On the Naming of UNIX Things
Mark Whetzel
I was having a discussion with some of my co-workers on WHY things
are named as they are in UNIX. Anybody have some of the
tales behind the odd things that bound in unix history?
One question we had of some students we are training:
Why are system processes called 'daemons'?
Some of the others I have heard about: the naming of 'biff', grep, yacc.
E-mail preferred, as I can then post a summary of the BEST of this quest.
References to printed material and other net resources appreciated.
--
Mark Whetzel My comments are my own, not my company's.
Western Geophysical - A division of Western Atlas International Inc.,
A Subsidiary of Western Atlas Inc. DOMAIN addr: mark.w...@waii.com
VOICE: (713) 963-2544 FAX: (713) 963-2758
Pete Barber
On 8 Nov 1996 14:57:51 GMT, Mark Whetzel <ma...@zeus.wg.waii.com>
wrote:
>I was having a discussion with some of my co-workers on WHY things
>are named as they are in UNIX. Anybody have some of the
>tales behind the odd things that bound in unix history?
>
>One question we had of some students we are training:
>
> Why are system processes called 'daemons'?
>
Daemons are from Greek Mythology and our 'little helpers' that
transport between 'Man' and 'The Gods'. I guess these chaps wern't
powerful enough be gods (i.e. kernel) and slightly more powerful than
'Man' (user process)
>Some of the others I have heard about: the naming of 'biff', grep, yacc.
biff - dunno
grep GetREgularePression
yacc - yet another compiler compiler (during the time of it's creation
lots of ppl were writing compiler (well parser) generators
Tony Finch
pete....@openv.co.uk (Pete Barber) wrote:
> Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
>
> >Some of the others I have heard about: the naming of 'biff', grep, yacc.
>
> biff - dunno
Supposedly after a dog that would bark at the postman.
> grep GetREgularePression
It's from ed syntax: g/re/p globally searches for the regular
expression re and prints lines that match.
Tony.
Mark Whetzel
Following up to my own post...
In comp.unix.misc Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
: I was having a discussion with some of my co-workers on WHY things
: are named as they are in UNIX. Anybody have some of the
: tales behind the odd things that bound in unix history?
: One question we had of some students we are training:
: Why are system processes called 'daemons'?
I am getting some GOOD feedback on several possible reasons...
: Some of the others I have heard about: the naming of 'biff', grep, yacc.
BY this sentence... what I was looking for was OTHER types of
'funny' unix commands/programs/procedures and their naming.. I know
about biff, grep and yacc...
Sort of like to make a collection of the 'in-jokes' and other strange
reasons things are like they are in unix..
E-mail prefered... I'll summarize the best 'funnies' that I get, and
that get the most common and/or strange reasons.
Thanks
--
Mark Whetzel My comments are my own, not my company's.
Western Geophysical - A division of Western Atlas International Inc.,
A Subsidary of Western Atlas Inc. DOMAIN addr: mark.w...@waii.com
Benjamin Hutchings
In article <328386bc....@news.ov.com>,
Pete Barber <pete....@openv.co.uk> wrote:
[...]
>On 8 Nov 1996 14:57:51 GMT, Mark Whetzel <ma...@zeus.wg.waii.com>
>wrote:
>
>>I was having a discussion with some of my co-workers on WHY things
>>are named as they are in UNIX. Anybody have some of the
>>tales behind the odd things that bound in unix history?
>>Some of the others I have heard about: the naming of 'biff', grep, yacc.
I heard that Biff was the name of Ken Thompson's dog, which would bark
loudly whenever the mailman came to deliver, but this could be another UL.
--
Ben Hutchings,|finger m95...@booth42.ecs.ox.ac.uk|mail benjamin.hutchings@
compsci&mathmo|lynx http://users.ox.ac.uk/~worc0223|worcester.oxford.ac.uk
Some mail sent on Saturday may have been lost, please resend if in doubt.
Asking whether machines can think is like asking whether submarines can swim.
John Rudd
In <560146$t...@mail1.wg.waii.com> Mark Whetzel wrote:
> : Some of the others I have heard about: the naming of 'biff', grep, yacc.
>
> BY this sentence... what I was looking for was OTHER types of
> 'funny' unix commands/programs/procedures and their naming.. I know
> about biff, grep and yacc...
>
You mean things like "awk" is the last initials of the 3 authors of the
program/language
(Aho, Weinberger, Kernighan)?
--
John "kzin" Rudd jr...@cygnus.com (ex- kz...@email.sjsu.edu)
=========Intel: Putting the backward in backward compatible.============
Spammers: I charge you for my time, disk, and bandwidth if you post off-
topic solicitations for money in the groups I read. $500/post/group.
Peter da Silva
Sorry, it's based on the ed(1) command that performs the same function:
g global
If you want to print all lines containing a regular expression (re), you
would type:
g/re/p
Where "re" is the placeholder for the regular expression. In version 6 unix
there was a second command, gres, based on the ed(1) global substitute command:
g/re/s//replacement/
But gres was replaced by sed in version 7. Grep was kept around because
it's a comman action and it's much faster than the more powerful Stream
EDitor.
--
</peter>
Melinda Shore
In article <328386bc....@news.ov.com>,
Pete Barber <pete....@openv.co.uk> wrote:
>grep GetREgularePression
Way incorrect. "Get Regular, Eat Prunes."
If you aren't sure about something, look it up before posting.
--
Melinda Shore - No Mountain Software - sh...@nomt.com
If you send me harassing email, I'll probably post it
Mark Greene
Mark Whetzel wrote:
>
> I was having a discussion with some of my co-workers on WHY things
> are named as they are in UNIX. Anybody have some of the
> tales behind the odd things that bound in unix history?
>
>
> Why are system processes called 'daemons'?
>
>
>
> References to printed material and other net resources appreciated.
>
> Mark Whetzel My comments are my own, not my company's.
> Western Geophysical - A division of Western Atlas International Inc.,
Check out the UNIX FAQ posted every month or so. Section 5 or 6, I think,
has a bit on unix command name history. Also, check out the book "A
Student's Guide to UNIX" by Harley Hahn. Very informative, witty, and
readable, with lots of historical tidbits. Including a couple of
paragraphs on deamons.
MJG
Robert Schuldenfrei
Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
>I was having a discussion with some of my co-workers on WHY things
>are named as they are in UNIX. Anybody have some of the
>tales behind the odd things that bound in unix history?
>One question we had of some students we are training:
> Why are system processes called 'daemons'?
>Some of the others I have heard about: the naming of 'biff', grep, yacc.
Yet Another Compiler Compiler (yacc)
According to folklore (that is the name of this NG is it not) Ken Thompson had
only a model 33 tty on his original system. Kept everything short! Do you know
what instuments are on Ken's car's dashboard? One light that illuminates "?" :)
Bob
Robert Schuldenfrei
S. I. Inc.
32 Ridley Road
Dedham, MA 02026
Voice: (617) 329-4828
FAX: (617) 329-1696
E-Mail: b...@s-i-inc.com
WWW: http://www.tiac.net/users/tangaroa/index.html
Gene Wirchenko
d...@world.std.com (Jeff DelPapa) wrote:
>In article <560146$t...@mail1.wg.waii.com>,
>Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
>>Following up to my own post...
>>
>>In comp.unix.misc Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
>>: I was having a discussion with some of my co-workers on WHY things
>>: are named as they are in UNIX. Anybody have some of the
>>: tales behind the odd things that bound in unix history?
>>
>Biff was someones dog that barked when the mailman came.
>The biggest conclusion that you can draw from the oldest of unix
>commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
^
Look! There's one now! Don't let it get away!
rm -r /usr
Oh, too late!
>original authors were vwl alrgc.
And Unix is usually written in C. Aha, a conspiracy.
><dp>
>(or possibly native Check speakers who had vowels, but never wrote
>them down. ex: Plzn, a town famous for its beer. If I remeber
>correctly, Dave Barry speculates that all the excess vowels were sold
>to the Hawiians)
Sincerely,
Gene Wirchenko
C Pronunciation Guide:
y=x++; "wye equals ex plus plus semicolon"
x=x++; "ex equals ex doublecross semicolon"
Matti Saarinen
worc...@sable.ox.ac.uk (Benjamin Hutchings) writes:
> >biff - dunno
>
> I heard that Biff was the name of Ken Thompson's dog, which would bark
> loudly whenever the mailman came to deliver, but this could be another UL.
Quote from A Quarter Century of UNIX ,an excellent book by Peter H.
Salus: (pp 169 - 170)
John Foderero, a graduate student at Berkeley in the summer of 1980,
wrote a progaram that checked whether new mail had arrived. It would
tell a user You have new mail. At the time, Heidi Stettner woked in
Evans Hall, prior to beginning graduate study. Heidi would bring her
dog with her to class and to her office. He was a very friendly dog,
and a lot of students enjoyed throwing a ball down the corridor for
him to fetch. He even had his picture on the bulletin board with the
graduate students: the legend read that he was working on his Ph.Dog.
John decided to name the program after the dog: Biff. According to
Heidi, John and Billy Joy then spent a lot of time trying to
compose an explanation to biff - they came up with "Be notified if
mail arrives." Biff, who died in August 1993, at 15, once got a B
in compiler class. According to Heidi, the story of Biff barking at
the mailman is a scurrilous canard.
--
- Matti -
Rick Hawkins
>The biggest conclusion that you can draw from the oldest of unix
>commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
>original authors were vwl alrgc.
Hey, why type 5 characters when 2 will do? It saves some memory, but a
lot of typing . . . the people that were using it were also writing it.
Almost all of my aliases (aliae?) and scripts are one or two letters.
If i don't use it often enough to remember whtat it is, it's time for it
to go . . .
>(or possibly native Check speakers who had vowels, but never wrote
>them down. ex: Plzn, a town famous for its beer.
