! Post it! :-) And I really think we ought to
issue a "call for discussion" on comp.lang.intercal, as I don't feel
that further work and discussion on Intercal really belongs in this
group.
How about it? Is there enough interest out there that we go ahead?
-tih
--
Tom Ivar Helbekkmo, NHH, Bergen, Norway. Telephone: +47-5-959205
t...@barsoom.nhh.no, thel...@norunit.bitnet, edb...@pluss.nhh.no
Ed McGuire
> And I really think we ought to
> issue a "call for discussion" on comp.lang.intercal, as I don't feel
> that further work and discussion on Intercal really belongs in this
> group.
>
> How about it? Is there enough interest out there that we go ahead?
I suggest that a mailing list would be sufficient.
---
peace. -- Ed
"So, when `thirtysomething' was over, she turned off the TV, collected
her thoughts and called 350,000 of her closest friends."
Ric Werme
>// MSGCLASS=A
>//COMP EXEC PGM=CC,PARM='-g'
>//SYSOUT DD DSN=TEMP,DISP=KEEP
>//SYSIN DD *
>#include <stdio.h>
>main()
>{
> printf("Hello World\n");
>}
>/*
Err, um, well, it's been 20 years (but they've been good years) since I punched
my last JCL card, but I distinctly recall no /usr/include directory on OS/360.
Seems to me that had there been a C compiler, you would also need:
//INCLUDE DD <I-can't-remember-any-of-this-nonsense!>
PARM=-g? Well, I used batch editors. I suppose I could use batch dbx. But
I'd rather not!
--
| A pride of lions | Eric J Werme |
| A gaggle of geese | uucp: decvax!linus!alliant |
| An odd lot of programmers | Phone: 603-673-3993 |
Peter Deutsch
> In article <83...@cognos.UUCP> rob...@cognos.UUCP (Robert Stanley) writes:
[ *** Hello world JCL deleted...*** ]
> PARM=-g? Well, I used batch editors. I suppose I could use batch dbx. But
> I'd rather not!
Ah, batch debugging. There's a phrase to throw around a
pint or two. I teach the Principles of Assembly Languages
course here at McGill and when I took over the course they
were using a PDP-11 simulator that ran under MUSIC (an O/S
that runs under VM on IBM mainframes). Imagine the
functionality of most mainframe operating systems. Strip
off most of it. That's MUSIC! :-)
(And yes, MUSIC was written here at McGill. And yes,it is
still used on campus, but not by me!)
Anyways, the way a student wrote an assembler program was
to type in the source file, which would include Assembler
directives, Simulator directives and MUSIC JCL, which was
translated to real JCL by MUSIC before it sent the job off
to be executed. This whole kit and kaboodle was then sent
off to be executed as a student priority batch job, with
the output automatically routed to a line printer.
The assembler tried to assemble the program into simulated
PDP-11 memory, then if it got that far it started the
simulator on this until it reached a HALT (or more likely,
crashed).
This output was an assembly listing, a dump of registers
when the program halted, along with an optional memory
dump. The student then picked over the resulting memory
dump to try and decide why the entire program had been
reduced to a series of 000000, 000000, 000000, etc. It was
sort of like examining the entrails of a chicken to
determine the future.
It was only the real hotshots who learned how to use
"OSJR" to get the result back into a file or (God forbid)
displayed onto the screen. Of course, if the output was
dumped to the screen, there was no way to save it or
scroll backwards and see it again.
Oh, yeah. I almost forgot. How do we simulate console I/O?
Well, output was simply tacked on to the end of the
listing, input was read as 80 column input from the
characters between the final .END and $END (ie between the
assembler directive and the simulator directive). The
simulator took care of translating this from EBCDIC and
ASCII, but couldn't include carriage returns so you had to
use something like a '*' to mark end of line. Reading past
the last character of the last input card was a fatal
error, with a mysterious error number associated with it.
We've replaced this monster with a UNIX-based simulator
written by a student for a project course. breakpoints,
single stepping and more. Although not perfect, it beats
the pants off the previous mainframe programming
environment.
The really sad thing was that this monster was in use up
until 1987! Ah, tempus fugit.
- peterd
Michael S. Pereckas
>I suggest that a mailing list would be sufficient.
