8 Nov
2017
8 Nov
'17
6:54 p.m.
Hi,
In looking around the system v7 environment, I don't see a more command anywhere. I
downloaded, converted, and attached 1bsd, 2bsd, and finally 3bsd and it was there that I
found source for more... 3bsd looks like it's for VAX, not PDP-11, and it
doesn't want to compile (looking for some externs that I gather are part of the
distro's clib).
I may jump ship on V7 and head over to 2.9BSD, which, as I understand it, is a V7 with
fixes and these kind of additional tools...
In the meantime, how did folks page through text like man sh and such before more? I know
how to view sections of text using sed and ed's ok for paging file text (painful, but
workable). I just can't seem to locate the idiomatic way of keeping everything from
constantly scrolling out of view! Obviously, this isn't a problem on my mac as
terminal works fine, but I like to try to stay in character as a 1970 time traveling unix
user :).
Thanks,
Will
Sent from my iPhone
8 Nov
8 Nov
6:59 p.m.
more(1ucb) was written by Eric Shienbrood, who modeled from the
functionality from the ITS terminal function he was used to at MIT when he
became a grad student around the same time as I did.
As for before more(1ucb), we used ^S/^Q and remember many of us were on
30/120 cps dialup terminals. So it was not a problem until direct lines @
9600 baud became available (plus having enough power in the processor to
drive them).
Clem
On Wed, Nov 8, 2017 at 6:54 PM, Will Senn <will.senn(a)gmail.com> wrote:
Hi,
In looking around the system v7 environment, I don't see a more command
anywhere. I downloaded, converted, and attached 1bsd, 2bsd, and finally
3bsd and it was there that I found source for more... 3bsd looks like it's
for VAX, not PDP-11, and it doesn't want to compile (looking for some
externs that I gather are part of the distro's clib).
I may jump ship on V7 and head over to 2.9BSD, which, as I understand it,
is a V7 with fixes and these kind of additional tools...
In the meantime, how did folks page through text like man sh and such
before more? I know how to view sections of text using sed and ed's ok for
paging file text (painful, but workable). I just can't seem to locate the
idiomatic way of keeping everything from constantly scrolling out of view!
Obviously, this isn't a problem on my mac as terminal works fine, but I
like to try to stay in character as a 1970 time traveling unix user :).
Thanks,
Will
Sent from my iPhone
7:06 p.m.
^S (stop) and ^Q (continue) at 1200 baud or less worked OK, especially
on paper terminals ;) -- (Clem just beat me to it)
head and tail would work too (although I'm not sure they existed at that
time, I'm no expert):
cat filename | head -20 # get the first page.
cat filename | head -40 | tail -20 # get the second page.
<ad infinitum>
etc.
On 11/8/2017 1:54 PM, Will Senn wrote:
Hi,
In looking around the system v7 environment, I don't see a more command anywhere. I
downloaded, converted, and attached 1bsd, 2bsd, and finally 3bsd and it was there that I
found source for more... 3bsd looks like it's for VAX, not PDP-11, and it
doesn't want to compile (looking for some externs that I gather are part of the
distro's clib).
I may jump ship on V7 and head over to 2.9BSD, which, as I understand it, is a V7 with
fixes and these kind of additional tools...
In the meantime, how did folks page through text like man sh and such before more? I know
how to view sections of text using sed and ed's ok for paging file text (painful, but
workable). I just can't seem to locate the idiomatic way of keeping everything from
constantly scrolling out of view! Obviously, this isn't a problem on my mac as
terminal works fine, but I like to try to stay in character as a 1970 time traveling unix
user :).
Thanks,
Will
Sent from my iPhone
7:31 p.m.
The first and second BSD has cr3 "This filter simulates chuck haleys
cr3, stopping output after each page (22 lines) to wait for a carriage
return, sending 22 more lines, or a EOF, sending 10 more lines."
From my Berkeley book in progress:
In the Berkeley terminal rooms, the dumb terminals beeped
incessantly so most of the bell speakers had been disconnected.