They're kind of pathological about those vowels. Irish (Guiness, et al)
started using unmalted roast barley a couple of centuries ago because it
was the malt, rather than the beer that was taxed; maybe the czechs
taxed vowels :)
When i listed a czech friend in the acknowledgements of the paper
several years ago, it didn't occur to me that the czech & german
spellings would be different. And then i botched the German. So
instead of Kutlvasr (4 syllables; 2 vowels!), i came up with something
like cuddelwasser, which was not correct german, either: instead of
"cool water" it came out "cuddle water". *sigh* And this was because
she helped with the editing . . .
--
R E HAWKINS
rhaw...@iastate.edu
These opinions will not be those of ISU until they pay my retainer.
John Everett
In article <328386bc....@news.ov.com>, pete....@openv.co.uk says...
>
>On 8 Nov 1996 14:57:51 GMT, Mark Whetzel <ma...@zeus.wg.waii.com>
>wrote:
>
>>I was having a discussion with some of my co-workers on WHY things
>>are named as they are in UNIX. Anybody have some of the
>>tales behind the odd things that bound in unix history?
>>
>>
>> Why are system processes called 'daemons'?
>>
>transport between 'Man' and 'The Gods'. I guess these chaps wern't
>powerful enough be gods (i.e. kernel) and slightly more powerful than
>'Man' (user process)
>
It should be pointed out that when UNIX was still a gleam in Ken Thompson's
eye the PDP-10 (later to be called DECsystem-10) monitor (later to be called
TOPS-10) was running a system program called DAEMON (pronounced deemon, NOT
daymon). DAEMON was considered to be a swappable part of the monitor, it
performed system functions which were required infrequently enough that it
wasn't considered prudent to permanently dedicate memory space. How do I
know? In the late '60s and early '70s I was a member of the PDP-10 monitor
group and wrote parts of DAEMON.
The daemon concept may predate its implementation on the PDP-10, I'm not
sure. It's entirely possible it came from either Project MAC or Multics. I
know that Tom Hastings and Dennis Ritchie both worked on Multics. Tom
rejoined the PDP-10 monitor group after his stint at Multics, and of course
Dennis joined Ken at Bell Labs and had a bit to do with the development of
UNIX. BTW, the name UNIX is a play on Multics.
--
jeve...@wwa.com (John V. Everett)
Ken Bibb
>Biff was someones dog that barked when the mailman came.
Heidi Stettner's dog. (See Peter Salus's "A Quarter Century of Unix" for
more details.)
>The biggest conclusion that you can draw from the oldest of unix
>commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
>original authors were vwl alrgc.
Most likely due to the amount of pressure required to depress the
keys on a Decwriter...
--
Ken Bibb "If the boundary breaks I'm no longer alone
kb...@arastar.com Don't discourage me
kb...@best.com Bring out the stars/On the first day"
kb...@csd.sgi.com David Sylvian--"The First Day"
Will Rose
Tony Finch (fa...@cam.ac.uk) wrote:: >
: > >Some of the others I have heard about: the naming of 'biff', grep, yacc.
: >
: > biff - dunno
: Supposedly after a dog that would bark at the postman.
After a dog, but not one that barked at the postman. According
to the dogs owner this was a foul canard, or dirty duck.
Ralph Wade Phillips
Howdy!
In article <328631...@jpmorgan.com>, Ling Wang
<wang...@jpmorgan.com> wrote:
> Melinda Shore wrote:
> >
> > In article <328386bc....@news.ov.com>,
> > Pete Barber <pete....@openv.co.uk> wrote:
> > >grep GetREgularePression
> >
> > Way incorrect. "Get Regular, Eat Prunes."
> >
> > If you aren't sure about something, look it up before posting.
>
> Shouldn't it be "Go Regularly, Eat Prunes"?
The "Olympus TuneUp" help file lists GREP as "Global Regular
Expression Parser".
Dunno how accurate, but it both fits the acronmym !AND! describes
what GREP does ... <B-)
(For the $64K question - The Apple Lisa had both a REAL and a
STATED reason for the name. Who knows both?)
RwP
--
Ralph Wade Phillips, CET #LA-82
ral...@gcstation.net
Eric Fischer
[followups directed to a.f.c]
Ralph Wade Phillips <ral...@gcstation.net> wrote:
> (For the $64K question - The Apple Lisa had both a REAL and a
> STATED reason for the name. Who knows both?)
It supposedly stands for "Locally Integrated Software Architecture"
(yeah, right) but the story goes that it's really the name of Steve
Jobs's daughter.
Eric
Jay R. Jaeger
kb...@best.com (Ken Bibb) wrote:
>
>Most likely due to the amount of pressure required to depress the
>keys on a Decwriter...
DecWriter nothing. Try TELETYPE. Every character counted, both
input and output.
---
Jay R. Jaeger The Computer Collection
Jay.J...@msn.fullfeed.com visit http://www.msn.fullfeed.com/~cube
DoN. Nichols
In article <kbibb.847601696@shellx>, Ken Bibb <kb...@best.com> wrote:
>In <E0L92...@world.std.com> d...@world.std.com (Jeff DelPapa) writes:
>>The biggest conclusion that you can draw from the oldest of unix
>>commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
>>original authors were vwl alrgc.
>
>keys on a Decwriter...
I *hope* that you mean a Teletype. Every Decwriter keyboard that I
have ever used had an excellent keyboard. However, on an ASR-33 Teletype,
you could balance a broom, handle down, on a key without applying enough
force to cause it to send the character. :-)
--
Email: <dnic...@d-and-d.com> | Donald Nichols (DoN.)
Voice Days: (703) 704-2280 | Eves: (703) 938-4564
My Concertina web page: | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
Ling Wang
Paul A Vixie
> I *hope* that you mean a Teletype. Every Decwriter keyboard that I
>have ever used had an excellent keyboard. However, on an ASR-33 Teletype,
>you could balance a broom, handle down, on a key without applying enough
>force to cause it to send the character. :-)
Wasn't it GREAT? It has a mechanical interlock so that no two keys could
be pressed at the same time. This interlock was gradiated, such that as
pressure was released on the "down" key, the "up but would like to be down"
key would slide downward. After a few years of practice with this feedback
loop I learned to type at exactly 10CPS and could continue so for minutes
at a time. THOSE were the days. (My friends tell me that this must be why
I love the VT100's keyboard so much.) Carpal what?
--
Paul Vixie
La Honda, CA "Illegitimibus non carborundum."
<pa...@vix.com>
pacbell!vixie!paul
Clinton Pierce
Benjamin Hutchings wrote:
>
> In article <328386bc....@news.ov.com>,
> Pete Barber <pete....@openv.co.uk> wrote:
>
> I heard that Biff was the name of Ken Thompson's dog, which would bark
> loudly whenever the mailman came to deliver, but this could be another UL.
Taken (without Permission) from "A Quarter Century of UNIX", which
the starter of this thread oughta try reading...would save him a lot
of work:
[...]John Fodero, a graduate student at Berkeley in the summer of
1980, wrote a program that checked whether new mail had arrived. It
would tell a user "you have new mail". At the time, Heidi Stettner
worked in Evans Hall, prior to beginning graduate study. Heidi would
bring her dog with her to and to her office. He was a very friendly
dog,and a lot of the students enjoyed throwing a ball down the
corridor for him to fetch. He even had his picture on the bulletin
board with the graduate students: the legend read that he was working
on his Ph.Dog. John decided to name the program after the dog: Biff.
According to Heidi, John and Bill Joy then spent a lot of time trying
to compose an explanation for biff--they came up with "Be notified
if mail arrives." Biff, who died in August 1993, at 15, once got a B
in a compiler class. According to Heidi, the story of Biff barking at
Mike Czaplinski
jeve...@wwa.com (John Everett) wrote:
>
[snip re daemons]
> The daemon concept may predate its implementation on the PDP-10, I'm not
> sure. It's entirely possible it came from either Project MAC or Multics. I
> know that Tom Hastings and Dennis Ritchie both worked on Multics. Tom
> rejoined the PDP-10 monitor group after his stint at Multics, and of course
> Dennis joined Ken at Bell Labs and had a bit to do with the development of
> UNIX. BTW, the name UNIX is a play on Multics.
I heard of an alternate origin for the term 'Unix' from a cow orker
who was around Bell Labs in the mid 70's:
UNiversal Information eXchange.
Perhaps it's a retronym, but it sure fits both the name itself (since
it was always capitalized in it's earliest appearances) and the
function of the OS.
Mike "But, maybe it fits too good..." Czaplinski
mike.cz...@washingtondc.ncr.com
Peter da Silva
> commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
> original authors were vwl alrgc.
The original developers were working on ASR33 teletypes at 110 baud
and slower.
--
</peter>
DoN. Nichols
In article <3286fd63...@news.ov.com>,
Pete Barber <pete....@openv.co.uk> wrote:
>On this theme, do you know wy the language forth is spelt such. It
>was supposed to be a foUrth generation language but the IBM it was
>written on only supported five letter commands hence forth.
Well ... close but ...
As I heard it -- at a seminar pushing the language by its commercial
side, the author was an astromomer, and the language sort of grew as it
followed him from site to site. *One* of the systems on which he did some of
the development had such a limitation, but it had started out on a system
which was a little more forgiving, and had been spelled normally then.
Chris
Mike Czaplinski (mike.cz...@washingtondc.ncr.com) wrote:
: I heard of an alternate origin for the term 'Unix' from a cow orker
: who was around Bell Labs in the mid 70's:
: UNiversal Information eXchange.
taken (also without permission) from _Unix System V Release 4: An
Introduction_, by Kenneth Rosen, Richard Rosinski, and James Farber:
p. 12:
"...the late 1960s when MIT, AT&T Bell Labs, and...General
Electric worked on an experimental operating system called MULTICS.
MULTICS, for MULTiplexed Information and Computing System..."