Probably more than sufficient...
--
Michael Pereckas * InterNet: m-per...@uiuc.edu *
NovaNet: m pereckas / imsa89 / cerl
+ My opinions are mine (no one else wants them, that's for sure). +
+ I will accept no criticism of my spelling of any words not in the OED2 +
Barry Shein
>(And yes, MUSIC was written here at McGill. And yes,it is
>still used on campus, but not by me!)
Wasn't MUSIC a rework of RAX?
Maybe we should have a special thread for all those one-shot O/S's
people wrote in dreams of fame and glory. Many of them run to this
day, usually wherever the original authors remain in control.
--
-Barry Shein
Software Tool & Die | {xylogics,uunet}!world!bzs | b...@world.std.com
Nik Conwell
>
>>(And yes, MUSIC was written here at McGill. And yes,it is
>>still used on campus, but not by me!)
>
>Wasn't MUSIC a rework of RAX?
>
The main O.S. here at B.U. (on the 3090) is called VPS. Many modules refer
to RAX, and I've heard that it was modeled on (from?) RAX. Anybody have any
interesting info on RAX, or can point me somewhere to it?
-nik n...@bu-it.bu.edu
Peter Deutsch
>
> >(And yes, MUSIC was written here at McGill. And yes,it is
> >still used on campus, but not by me!)
>
> Wasn't MUSIC a rework of RAX?
It began life as a rework of something from IBM, I can't
remember what. If someone in our Computing Centre ever
get's news access, reads this and calls to scream at me
I'll ask them (no really, we have a pretty good
relationship. We do. NO! NOT THE CATTLE PROD! ARRGGGHHHH!)
There, I feel better now. MUSIC is a wonderful O/S.
twitch.
twitch.
> Maybe we should have a special thread for all those one-shot O/S's
> people wrote in dreams of fame and glory. Many of them run to this
> day, usually wherever the original authors remain in control.
Actually, MUSIC has been renamed from the McGill
University System for Interactive Computing (MUSIC) to the
Multi-User System for Interactive Computing (MUSIC), or
something like that and is sold by IBM. I understand it
runs at hundreds of sites. The MUSIC Product Group
apparently runs at a profit!
I must say (since I _never_ put a standard disclaimer in
my postings and really do like my job) that MUSIC is
probably no worse than any other mainframe O/S designed to
do the same things. It is supposed to be for novices and
students and the way you stop novices from destroying
things is probably to stop them from doing things. With
UNIX we can set it up this way and let the novices remove
protection when they're ready. This goes against the grain
of mainframe philosophy (WHAT! Let users change
things?!!?) but that's not MUSIC's fault. Anyways, I hate
mainframe O/S's, so there.
- peterd
Simon E Spero
emcg...@cadfx.ccad.uiowa.edu (Ed McGuire) writes:
>I suggest that a mailing list would be sufficient.
Probably more than sufficient...
But not nearly silly enough....
--
------------------------------------------------------------------------------
zma...@uk.ac.ic.doc | sispero%c...@specialix.co.uk | ..!ukc!slxsys!cix!sispero
------------------------------------------------------------------------------
The Poll Tax. I'm Not. Are you?| Snail: DoC, 180 Queensgate, London SW7 2BZ
"The GNU Manifesto refers to all Software, not just Editors" | (I'm the FSF)
Edward Vielmetti
Maybe we should have a special thread for all those one-shot O/S's
people wrote in dreams of fame and glory. Many of them run to this
day, usually wherever the original authors remain in control.
You mean, like the Michigan Terminal System?
--Ed
Edward Vielmetti, U of Michigan math dept.
e...@math.lsa.umich.edu
jim frost
>Maybe we should have a special thread for all those one-shot O/S's
>people wrote in dreams of fame and glory. Many of them run to this
>day, usually wherever the original authors remain in control.
How about NATS (Naval Academy Time Sharing System)? Is that still
around?
Patrick B. Haggood
MTS can't qualify, it's quite popular, being used on at least 25
machines on the planet. Doesn't that make it a best seller?
On another note, how about OS's with lots of users/hackers, but
no software? For Minix, there's about a new version every two
weeks....
--
Patrick B. Haggood
Wayne STate University
Detroit, MI
Physics - Class of 1991 (-2?)