Their \emph{cr3} pager tool rang the terminal bell and waited for a
carriage
return after every 24 lines.
The terminals also rang the bell when the cursor advanced near the
right margin on output or keyboard input (like a typewriter
bell).\cite{halbert-jchac1-4}
% TODO: mention cr3 stty mode?
% archives/1970s/2bsd/src/cr3.c says 22 lines and no mention of sound
% what about that?
So Dan Halbert\index{Halbert, Dan}.
who arrived in 1978 as a first-year graduate student,
wrote a pager called \emph{more} that printed ``--More--''
instead of ringing the bell and accepted the space
instead of carriage return to continue. Plus it could take multiple
filenames and print a line of colons around the filenames.
This was inspired by his use of the ITS timesharing systems as an
undergraduate at MIT that put a ``--MORE--'' prompt at the bottom
of the screen when displaying files.\cite{halbert-jchac1-4}
His friends and fellow graduate students, Geoff
Peck\index{Peck, Geoff} and Eric Shienbrood\index{Shienbrood, Eric},
greatly expanded it, adding various
options -- and \emph{more} was added into the next
distribution.\cite{halbert1}
Date: 17 Jun 2010
@MISC{halbert1,
author = {Dan Halbert},
howpublished = "Personal correspondence",
year = 2010,
month = jun
}
@ARTICLE{halbert-jchac1-4,
author = "Dan Halbert",
title = "{THE "MORE" COMMAND IN UNIX}",
journal = "Journal of the Computer History Association of California",
year = 1994,
month = "April-June",
volume = "1",
number = "4" }
10:43 p.m.
On Wed, 8 Nov 2017, Arthur Krewat wrote:
^S (stop) and ^Q (continue) at 1200 baud or less
worked OK, especially
on paper terminals ;) -- (Clem just beat me to it)
I had to implement XON/XOFF in our Edition 6.
head and tail would work too (although I'm not
sure they existed at that
time, I'm no expert):
When did "pg" come about?
cat filename | head -20 # get the first page.
cat filename | head -40 | tail -20 # get the second page.
Arrgghh!! Unnecessary use of a cat pipe!
--
Dave Horsfall DTM (VK2KFU) "Those who don't understand security will
suffer."
10:47 p.m.
On 11/8/2017 5:43 PM, Dave Horsfall wrote:
cat added for clarity. Ok, not. I'm just dumb.
head -20 < filename
head -40 < filename | tail -20
Better? ;)
cat filename | head -20 # get the first page.
cat filename | head -40 | tail -20 # get the second page.
Arrgghh!! Unnecessary use of a cat pipe!
9 Nov
9 Nov
4:38 a.m.
On 11/8/17 5:21 PM, Dave Horsfall wrote:
On Wed, 8 Nov 2017, Arthur Krewat wrote:
too bad head doesn't exist on v7. I did have to look up the incantation
in Intro to the Shell by Bourne, that's fascinating stuff and it still
works on my mac - nuts!
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
head -20 < filename
Or if you really want to confuse the newbies:
< filename head -20
1:18 p.m.
Head is easy to emulate: sed 10q
On Nov 9, 2017 12:18 AM, "Will Senn" <will.senn(a)gmail.com> wrote:
On 11/8/17 5:21 PM, Dave Horsfall wrote:
On Wed, 8 Nov 2017, Arthur Krewat wrote:
head -20 < filename
Or if you really want to confuse the newbies:
< filename head -20
too bad head doesn't exist on v7. I did have to look up the incantation
in
Intro to the Shell by Bourne, that's fascinating stuff and it still
works on my mac - nuts!
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
2:29 p.m.
It's amazing how often I think - I need to spend more time on sed. I am
also just now beginning to understand that it's a stream ed-itor and
therefore intimately related to ed, which isn't a half editor.