[describes Ken Thompson and Dennis Ritchie rewriting Space Travel for
the DEC PDP-7, followed by an OS with file system and command
interpreter]
"Since the new multi-tasking operating system for the PDP-7
could support two simultaneous users, it was humorously called UNICS
for the UNiplexed Information and Computing System; the first use of
this name is attributed to Brian Kernighan. The name was changed
slightly to UNIX in 1970, and that has stuck ever since."
given that these guys have written an "introduction" that is roughly
1200 pages long, I'll go with their story...
cheers,
chris
Ken Bibb
>In article <kbibb.847601696@shellx>, Ken Bibb <kb...@best.com> wrote:
>>In <E0L92...@world.std.com> d...@world.std.com (Jeff DelPapa) writes:
>>>The biggest conclusion that you can draw from the oldest of unix
>>>commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
>>>original authors were vwl alrgc.
>>
>>Most likely due to the amount of pressure required to depress the
>>keys on a Decwriter...
> I *hope* that you mean a Teletype.
Yes, I meant a teletype--we used them on an HP mainframe during the
70s. The Decwriters I was exposed to were attached to a TOPS-20 system...
Pete Barber
Mike Czaplinski
Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
>
[snip]
> Sort of like to make a collection of the 'in-jokes' and other strange
> reasons things are like they are in unix..
There's always the fuser command.....
Mike "Check man fuser on a system to see what I mean" Czaplinski
mike.cz...@washingtondc.ncr.com
Tony Finch
Mike Czaplinski <mike.cz...@washingtondc.ncr.com> wrote:
>
> I heard of an alternate origin for the term 'Unix' from a cow orker
> who was around Bell Labs in the mid 70's:
>
> UNiversal Information eXchange.
>
> it was always capitalized in it's earliest appearances) and the
> function of the OS.
ISTR a paper by dmr (I think) in which he mentions that Unix became
UNIX only because when one of the earliest papers was written they had
recently become able to use small caps and therefore did so at the
slightest opportunity.
Tony.
Lew Pitcher
ral...@gcstation.net (Ralph Wade Phillips) wrote:
>Howdy!
snip
> The "Olympus TuneUp" help file lists GREP as "Global Regular
>Expression Parser".
> Dunno how accurate, but it both fits the acronmym !AND! describes
>what GREP does ... <B-)
Not very accurate. According to "The Unix Programmer's Manual - Volume
2" (Bell Labs, 1979) in the article "Advanced Editing on UNIX" by
Brian W. Kernighan (Aug. 4 1978), "The program grep was invented to
get around these limitations <in multifile editing LP>. The search
patterns that we have described in the paper are often called 'regular
expressions', and 'grep' stands for g/re/p <UNIX ed command string
LP> That describes exactly what grep does -- it prints every line in
a set of files that contains a particular regular expression."
g/re/p was the mnemonic for the ed command string that located and
printed strings: g/UNIX/p would display every occurrance of the string
'UNIX' in the file being edited, while g/fubar/p would do the same for
the string 'fubar'. This could be extended from literal matches to
regular expression matches a la g/U*X/p or g/f*bar/p.
Thus endeth the history lesson. ;-)
> (For the $64K question - The Apple Lisa had both a REAL and a
>STATED reason for the name. Who knows both?)
> RwP
>--
>Ralph Wade Phillips, CET #LA-82
>ral...@gcstation.net
Lew Pitcher | "I'm a little source code
Toronto Dominion Bank | Short and Stout
======================= | Here is my Input,
Enzo Matrix - Reboot | And here is my out"
Shawn Barnhart
> In <E0L92...@world.std.com> d...@world.std.com (Jeff DelPapa) writes:
>
> >The biggest conclusion that you can draw from the oldest of unix
> >commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
> >original authors were vwl alrgc.
I thought I read someplace that one of the original incarnations of Unix
could only support two character file names. Hence many of the basic
system commands (ls, mv, cp, rm, cd, and so forth) are only two
characters. But I like the idea that it was attributed to the amount of
effort to type the commands into a printing terminal.
--
Shawn Barnhart
s...@mercury.campbell-mithun.com
Shawn Barnhart
Hugh J.E. Davies <hda...@kzin.mon.rnb.com> wrote:
> s...@mercury.campbell-mithun.com (Shawn Barnhart) writes:
> >I thought I read someplace that one of the original incarnations of Unix
> >could only support two character file names.
>
> Err, no.
>
> > Hence many of the basic
> >system commands (ls, mv, cp, rm, cd, and so forth) are only two
> >characters. But I like the idea that it was attributed to the amount of
> >effort to type the commands into a printing terminal.
>
> And this is the accepted explaination.
Are there many other OS of Unix vintage that have as many two character
commands? I can't remember the HP2000 we used in elementary school (ca
1976) having two character commands, but I do remember using clunky
teletypes on them. Certainly when that machine was developed, there
were no more high speed video terminals in use than in the Unix
development time period.
--
Shawn Barnhart
s...@mercury.campbell-mithun.com
Alan Bowler
In article <E0qH5...@world.std.com> d...@world.std.com (Jeff DelPapa) writes:
>
>Make that an ASR-33. The decwriter didn't exist when unix got
>written. 10 CPS, and oil everything but the ribbon.
Why not not the ribbon? A few drops of three in one was an easy
way to squeeze a little more life out of a ribbon cartridge on a
daisy wheel printer, it would probably work on TTY 33 or TTY 35 just
as well.
John Bayko
In article <1996111206190712643@[192.159.32.2]>,
Shawn Barnhart <s...@mercury.campbell-mithun.com> wrote:
>
>I thought I read someplace that one of the original incarnations of Unix
>system commands (ls, mv, cp, rm, cd, and so forth) are only two
>characters.
Except that they were all kept in /bin, which is a three
character name.
No serious operating system would ever place such a restrictive
limit on file names.
--
John Bayko (Tau).
ba...@cs.uregina.ca
http://www.cs.uregina.ca/~bayko
Gene Wirchenko
s...@mercury.campbell-mithun.com (Shawn Barnhart) wrote:
[snip]
>Are there many other OS of Unix vintage that have as many two character
>commands? I can't remember the HP2000 we used in elementary school (ca
>1976) having two character commands, but I do remember using clunky
>teletypes on them. Certainly when that machine was developed, there
>were no more high speed video terminals in use than in the Unix
>development time period.
On HP2000F BASIC, only the first three chars of a command were
significant. Filenames were up to six (I think) chars in length.
Sincerely,
Gene Wirchenko
C Pronunciation Guide:
y=x++; "wye equals ex plus plus semicolon"
x=x++; "ex equals ex doublecross semicolon"
Peter da Silva
In article <1996111212371894551@[192.159.32.153]>,
Shawn Barnhart <s...@mercury.campbell-mithun.com> wrote:
> Are there many other OS of Unix vintage that have as many two character
> commands?
Most of the ones I used allowed 1- or 2- character abbreviations for most
commands. This was resurrected later on when membrane keyboards became
briefly popular.
--
</peter>
Tim Hollebeek
In article <ralphp-1011...@du-18.gcstation.net>, ral...@gcstation.net writes:
> Howdy!
>
> In article <328631...@jpmorgan.com>, Ling Wang
> <wang...@jpmorgan.com> wrote:
>
> > Melinda Shore wrote:
> > >
> > > In article <328386bc....@news.ov.com>,
> > > Pete Barber <pete....@openv.co.uk> wrote:
> > > >grep GetREgularePression
> > >
> > > Way incorrect. "Get Regular, Eat Prunes."
> > >
> > > If you aren't sure about something, look it up before posting.
> >
> > Shouldn't it be "Go Regularly, Eat Prunes"?
>
> The "Olympus TuneUp" help file lists GREP as "Global Regular
> Expression Parser".
Bah. Doesn't anyone know 'ex' (the non-visual version of vi) any more?
Objective: to print every line an a file that contains the regular
expression 're'
Solution: type in "g/re/p"
Example:
---
franck:/var/adm:123> ex SYSLOG
"SYSLOG" [Read only] 73 lines, 6043 characters
:g/Xsession/p
10 Nov 3 06:19:35 6B:franck Xsession: tim: logout
11 Nov 3 13:48:46 6B:franck Xsession: tim: login
18 Nov 4 01:40:42 6B:franck Xsession: tim: logout
19 Nov 4 12:00:25 6B:franck Xsession: tim: login
24 Nov 4 15:56:21 6B:franck Xsession: tim: logout
36 Nov 6 13:48:29 6B:franck Xsession: tim: login
37 Nov 7 00:19:04 6B:franck Xsession: tim: logout
38 Nov 7 12:58:00 6B:franck Xsession: tim: login
68 Nov 10 17:47:47 6B:franck Xsession: tim: logout
69 Nov 11 13:40:36 6B:franck Xsession: tim: login
---
Hence the name. 'g' stands for "global"; the 'p' means "print".
So roughly, "global, search for re, and print".
---------------------------------------------------------------------------
Tim Hollebeek | Disclaimer :=> Everything above is a true statement,
Electron Psychologist | for sufficiently false values of true.
Princeton University | email: t...@wfn-shop.princeton.edu
----------------------| http://wfn-shop.princeton.edu/~tim (NEW! IMPROVED!)
William J. Hayes
Mark Whetzel (ma...@zeus.wg.waii.com) wrote:
: I was having a discussion with some of my co-workers on WHY things
: are named as they are in UNIX. Anybody have some of the
: tales behind the odd things that bound in unix history?
: One question we had of some students we are training:
: Why are system processes called 'daemons'?
After the traditional meaning, "A diabolic creature, of minimal
intelligence, bound to ones will, to perform boring or repetitive
chores".
: Some of the others I have heard about: the naming of 'biff', grep, yacc.
biff:
The name of the dog which belonged to the programmer at UC
Berkeley who wrote the program. Dog would bark whenever the mailman
showed up.
grep:
"Generic Regular Expression Printer"
yacc:
"Yet Another Compiler Compiler"
: E-mail preferred, as I can then post a summary of the BEST of this quest.
These are easy, with definitive answers.
: References to printed material and other net resources appreciated.
Find "The Jargon File"
: --
: Mark Whetzel My comments are my own, not my company's.
: Western Geophysical - A division of Western Atlas International Inc.,
: A Subsidiary of Western Atlas Inc. DOMAIN addr: mark.w...@waii.com
: VOICE: (713) 963-2544 FAX: (713) 963-2758
Ron Hunsinger
In article <1996111212371894551@[192.159.32.153]>,
> Are there many other OS of Unix vintage that have as many two character
> commands? I can't remember the HP2000 we used in elementary school (ca
> 1976) having two character commands, but I do remember using clunky
> teletypes on them. Certainly when that machine was developed, there
> were no more high speed video terminals in use than in the Unix
> development time period.
The Burroughs (now Unisys) Medium Systems MCP (Master Control Program) uses
ONLY two-letter commands at the ODT (Operator Display Terminal).
Thus you get things like:
DS - DiScontinue program
DM - DuMp memory for program
DP - DumP and discontinue
Most commands took parameters, like which program was to be affected. Some
did not, and would ignore any you supplied. Since all commands were exactly
two letters long, there was no need to separate the command from the
parameters.
For example, the SL command (SPO List) would produce a list of all the
remote SPOs (SuPervisOry consoles). Since there usually weren't any, the
usual response was "NONE".
It was also an obscure enough command that not many people knew it. You
could (momentarily) blow someone's mind by typing in
SLEPT WITH ANY VIRGINS LATELY?
or
SLAIN ANY DRAGONS LATELY?
and letting them figure out how the computer knew to say: NONE.
-Ron Hunsinger
Alan Bowler
In article <1996111206190712643@[192.159.32.2]> s...@mercury.campbell-mithun.com (Shawn Barnhart) writes:
>> In <E0L92...@world.std.com> d...@world.std.com (Jeff DelPapa) writes:
>>
>> >The biggest conclusion that you can draw from the oldest of unix
>> >commands, (cp,rm,ln,cmp, and directories like /tmp /usr) is that the
>> >original authors were vwl alrgc.
>
>could only support two character file names. Hence many of the basic
>system commands (ls, mv, cp, rm, cd, and so forth) are only two
>effort to type the commands into a printing terminal.
>
described the function and a short abbreviation that you could type
in a reasonable amount of time. People of course only remembered
the abbreviations that they actually used. Many of the Unix commands
are about the same name as the Multics abbreviations.
Steve Blinkhorn
This discussion is getting seriously nostalgic for me, what with
ASR-33s (which were in fact introduced as subversive micro-confetti
manufacturing devices, since any attempt to *read* a paper tape of any
length invariably resulted in two shorter paper tapes), PDP-10s
(favourite memory: DEC engineer leaving computer room talking to chief
systems programer ``It's a software problem'' as smoke started to
issue from cabinet), someone who knows how to pronounce daemon
properly.
Next we'll be hearing from people who know what it was like to have no
more than 100K 36-bit words to do real programming in.
Is there a comp.nostalgia group (INPUT IN CAPS ONLY at 10cps, wearing
ear defenders)?
--
Steve Blinkhorn <st...@prd.co.uk>
Mark Sienkiewicz
In article <1996111212371894551@[192.159.32.153]>,
>
>Are there many other OS of Unix vintage that have as many two character
>commands? I can't remember the HP2000 we used in elementary school (ca
>1976) having two character commands, but I do remember using clunky
>teletypes on them. Certainly when that machine was developed, there
>were no more high speed video terminals in use than in the Unix
>development time period.
It used to be common to allow abbeviations of command names. I
recall seeing this in RT-11, TOPS-10, RSTS, VMS, VM/CMS, CANDE,
and some HP thing (MPE?) to name a few. You only had to type a
unique prefix of the command name, so if the command to copy a file
was "COPY", you could just type
C FILE.1 FILE.2 (this was before lower case was invented :)
but it would break horribly when the next release of the OS came out
with a command named "COMPARE" and now you have to type
COP FILE.1 FILE.2
The explicit explanation for this feature was to make it easier to type
commands.
On Unix, you just type a program name to run it. On some of these
other systems, there is a list of valid commands that the user can
type. From the list, you can recognize unique abbreviations
pretty easily. Of course, one of those is probably "RUN programname",
usually abbreviated "R".
It makes as much sense to give the programs shorter names as it
does to read all the file names on your path to determine unique
abbreviations.
You were also dealing with a mindset at the time that names had to
be short. There were lots of things like the 6 character variable
names in FORTRAN, the 2 and 3 character mnemonics for machine
instructions, etc etc that made people tend to shorten names
automatically. DEC even had a character set encoding that
let you squeeze 3 characters into 2 bytes (thus the 3 characters
extension in file names).
I'd be surprised if this mindset isn't also part of it, at least
subconciously.
Simon Slavin
In article <564jdb$4...@ordeal.cts.com>,
c...@cts.com (Will Rose) wrote:
> Tony Finch (fa...@cam.ac.uk) wrote:
> : pete....@openv.co.uk (Pete Barber) wrote:
> : > Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
> : >
> : > >Some of the others I have heard about: the naming of 'biff',
> : >
> : > biff - dunno
>
> : Supposedly after a dog that would bark at the postman.
>
> After a dog, but not one that barked at the postman. According
> to the dogs owner this was a foul canard, or dirty duck.
Hold on, there's an evil-smelling duck that thinks it's a dog and
barks at postmen. Does anyone have the film rights to this ?
Simon.
--
... or, here's a thought, why don't we talk about SF for a while?
-- Simon van Dongen <sg...@pi.net> on rec.arts.sf.written
Simon Slavin -- Computer Contractor Ordinaire. Junk email not welcome here.
Will administer ISO 9000 and year 2000 certification tests for food.
Perry Hutchison
Peter da Silva wrote:
> gres was replaced by sed in version 7 ...
I think you may be off by one version. I distinctly recall
sed being in 6th Edition (and I thereby date myself :-).
--
Not speaking for Intel.
Ųyvind Gjerstad
s...@mercury.campbell-mithun.com (Shawn Barnhart) writes:
> Are there many other OS of Unix vintage that have as many two character
> commands?
The DNOS from Texas Instruments also used a lot of 2-character commands,
but it could use up to 6 characters (uppercase only) for command names.
LF=List Files
SF=Show Files
LD=List Directory
CC=Copy/Concatenate
LS=List Synonyms (synonyms are kind of persistent environment-variables)
DD=Delete Directory
CD=Copy Directory
LJ=List Jobs
XE=eXecue Editor
(the list goes on)
All commands are interaktive, that means you type the command
and is prompted for arguments. Filenames will default to the last file
you visited, and if you just want to use the default arguments, you
can enter at "." after the command.
--
Øyvind Gjerstad Systems dept Tollpost-Globe AS N-6301 Åndalsnes/Norway
E-mail: o...@it.tollpost.no Phone: +47 7122 6663 Fax: +47 7122 6694
Glenn Chambers
> grep:
> "Generic Regular Expression Printer"
Clever, but not correct.
grep is from the ed 'g' command, which had the syntax:
g/<regular expression>/<command>
where 'command' was any normal ed command, such as 'p', or 's/old/new/'.
Thus,
g/a[0-9]*/p
printed all lines containing the letter 'a' followed by zero or more digits.
If you describe this command as 'g/re/p', you'll immediately see where the
name of the program came from.
--
Glenn Chambers
gcha...@mail.bright.net
Toledo, OH
Dik T. Winter
In article <steve.8...@fastnet.prd.co.uk> st...@fastnet.prd.co.uk (Steve Blinkhorn) writes:
> Next we'll be hearing from people who know what it was like to have no
> more than 100K 36-bit words to do real programming in.
Does 32K 27-bit words count?
--
dik t. winter, cwi, kruislaan 413, 1098 sj amsterdam, nederland, +31205924131
home: bovenover 215, 1025 jn amsterdam, nederland; http://www.cwi.nl/~dik/
Steinar Haug
[Dik T. Winter]
| > Next we'll be hearing from people who know what it was like to have no
| > more than 100K 36-bit words to do real programming in.
|
| Does 32K 27-bit words count?
How about 64K 16-bit words? Are we trying to set a new record here? :-)
Steinar Haug, Nethelp consulting, sth...@nethelp.no
Dik T. Winter
In article <56fdha$r...@verdi.nethelp.no> sth...@nethelp.no (Steinar Haug) writes:
> [Dik T. Winter]
>
> | > Next we'll be hearing from people who know what it was like to have no
> | > more than 100K 36-bit words to do real programming in.
> |
> | Does 32K 27-bit words count?
>
> How about 64K 16-bit words? Are we trying to set a new record here? :-)
We are talking real programming. That 32K 27-bit word machine was used
as main (non-administrative) computer for two universities and one (or
perhaps more) research institute.
Peter Ludemann
s...@mercury.campbell-mithun.com (Shawn Barnhart) writes:
> Are there many other OS of Unix vintage that have as many two character
> commands? [snip]
The operator's console for IBM's OS/360 (MFT, MVT, etc., and as far as
I know this is still true for MVS) would accept single-letter
abbreviations for commands. Some were obvious, such "R" for "REPLY"
and "Q" for "QUERY", but others were more obscure such as "A" for
"RELEASE". The oldest consoles were 1050s, essentially modified
Selectric typewriters (probably the greatest design ever in terms of
keyboard feel), so clunkiness of input was not an excuse, as was
(allegedly) the case with Unix's 2-letter commands.
I seem to remember 3-letter commands for RT-11, but the only one that
comes to mind now is "PIP" (Peripheral Interchange Program"). On
PDP-11s, there was a straightforward method of packing 3 characters
into a 2-byte word, so file names were 6 characters, with a
3-character extension (I suppose MS-DOS copied the 3-character
extension, but without the 3-to-2 packing).