Will
On 11/9/17 7:18 AM, Dan Cross wrote:
Head is easy to emulate: sed 10q
On Nov 9, 2017 12:18 AM, "Will Senn" <will.senn(a)gmail.com
<mailto:will.senn@gmail.com>> wrote:
On 11/8/17 5:21 PM, Dave Horsfall wrote:
On Wed, 8 Nov 2017, Arthur Krewat wrote:
head -20 < filename
Or if you really want to confuse the newbies:
< filename head -20
too bad head doesn't exist on v7. I did have to look up the
incantation in Intro to the Shell by Bourne, that's fascinating
stuff and it still works on my mac - nuts!
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
8:53 p.m.
On Wed, 8 Nov 2017, Will Senn wrote:
If you're interviewing a new hire, see how long it takes 'em to work out
what this does:
:; </etc/motd> /dev/tty sed 1q ;: echo "Hello, world"
Tested on all shells at my disposal; might even work on V7. Did I mention
that I use Perl, and used to use APL\360? The only reason I don't write
obfuscated code is that I might be the one who has to maintain it later...
--
Dave Horsfall DTM (VK2KFU) "Those who don't understand security will
suffer."
<
filename head -20
too bad head doesn't exist on v7. I did have to look up the incantation
in Intro to the Shell by Bourne, that's fascinating stuff and it still
works on my mac - nuts!
10 Nov
10 Nov
12:15 a.m.
On 11/9/2017 3:53 PM, Dave Horsfall wrote:
If you're interviewing a new hire, see how long
it takes 'em to work
out what this does:
:; </etc/motd> /dev/tty sed 1q ;: echo "Hello, world"
For me, there are other criteria for hiring than telling me what some
obscure string of stuff will do.
It's more important that they can take the above or something similar
and shorten/simplify it to:
head -1 /etc/motd
or
sed 1q < /etc/motd
Because I just ran the above on a few different platforms, and they all
just give the first line of /etc/motd, which I expected.
I might be missing something :P - but having them solve a problem the
shortest way possible is more in my interests.
art k.
PS: I've been in interviews where I was given something like that. A
notable one was NYNEX Science and Technology. One of the sysadmins
interviewing me asked about some obscure command or maybe an option, in
SunOS. My answer was "well, I don't remember that, having already worked
with SunOS for a few years, but see that row of manuals behind you? I
bet I can find it in there quick enough". I did, within a minute or two.
I got the consulting gig.
--
Again, no disrespect intended.
2:14 a.m.
Arthur wrote:
head -1 /etc/motd
or
sed 1q < /etc/motd
It's good conversation has moved onto sed given head was presumably
created by someone that saw tail existed and didn't know their sed well
enough, but the `1q' that keeps appearing can be just `q'. :-)
IIRC Brian Cantrell complains in a video interview that head didn't
exist, even though tail did, and it took "them" to create it, presumably
the BSD or Sun lot? (He's also one of the few interviewees that talk
fast enough for a video to be bearable compared to reading at one's own
pace.)
I use `sed 3q', etc., so often that I've ~/bin/3q for various sizes just
to save that typo of seq for sed; handy for checking the output of a
pipeline so far when a screenful would push off useful history.
--
Cheers, Ralph.
https://plus.google.com/+RalphCorderoy
8 Nov
8 Nov
7:12 p.m.
On Nov 8, 2017, at 10:54 AM, Will Senn
<will.senn(a)gmail.com> wrote:
Hi,
In looking around the system v7 environment, I don't see a more command anywhere. I
downloaded, converted, and attached 1bsd, 2bsd, and finally 3bsd and it was there that I
found source for more... 3bsd looks like it's for VAX, not PDP-11, and it
doesn't want to compile (looking for some externs that I gather are part of the
distro's clib).
I may jump ship on V7 and head over to 2.9BSD, which, as I understand it, is a V7 with
fixes and these kind of additional tools...
In the meantime, how did folks page through text like man sh and such before more? I know
how to view sections of text using sed and ed's ok for paging file text (painful, but
workable). I just can't seem to locate the idiomatic way of keeping everything from
constantly scrolling out of view! Obviously, this isn't a problem on my mac as
terminal works fine, but I like to try to stay in character as a 1970 time traveling unix
user :).