Peter da Silva
In article <328A06...@ccm.hf.intel.com>,
Perry Hutchison <perry_h...@ccm.hf.intel.com> wrote:
> I think you may be off by one version. I distinctly recall
> sed being in 6th Edition (and I thereby date myself :-).
There was no "sed" on the Computer Center computers at UCB in 1980. It
was on the EECS computer. The CC machines ran V6, the EECS machine
(Cory) was running V7. It is possible that the CC staff removed "sed"...
there were a number of oddball changes in the sommands available there...
but unless I can dig up my old V6 manual there's no way I can prove it
one way or the other.
--
</peter>
Peter da Silva
In article <E0tAts.BA...@torfree.net>,
No, no, you're wrong. It's from "Generic Repeated Pattern Evaluator", an old
IBM mainframe program that searched for text in 80 column card decks and
edited them into new decks. When it was ported to UNIX (using the "struct"
tool to convert the original Fortran to Ratfor, then to C) they couldn't
fit the whole functionality in so they only provided the "print" edit.
So it was called "grpep", then shortened to "grep" after Bill Gates made
a hilarious typo in the Xenix-86 version at a trade show. The actual details
were hushed up, and they changed the command to make sure it never happened
again.
The story was created that it was shortened because it was easier to type
"grep" than "grpep" on an IBM card punch, but this was an obvious cover-up,
since the "grpep" version was never used on an IBM.
Ironically the full functionality of "grpe" was ported to micros for the
first time on OS/2 in an implementation written in REXX (named after a dog
that barked when his owner sat down at a computer).
--
</peter>
Melinda Shore
In article <56g0mm$g...@web.nmti.com>, Peter da Silva <pe...@nmti.com> wrote:
>In article <E0tAts.BA...@torfree.net>,
>William J. Hayes <ab...@torfree.net> wrote:
>> grep:
>> "Generic Regular Expression Printer"
>No, no, you're wrong. It's from "Generic Repeated Pattern Evaluator", an old
>IBM mainframe program that searched for text in 80 column card decks and
>edited them into new decks.
Huh. The way *I've* heard it (so it must be true and I
don't actually have to look it up or anything) is that it
actually was derived from "regular global pattern
evaluator," but that at that trade show you mentioned Gates
was embarrassed to find "rgpe" nuxied into "grep" by a
filesystem bug in the 8086 port. The rest, as they say, is
history.
--
Melinda Shore - No Mountain Software - sh...@nomt.com
If you send me harassing email, I'll probably post it
Perry Hutchison
Tim Hollebeek wrote:
> Bah. Doesn't anyone know 'ex' (the non-visual version of vi) any more?
and humbug. Even the curmudgeons seem to have forgotten ed(1), the
precursor of ex. ed was the only editor supplied with 6th Edition.
Adam J. Thornton
>as main (non-administrative) computer for two universities and one (or
>perhaps more) research institute.
If you mean the machine I think you do, it also supported an incredibly
elegant hierarchical batch operating system, whose algorithms were provably
correct, and which is at the core of one of the finest CS papers ever
written.
Adam
--
"I'd buy me a used car lot, and | ad...@princeton.edu | As B/4 | Save the choad!
I'd never sell any of 'em, just | "Skippy, you little fool, you are off on an-
drive me a different car every day | other of your senseless and retrograde
depending on how I feel.":Tom Waits| little journeys.": Thomas Pynchon | 64,928
Marco S Hyman
Jay.J...@msn.fullfeed.com (Jay R. Jaeger) writes:
> st...@fastnet.prd.co.uk (Steve Blinkhorn) wrote:
>
> <snip>
> > someone who knows how to pronounce daemon
> >properly.
> >
>
> According to whom? The dictionary I ahve lists the pronunciation as
> dEmen (long E). However, I know lots of would be Unix hacks who, for
> some reason, think it should be pronounced dAmen (long A). Or is this
> one of those British English vs. US English debates?
Ahhh, I knew there was a second reason I keep the Compact OED around.
(The first reason is the magnifier that comes with it :-) Lets see what
it has to say:
daemon, daemonic, etc: see DEMON, etc.
There, that settles it!
// marc
Paul Eggert
pe...@nmti.com (Peter da Silva) writes:
> There was no "sed" on the Computer Center computers at UCB in 1980. It
> was on the EECS computer. The CC machines ran V6, the EECS machine
> (Cory) was running V7. It is possible that the CC staff removed "sed"...
I remember "sed" being on the V6 PDP-11/45s at UCLA in the late 1970s.
(I noted it because sed's author Lee McMahon and I had been fellow students
at Rice in the early 1970s.) I suppose it could have been an update to V6,
as opposed to the original V6 -- or perhaps it was lifted from V7 by the
UCLA sysadmins. V6 was sort of a moving target, after all.
Jay R. Jaeger
st...@fastnet.prd.co.uk (Steve Blinkhorn) wrote:
<snip>
> someone who knows how to pronounce daemon
>properly.
>
<snip>
According to whom? The dictionary I ahve lists the pronunciation as
dEmen (long E). However, I know lots of would be Unix hacks who, for
some reason, think it should be pronounced dAmen (long A). Or is this
one of those British English vs. US English debates?
---
Jay R. Jaeger The Computer Collection
Jay.J...@msn.fullfeed.com visit http://www.msn.fullfeed.com/~cube
Jay R. Jaeger
pe...@nmti.com (Peter da Silva) wrote:
>In article <E0tAts.BA...@torfree.net>,
>William J. Hayes <ab...@torfree.net> wrote:
>> grep:
>> "Generic Regular Expression Printer"
>
>
>So it was called "grpep", then shortened to "grep" after Bill Gates made
>a hilarious typo in the Xenix-86 version at a trade show. The actual details
>were hushed up, and they changed the command to make sure it never happened
>again.
>
Baloney. grep existed in Unix 6th edition, cir. 1976, long before
there was a PC. I'd have to check, but I'd bet it was even in the
5th edition, a year before. Or was your explanation intended to be
flame bait?
It is, of course, as others here have pointed out, really derived from
the "ed" command sequence: g/<regular expression/p, i.e. Globally
search for a Regular Expression, and Print.
Sigh, these folks that started on PC's need some real history of
computing education ;-)
Kin Hoong CHUNG
Steve Blinkhorn (st...@fastnet.prd.co.uk) wrote:
: This discussion is getting seriously nostalgic for me, what with
: ASR-33s (which were in fact introduced as subversive micro-confetti
: manufacturing devices, since any attempt to *read* a paper tape of any
: length invariably resulted in two shorter paper tapes), PDP-10s
: (favourite memory: DEC engineer leaving computer room talking to chief
: systems programer ``It's a software problem'' as smoke started to
: properly.
: Next we'll be hearing from people who know what it was like to have no
: more than 100K 36-bit words to do real programming in.
: Is there a comp.nostalgia group (INPUT IN CAPS ONLY at 10cps, wearing
: ear defenders)?
Don't know where you are reading this from... but this is supposed to
be the purpose of alt.folklore.ocmputers, and the moderated
comp.society.folklore are for. Well, in principle anyway, as it is clear
that too many people misuse alt.folklore.computers.
ObTopic: more on UNIX has an obvious name, and less is a pun on more.
ObFolklore: remember when more was a simple pager which did nothing other
than stopping the page running away from you?
Cheers,
Kin Hoong
Faried Nawaz
one fish caught.
John Winters
In article <56fdha$r...@verdi.nethelp.no>,
Steinar Haug <sth...@nethelp.no> wrote:
>[Dik T. Winter]
>
>| > Next we'll be hearing from people who know what it was like to have no
>| > more than 100K 36-bit words to do real programming in.
>|
>
How about 1K 8-bit bytes (and that included the display memory)? :-)
John
(Yes, yes, I know. We're talking about real computers, not things
that make even an IBM PC look sophisticated.)
--
John Winters. Wallingford, Oxon, England.
Dwight Tovey
My first "computer" had 256 bytes of ram. Had one register to hold a return
address, so the call stack was only one deep. The program code had to live in
rom, so the code had to be tight to save space. If you wanted to add something,
you first had to go through the existing code to see if you could do something in
fewer instructions so that you could make room for the addition.
The thing was used to generate line-printer test patterns. Barely qualifies as a
computer, but it was an interesting way to learn programming.
/dwight
--
Dwight N. Tovey H&W Computer Systems, Inc.
Software Specialist III 12438 W. Bridger St. Suite 100
dwi...@hwcs.com Boise, ID. 83713
or dwi...@micron.net (208)377-0336
I didn't claw my way to the top of the food chain to eat vegetables!!
Peter da Silva
In article <328cb5ad...@news.msn.fullfeed.com>,
Jay R. Jaeger <Jay.J...@msn.fullfeed.com> wrote:
> Baloney. grep existed in Unix 6th edition, cir. 1976, long before
> there was a PC. I'd have to check, but I'd bet it was even in the
> 5th edition, a year before. Or was your explanation intended to be
> flame bait?
Spock: Captain, I believe this planet's life-forms are
capable of delayed clue reception.
Kirk: Damn the prime directive, then, beam me down so
I can get me some alien babes!
Spock: This is most illogical behaviour, Captain. Or are
you attempting to use parody to make a point.
--
</peter>
Joe Morris
Peter Ludemann <lude...@expernet26.expernet.com> writes:
>The operator's console for IBM's OS/360 (MFT, MVT, etc., and as far as
>I know this is still true for MVS) would accept single-letter
>abbreviations for commands. Some were obvious, such "R" for "REPLY"
>and "Q" for "QUERY", but others were more obscure such as "A" for
>"RELEASE". The oldest consoles were 1050s, essentially modified
>Selectric typewriters (probably the greatest design ever in terms of
>keyboard feel), so clunkiness of input was not an excuse, as was
>(allegedly) the case with Unix's 2-letter commands.
Minor quibble, but the S/360 console was a type 1052 machine (model 7),
not a 1050. Like all Selectrics, it ran at 14.6 characters/second,
so while it was faster than a 10 cps TeleType, it still was slower
than anything we're used to today. The machine type (1052) came from
its use as one of the components of the 1050 family of remote terminals
components, from which it was taken and modified to become the standard
S/360 console.
The abbreviations (as noted above) for the commands were single-character
and where possible used the first character of the full name, without
worrying about ambiguity. (R was the abbreviation for "REPLY" and not
for "RELEASE"). Where the first letter was already in use the abbreviation
would be the first letter of the full command name that wasn't already
spoken for -- for example, "A" was used for RELEASE because "R", "E",
and "L" were already in use.
In some cases even this failed: the HALT command ("HALT EOD" was roughly
analogous to the UNIX command 'shutdown -h'; "EOD" stood for "End Of Day")
had *no* characters not already in use, so it got the abbreviation "Z".
Joe Morris / MITRE
Auke van Slooten
In article <wzzq0kf...@expernet26.expernet.com>,
Peter Ludemann <lude...@expernet26.expernet.com> writes:
>
>I seem to remember 3-letter commands for RT-11, but the only one that
>comes to mind now is "PIP" (Peripheral Interchange Program"). On
>PDP-11s, there was a straightforward method of packing 3 characters
>into a 2-byte word, so file names were 6 characters, with a
>3-character extension (I suppose MS-DOS copied the 3-character
>extension, but without the 3-to-2 packing).
>
command. It is however a utility used by a lot of the RT-11 commands, and
indeed, according to my RT-11 mini reference manual all the commands could
be abreviated to 3 or even 2 letter. The only exeption being FORMAT, which
took 4 letters.
Hmm.. maybe I'll take another shot at bringing one of my pdp's online...
Auke van Slooten
Adam J. Thornton
In article <56g31n$9...@panix2.panix.com>, Melinda Shore <sh...@nomt.com> wrote:
>evaluator," but that at that trade show you mentioned Gates
>was embarrassed to find "rgpe" nuxied into "grep" by a
>filesystem bug in the 8086 port. The rest, as they say, is
>history.
It was just an endianness problem, anyway. Only a filesystem bug in the
most tangential sense.
Alan Bowler
In article <E0v6z...@cwi.nl> d...@cwi.nl (Dik T. Winter) writes:
>In article <steve.8...@fastnet.prd.co.uk> st...@fastnet.prd.co.uk (Steve Blinkhorn) writes:
> > more than 100K 36-bit words to do real programming in.
>
>Does 32K 27-bit words count?
I still live with a 180k 36-bit limit.
Tom I Helbekkmo
[Perry Hutchison]
> I think you may be off by one version. I distinctly recall
> sed being in 6th Edition (and I thereby date myself :-).
[Peter da Silva]
> There was no "sed" on the Computer Center computers at UCB in 1980. It
> was on the EECS computer. The CC machines ran V6, the EECS machine
> (Cory) was running V7. It is possible that the CC staff removed "sed"...
There's no sed in the V6 distribution I've got running here, and it's
supposed to be a complete V6 from Bell. Of course, there weren't
clear cut V? distributions; the number just changed from time to time.
-tih
--
Popularity is the hallmark of mediocrity. --Niles Crane, "Frasier"
Alan Bowler
In article <56ai85$i...@sue.cc.uregina.ca> ba...@BOREALIS.CS.UREGINA.CA (John Bayko) writes:
>In article <1996111206190712643@[192.159.32.2]>,
> Shawn Barnhart <s...@mercury.campbell-mithun.com> wrote:
>>
>>I thought I read someplace that one of the original incarnations of Unix
>>could only support two character file names. Hence many of the basic
>>system commands (ls, mv, cp, rm, cd, and so forth) are only two
>>characters.
>
> Except that they were all kept in /bin, which is a three
>character name.
> No serious operating system would ever place such a restrictive
>limit on file names.
These days when most command names are just the names of files in some
directory we have lost the distinction between command names and file
names. It was not always so. In many of the older systems you have a
set of commands known to the command interpreter. One of the commands
is likely "run <filename>" or something like that, so there is a clear
distinction between system commands and user programs. The length of
these command names was/is often limited so it will fit in a table.
On Gcos8 timesharing it is 4 characters. On MarkIII (the system that
underlies the Genie service), it is three characters.
Brendan McKeon
c...@bankersnet.co.uk (Chris Hedley) writes:
> I'm sorry, you're totally wrong. GREP was ported to UNIX from its
> original inception on NT. [...]
> The GROPE program, along with the rest of VMS, was written for Sega's
> PDP-10 "mainframe" (a rackmounted PC) back in 1988 by Bill Gates and
> Alan Turing when they worked together sweeping floors at a Burger King
> in Istanbul.
Indeed. And it was while cleaning under the tables at that restaurant
that Bill got the idea for Bubble Gum memory, which he later sold to
IBM, - in exchange, they would use his fries. While the bubble gum
memory was a failure (it was only useful for storing sticky bits), IBM
later contracted him to provide large quantaties of sauce. (I can't
remember what sort of sauce it was - it started off with plenty of
character, then later had a sour apple taste, but anyhow took years to
ketchup.) He later went into the glazing business, and hired a vaccuum
cleaner engineer to create a self-cleaning window (it sucked) called
'Aunty'. (Was sort of a success, although the special glass used was
very heavy.) Many people referred to it the name of country in which
that Burger King was situated.
As an aside, Alan was later promoted to Human Resources at that
particular burger joint (involved hiring and training cash-register
attendants to grin in that special way, etc). He gained fame for
creating a simple test for determining if a certain person *lacked* the
qualities necessary for the job. Hence only applicants who failed the
so-called Turing test got the job. (Oddly enough, the first person to
get a job after the test was put in place was called Elisabeth, who was
at the time majoring in psychoanalysis. She had an interesting way of
daling with the customers:
Customer: Quarter pounder and large cola, please.
Elizabeth: HOW DO YOU FEEL ABOUT DEATH?
Customer: Err, ok, make it a veggie burger instead...
Elizabeth: YOU WANT FRIES WITH THAT?
... and so on. She was fired after a guy called Parry joined up - they
spent the whole time arguing. Last I heard, she was working in the
Chineese next door, owned by some guy called Searle...)
HTH, HAND.
Brendan.
--
char c[39],i,j;main(){srandom(time(0)/*/ Brendan McKeon /*/);for(
;i++<11;){for(j=39;j--;)printf((c[j]/*/ bmc...@tcd.ie /*/=random
()%2)?"/ ":" \\");putchar('\n');for/*/ http://alf2.tcd.ie/~bmckeon /*/(j=39;j;
)printf(c[--j]?" /":"\\ ");putchar/*/ /*/('\n');}}
John G Dobnick
From article <1996111206190712643@[192.159.32.2]>, by s...@mercury.campbell-mithun.com (Shawn Barnhart):
> characters. But I like the idea that it was attributed to the amount of
> effort to type the commands into a printing terminal.
Perhaps the most important reason, or at least the driving force behind
short command names, was the TELETYPE itself! You can type at a
_maximum_ rate of 10 characters per second. Period. You have no-key
rollover -- mechanical interlocks prevent depressing one key until
after the previous key has released. (Anyone timed their "burst rate"
typing recently? I know I can type, in bursts, much faster than 10 CPS.)
So, given a fixed, and slow, rate of typing being inposed on you, and
probably being the impatient type (we're talking programmers here,
after all :-) ), you do what you can to reduce keystrokes. Reducing
command names to minimal length helps, especially for the "common"
programmer commands.
That's my theory, at least. (Based on personal experience with
TELETYPE devices.)
--
John G Dobnick "Knowing how things work is the basis
Information & Media Technologies for appreciation, and is thus a
University of Wisconsin - Milwaukee source of civilized delight."
j...@csd.uwm.edu ATTnet: (414) 229-5727 -- William Safire
John G Dobnick
From article <569rkk$g...@orac.mon.rnb.com>, by hda...@kzin.mon.rnb.com (Hugh J.E. Davies):
> In article <1996111206190712643@[192.159.32.2]>, s...@mercury.campbell-mithun.com (Shawn Barnhart) writes:
>
[words about why Unix command names area short and "cryptic",
because programmers don't like to type.]
> And this is the accepted explaination.
You bet it is! Why type three or four characters when
two will do? Why "copy" when "cp" suffices? Why "list" when
"ls" does fine? (Although I aliases this to "l" -- why type 2
characters when 1 willdo? :-) )
Programmers are lazy [1] typists. You like to type, become a
stenographer.
[1] lazy -- efficient of keystroke,
conservative of finger motion,
reducer of muscular activity,
preventer of keyboard wear and tear,
lovers of abbreviations,
attempting to type as fast as they think.
What? You *never* use abbreviations?
Jonathan Dursi
In article <84807666...@balti.bankersnet.co.uk>,
Chris Hedley <c...@bankersnet.co.uk> wrote:
>
>I'm sorry, you're totally wrong. GREP was ported to UNIX from its
Revisionists.
Even if there was a VMS program called GROPE, that clearly wasn't
the origin of the command. It predates VMS by quite some time.
In the days not too long after ENIAC, it soon became clear to such
luminaries as von Neumann that a quick, efficient way of scanning
memory for certain patterns would be extremely useful; one could set up
`bit-patterns' that would be recognizable as a flag that some event had
occurred in the system, such as a calculation was complete. Then a
bell could be run, or a light turned on, or an electric charge could be
run through the cabinet; anyway, you get the idea, some way of letting
the `operators' know that something had occurred could be done.
A circuit, then, that examined the voltage patterns of the various
electron tubes was occasionly wired up by the technicians; it
was, originally, labelled a `Vaccuum Tube Grid Electric Charge Probe'
by the Americans. The name was was both awkward, and confused the
Brits who didn't use nice reasonable terms like `Vaccuum tube'.
Eventually, it got cut down to `Grid electric charge probe',
`GRid Electric Probe', and then, finally, grep.
This was pretty much nearly forgotten until the days of the TRS-80s,
when designers put the same sort of function into early ROMS of some of
these machines; it searched RAM for specified bit-patterns. It wasn't
intended to be a feature for use by users; it was solely a system thing,
but it wasn't long before clever users discovered what it could do,
and it often found its way into `The super duper book of pokes and peeks',
and literature of that nature. It was sometimes labled `the Great Poke',
as it was considerably more interesting than pokes which merely changed
colours; thus, ironically, the pseudo-acronym `grep' occasionally occurred
in this context, as well.
Finally, Kernigan and Ritchie when they were first implementing UNIX
for the TRS-80 model 4s, (it never really worked as well; that's why
they switched over to PDP-10s), discovered this feature, and started
figuring out how it could be used. Early implementations loaded
a file entirely into memory and called the Great Poke to look for
fixed strings.
After all these years, it's about time we got this straight. Geez,
people.
- Jon
--
Jonathan Dursi | ``..And the world will be better for this;
ljd...@yoho.uwaterloo.ca | that one man, scorned and covered with scars,
| still strove with his last ounce of courage.''
Marco S Hyman
j...@alpha1.csd.uwm.edu (John G Dobnick) writes:
> Perhaps the most important reason, or at least the driving force behind
> short command names, was the TELETYPE itself! You can type at a
> _maximum_ rate of 10 characters per second. Period. You have no-key
10 CPS... you got to use 10 CPS :-) You mean the reason I type
slow today is because of the years spent using a TELETYPE model
28 KSR? 60 wpm, a whopping 45.45 baud -- 6 cps. It spoke baudot
code, a 5 bit code that required two codes (FIGS and LTRS) to
give a reasonable number of characters (assuming you wanted to
use the numbers 0-9 in addition to the letters A-Z).
The model 28 and learning how to type on the old 02x family of
keypunches probably explains why I still think there is only
one shift key on a keyboard.
// marc
Chris Hedley
In article <328cb5ad...@news.msn.fullfeed.com>,
>>a hilarious typo in the Xenix-86 version at a trade show. The actual details
>>were hushed up, and they changed the command to make sure it never happened
>>again.
>>
>
> there was a PC. I'd have to check, but I'd bet it was even in the
> 5th edition, a year before. Or was your explanation intended to be
> flame bait?
>
> the "ed" command sequence: g/<regular expression/p, i.e. Globally
> search for a Regular Expression, and Print.
>
> Sigh, these folks that started on PC's need some real history of
> computing education ;-)
I'm sorry, you're totally wrong. GREP was ported to UNIX from its
original inception on NT. IBM's NT system took the basics for its
GREP command from a file scanning program on VMS called "GROPE", but
they had to change the spelling to the phonetic version (as pronounced
in Ulan-Bator, the site of IBM's head office) of "grep" for copyright
reasons.
The GROPE program, along with the rest of VMS, was written for Sega's
PDP-10 "mainframe" (a rackmounted PC) back in 1988 by Bill Gates and
Alan Turing when they worked together sweeping floors at a Burger King
in Istanbul.
Hope this helps.
Regards,
Chris.
Peter Hesketh
In article <steve.8...@fastnet.prd.co.uk>
st...@fastnet.prd.co.uk "Steve Blinkhorn" writes:
> Next we'll be hearing from people who know what it was like to have no
> more than 100K 36-bit words to do real programming in.
>
> Is there a comp.nostalgia group (INPUT IN CAPS ONLY at 10cps, wearing
> ear defenders)?
Ah! Memories of the KDN2 come rushing back. 4K by 18 bits which controlled
the hot saw which cut up girders and paid for itself every three months for
the whole of its 10 year life. It was installed in a steelworks where they
rolled ingots into H-section rolled-steel-joists. In the old days, a
steelworker measured the resulting length, a clerk looked at current orders
to see how the RSJ could be sliced into sections in the most economical way,
and the hot saw operator cut it up. When the KDN2 arrived, with its
length-measuring photocells, it increased yield from 80% to 95% overnight.
That was a REAL computer. Hardware divide in 25 milliseconds.
But not as old as the machine which printed my first pay-slip. That was
an English Electric Deuce, which used valve logic (translated as toob logic
for those separated by a common language)
I think it was the Deuce which helped originate the saying, "Ouch! I'm glad
that pin was binary 0. A binary 1 would have killed me."
--
Peter Hesketh from Mynyddbach, Monmouthshire, UK
Mike Czaplinski
Perry Hutchison <perry_h...@ccm.hf.intel.com> wrote:
>
> Tim Hollebeek wrote:
> > Bah. Doesn't anyone know 'ex' (the non-visual version of vi) any more?
>
> and humbug. Even the curmudgeons seem to have forgotten ed(1), the
> precursor of ex. ed was the only editor supplied with 6th Edition.
Speaking as a 'young whipper snapper' who's only been working with
Unix for a little over two years, ed(1) is a regular part of my
arsenal, since my job is to resurrect sick machines, and often it's
the only text editor I have available.
I don't ENJOY using it, but I'm somewhere between functionally
illiterate and fluent in it.
> Not speaking for Intel.
Tell you what, I'll speak for Intel, and you speak for NCR....
Mike "NAW!!!" Czaplinski
mike.cz...@washingtondc.ncr.com
Mike Czaplinski
pe...@nmti.com (Peter da Silva) wrote:
>
> In article <E0tAts.BA...@torfree.net>,
> William J. Hayes <ab...@torfree.net> wrote:
> > grep:
> > "Generic Regular Expression Printer"
>
> IBM mainframe program that searched for text in 80 column card decks and
> edited them into new decks. When it was ported to UNIX (using the "struct"
> tool to convert the original Fortran to Ratfor, then to C) they couldn't
> fit the whole functionality in so they only provided the "print" edit.
>
> So it was called "grpep", then shortened to "grep" after Bill Gates made
> a hilarious typo in the Xenix-86 version at a trade show. The actual details
> were hushed up, and they changed the command to make sure it never happened
> again.
>
> "grep" than "grpep" on an IBM card punch, but this was an obvious cover-up,
> since the "grpep" version was never used on an IBM.
>
> Ironically the full functionality of "grpe" was ported to micros for the
> first time on OS/2 in an implementation written in REXX (named after a dog
> that barked when his owner sat down at a computer).
> --
> </peter>
Hmm....good form.....earnest delivery...
Judges?
FWIW: I heard that some renegade programmer originally called it
'gripe', after he was denied some stock during MS's IPO......
Mike "And he had a hook on his foot...." Czaplinski
mike.cz...@washingtondc.ncr.com
Will Rose
Jay R. Jaeger (Jay.J...@msn.fullfeed.com) wrote:
: st...@fastnet.prd.co.uk (Steve Blinkhorn) wrote:
: <snip>
: > someone who knows how to pronounce daemon
: >properly.
: >
: <snip>
: According to whom? The dictionary I ahve lists the pronunciation as
: dEmen (long E). However, I know lots of would be Unix hacks who, for
: some reason, think it should be pronounced dAmen (long A). Or is this
: one of those British English vs. US English debates?
Daemons and demons differ. Unix uses daemons (the ae is a dipthong)
but through time and casual usage I think most people now refer to
them as demons. Pity, since the earlier usage makes more sense.
Jay R. Jaeger
c...@bankersnet.co.uk (Chris Hedley) wrote:
>
>The GROPE program, along with the rest of VMS, was written for Sega's
>PDP-10 "mainframe" (a rackmounted PC) back in 1988 by Bill Gates and
>Alan Turing when they worked together sweeping floors at a Burger King
>in Istanbul.
>
Allright, allright, so you all p*ss on me for being just unbusy enough
to read new messages, and for having an ISP with a slow news setup
that gets messages out of order, and missed the joke. Hope y'all
enjoyed it.
John Everett
In article <56jl24$h...@ordeal.cts.com>, c...@cts.com says...
As I've posted here before, DAEMON was pronounced deemon (not daymon) when it
was implemented on the PDP-10; back when UNIX was just a gleam in Ken
Thompson's eye.
--
jeve...@wwa.com (John V. Everett)
Brian Moffet
In article <328bb47a...@news.msn.fullfeed.com>,
Jay.J...@msn.fullfeed.com says...
cording to whom? The dictionary I ahve lists the pronunciation as
>dEmen (long E). However, I know lots of would be Unix hacks who, for
>some reason, think it should be pronounced dAmen (long A). Or is this
>one of those British English vs. US English debates?
dAmon, and I'm not british, been a Unix "hack" for about 15 years. :-)
brian
Jay R. Jaeger
Tom I Helbekkmo <t...@hamartun.priv.no> wrote:
>[Perry Hutchison]
>
>
>There's no sed in the V6 distribution I've got running here, and it's
>supposed to be a complete V6 from Bell.
Doesn't appear in my 6th Edition man pages (1975) or my "Unix Summary"
either.
Ken Bibb
>In article <56jl24$h...@ordeal.cts.com>, c...@cts.com says...
>>
>>Jay R. Jaeger (Jay.J...@msn.fullfeed.com) wrote:
>>: st...@fastnet.prd.co.uk (Steve Blinkhorn) wrote:
>>
>>: <snip>
>>: > someone who knows how to pronounce daemon
>>: >properly.
>>: >
>>: <snip>
>>
>>: According to whom? The dictionary I ahve lists the pronunciation as
>>: dEmen (long E). However, I know lots of would be Unix hacks who, for
>>: some reason, think it should be pronounced dAmen (long A). Or is this
>>: one of those British English vs. US English debates?
>>
>>Daemons and demons differ. Unix uses daemons (the ae is a dipthong)
>>but through time and casual usage I think most people now refer to
>>them as demons. Pity, since the earlier usage makes more sense.
>As I've posted here before, DAEMON was pronounced deemon (not daymon) when it
>was implemented on the PDP-10; back when UNIX was just a gleam in Ken
>Thompson's eye.
But daemon is derived from the greek "daimon" (benevolent spirit).
Hence the correct (greek) pronounciation has the dipthong.
--
Ken Bibb "If the boundary breaks I'm no longer alone
kb...@arastar.com Don't discourage me
kb...@best.com Bring out the stars/On the first day"
kb...@csd.sgi.com David Sylvian--"The First Day"
Curt Sampson
In article <kbibb.848172436@shellx>, Ken Bibb <kb...@best.com> wrote:
>In <56kli3$3...@kirin.wwa.com> jeve...@wwa.com (John Everett) writes:
>
>>As I've posted here before, DAEMON was pronounced deemon (not daymon) when it
>>was implemented on the PDP-10; back when UNIX was just a gleam in Ken
>>Thompson's eye.
>
>But daemon is derived from the greek "daimon" (benevolent spirit).
>Hence the correct (greek) pronounciation has the dipthong.
In which case it's neither /DEEmon/ nor /DAYmon/, but /DIEmon/, in
ancient Greek, or with a short e, /DEHmon/, in modern. (Actually,
the stress on that last one may be backwards; modern Greek has
merged the pronunciations of a lot of the dipthongs and now uses
stress to determine which word the speaker is saying.) Oh, for an
ISO standard character set for IPA....
cjs
--
Curt Sampson c...@portal.ca Info at http://www.portal.ca/
Internet Portal Services, Inc.
Vancouver, BC (604) 257-9400 De gustibus, aut bene aut nihil.
Curt Sampson
In article <wzzq0kf...@expernet26.expernet.com>,
Peter Ludemann <lude...@expernet26.expernet.com> wrote:
>I seem to remember 3-letter commands for RT-11, but the only one that
>comes to mind now is "PIP" (Peripheral Interchange Program"). On
>PDP-11s, there was a straightforward method of packing 3 characters
>into a 2-byte word, so file names were 6 characters, with a
>3-character extension (I suppose MS-DOS copied the 3-character
>extension, but without the 3-to-2 packing).
Well, MS-DOS copied it though CP/M. CP/M seems to have taken a lot
from RT-11; it even had a PIP program.
Christoph Badura
>Speaking as a 'young whipper snapper' who's only been working with
>Unix for a little over two years, ed(1) is a regular part of my
>arsenal, since my job is to resurrect sick machines, and often it's
>the only text editor I have available.
>I don't ENJOY using it, but I'm somewhere between functionally
>illiterate and fluent in it.
You're not supposed to enjoy using it. It isn't meant to be used on these
newfangled glass TTYs. It works much better on real TTYs.
--
Christoph Badura
Now available in print: Lion's Commentary on UNIX 6th Edition, with Source Code
http://www.peer-to-peer.com/
Tim Shoppa
In article <wzzq0kf...@expernet26.expernet.com>,
Peter Ludemann <lude...@expernet26.expernet.com> wrote:
>I seem to remember 3-letter commands for RT-11, but the only one that
>comes to mind now is "PIP" (Peripheral Interchange Program").
Most RT-11 Keyboard monitor commands can be abbreviated to 3
characters, but you are talking about the system utilities if
you are thinking of PIP. Most of the commonly used system
utilities are only 3 characters - here's the list for RT-11 v5.6:
BINCOM (Binary File Comparison Program)
BUP (Backup utility program)
DIR (Directory program)
DUMP (Dump program)
DUP (Device Utility program)
FILEX (File Exchange Program)
FORMAT (Volume formatting program)
LD (Logical Disk Subsetting Utility)
LIBR (Librarian)
LINK (Linker)
MACRO (Macro-11 assembler)
PIP (Peripheral Interchange Program)
RESORC (Resource Utility Program)
SRCCOM (Source Comparison)
As you can see, only 4 of the 14 are 3 characters long.
Note that the above utility programs can be called through Keyboard
monitor commands, which can be abbreviated to 1 or 2 characters
in many cases. The concept of a "keyboard monitor" with
abbreviable commands is completely foreign to Unix - but this
will hardly matter after January 17, 2038!
Tim. (sho...@triumf.ca)
James M. Curran
In <<55vhpf$q...@mail1.wg.waii.com>>,
Mark Whetzel <ma...@zeus.wg.waii.com> wrote:
> Why are system processes called 'daemons'?
"Daemons" are called such because, just like their "real-world"
counterparts, they are little imps, which hide unnoticed and do
something in the background.
Truth,
James
[alt.folklore.computer]
Robert Bonomi
>dEmen (long E). However, I know lots of would be Unix hacks who, for
>some reason, think it should be pronounced dAmen (long A). Or is this
>one of those British English vs. US English debates?
Neither. In the _context- of "Operating systems design" and utilities
discussions, it is traditionally prounounced "day-mon". Specifically
do distinguish it from 'demon'. This is -not- unique to UNIX, and can
be found in several O/S's that predate it -- e.g. TOPS (-10 or -20), MULTICS,
just to name a couple..
Steve Blinkhorn
In <E0v6z...@cwi.nl> d...@cwi.nl (Dik T. Winter) writes:
>In article <steve.8...@fastnet.prd.co.uk> st...@fastnet.prd.co.uk (Steve Blinkhorn) writes:
> > Next we'll be hearing from people who know what it was like to have no
> > more than 100K 36-bit words to do real programming in.
>Does 32K 27-bit words count?
Oh, all right, I suppose so, but you'll have to put up with me being
boring about writing natural language generation software in 56Kbytes
on an Apple][ that still works 15 years on on a Unix box.
--
Steve Blinkhorn <st...@prd.co.uk>
John Everett
In article <56n415$5...@news.ece.nwu.edu>, bon...@eecs.nwu.edu says...
What a load of rubbish! As a former member of the PDP-10 Monitor (later
called TOPS-10) development group, and author of significant parts of DAEMON,
let me again state categorically that we pronounced it "deemon". Again let me
state that this use predates UNIX by several years. When those of us
in the monitor group attended DECUS conferences we invariably discovered that
some customer sites had taken to calling it "daymon", an error we took pains
to correct.
While it may be true that in the _context- of "Operating systems design" and
utilities discussions, it is traditionally prounounced(sic) "day-mon"; those
of us actually engaged in operating systems and utilities design and
implementation in the late 60s and early 70s pronounced it "deemon".
John Hobson
Steve Blinkhorn wrote:
properly.
>
> Next we'll be hearing from people who know what it was like to have no
> more than 100K 36-bit words to do real programming in.
How about writing FORTRAN on an IBM 1620 with 4000 12-bit words
of memory? Been there, done that. (As a line from one of my
favourite Monty Python skits goes, "Oh, we used to dream of
living in a corridor. Would have been a palace to us.")
--
John Hobson |Short dayes, sharp dayes,
Unix Support Group |Long nights come on apace.
ComEd, Chicago, IL, USA |Ah, who will hide us from
jho...@ceco.ceco.com |The winter's face? -- Thomas Nashe
Shez
sth...@nethelp.no (Steinar Haug) writes:
> [Dik T. Winter]
> | > Next we'll be hearing from people who know what it was like to have no
> | > more than 100K 36-bit words to do real programming in.
> |
>
Oh come on! That's more than I have even now! (64K 8-bit words on CP/M+)
Sheesh, not all of us can afford this week's model you know.
-Shez.
Micah James Cooper
Ken Bibb (kb...@best.com) wrote:
>
> But daemon is derived from the greek "daimon" (benevolent spirit).
> Hence the correct (greek) pronounciation has the dipthong.
>
Daimon in Greek would be pronounced dyemon, actually.
--Micah
---------------------------------------------------------------------------
Micah James Cooper If a liar and deceiver comes and says, 'I will
Coop...@MUOhio.Edu prophesy for you plenty of wine and beer,' he would
be just the prophet for this people! Micah 2:11 (NIV)
Dik T. Winter
In article <56ga2t$8...@cnn.Princeton.EDU> ad...@yuma.Princeton.EDU (Adam J. Thornton) writes:
> In article <E0vDI...@cwi.nl>, Dik T. Winter <d...@cwi.nl> wrote:
> > > | Does 32K 27-bit words count?
> >as main (non-administrative) computer for two universities and one (or
> >perhaps more) research institute.
>
> If you mean the machine I think you do, it also supported an incredibly
> elegant hierarchical batch operating system, whose algorithms were provably
> correct, and which is at the core of one of the finest CS papers ever
> written.
Might well be. It was the Electrologica X8, one of the best early designs
I have seen.
--
dik t. winter, cwi, kruislaan 413, 1098 sj amsterdam, nederland, +31205924131
home: bovenover 215, 1025 jn amsterdam, nederland; http://www.cwi.nl/~dik/
J Wunsch
ma...@troma.rv.tis.com (Mark Sienkiewicz) wrote:
> You were also dealing with a mindset at the time that names had to
> be short. There were lots of things like the 6 character variable
> names in FORTRAN, the 2 and 3 character mnemonics for machine
> instructions, etc etc that made people tend to shorten names
> automatically. DEC even had a character set encoding that
> let you squeeze 3 characters into 2 bytes (thus the 3 characters
> extension in file names).
Radix-50, meaning you could encode octal 50 == decimal 40 different
characters in one digit: 26 letters, 10 decimal digits, dollar, dot,
underscore and space, IIRC. I think two thirds of the RSX-11 names
were encoded R50... That's also why RSX-11 command names were three
letters.
--
J"org Wunsch Unix support engineer
joerg_...@interface-business.de http://www.interface-business.de/~j