Thanks,
Will
Sent from my iPhone
Terminals were slow enough that you could hit ^s to stop scrolling
and ^q to continue. At Fortune Systems, when Dave Yost speeded
up serial io to run 9600 baud terminals at full blast, scrolling was
too fast. So then he added “page mode” to the tty line discipline,
where output stopped after N lines until you hit a key. This was nice
as you didn't have to rerun a command that spewed more than a
screenful but could interfere with type ahead.
7:22 p.m.
On Wed, Nov 8, 2017, at 13:54, Will Senn wrote:
In the meantime, how did folks page through text like
man sh and such
before more? I know how to view sections of text using sed and ed's ok
for paging file text (painful, but workable). I just can't seem to locate
the idiomatic way of keeping everything from constantly scrolling out of
view! Obviously, this isn't a problem on my mac as terminal works fine,
but I like to try to stay in character as a 1970 time traveling unix user
:).
Well, in the beginning, there was paper. You didn't need to "page"
because you had actual pages you could hold in your hand (if not the
manual typeset as an actual book, then the printed output from the
teletype). The utility of a 'pager' as such came from the proliferation
of CRT terminals. To that extent, using the scroll bar in your terminal
to imagine you're going through a roll (well, it was actually fanfold
AIUI) of paper may actually be more "in character" than installing a
pager.
The earliest (mainline BSD, anyway, SRI-NOSC "pg" is dated earlier)
pager in the archives is "cr3" in 1BSD - it alludes to the existence of
an earlier kernel hack with the same effect (enabled, according to
legend, by setting the CR delay in the tty line discipline to mode 3
i.e. "stty cr3").
There are also four different programs in the archive called "pg", all
for the same purpose, that appear otherwise unrelated to each other. The
OpenSolaris one comes from some version of System V, which is the one
that the one available on modern-day Linux is a clone of - the others
are of far earlier vintage. Other programs include "more" (naturally)
from 3BSD, and "p" (which is very minimalistic, despite using a few
V8-specific library features, but V8 isn't in the web-accessible source
archive) from Version 8 research unix.
Some of these are for V7 or V6 contemporary PDP-11 Unixes, and most of
them are far simpler than more so you should have little difficulty
getting them working anyway.
Links:
http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2BSD/src/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/src/ucb/more.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=32V/usr/src/cmd/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=AUSAM/source/S/pg.c
(apparently unrelated to the 32V version)
http://minnie.tuhs.org/cgi-bin/utree.pl?file=SRI-NOSC/s2/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=OpenSolaris_b135/cmd/pg/pg.c
https://danhalbert.org/more.html - mentions cr3
9:18 p.m.
On 11/8/17 1:22 PM, Random832 wrote:
Some of these are for V7 or V6 contemporary PDP-11
Unixes, and most of
them are far simpler than more so you should have little difficulty
getting them working anyway.
Links:
http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2BSD/src/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/src/ucb/more.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=32V/usr/src/cmd/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=AUSAM/source/S/pg.c
(apparently unrelated to the 32V version)
http://minnie.tuhs.org/cgi-bin/utree.pl?file=SRI-NOSC/s2/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=OpenSolaris_b135/cmd/pg/pg.c
https://danhalbert.org/more.html - mentions cr3
It makes sense that it used to be paper based and slow on screen.
Regarding the links and old bsd's. The binary cr3 on 1bsd works in v7.
Also, the book I'm reading has a c program that does paging. But, I'm
always off down the rabbit hole... I tried to compile the cr3.c source
and I get this error:
# cc cr3.c
Undefined:
_fout
_flush
_getc
_end
My understanding is that cc includes libc by default, so these must not
be in libc. But getc is standard lib, so what am I missing?
Thanks,
Will
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
10 p.m.
fout and flush -- Looks like the V6 version of cr3 - which is using the
portable C library which was done before the standard C library.
I bet if you grab the one from 2BSD that should have been updated to V7
On Wed, Nov 8, 2017 at 9:18 PM, Will Senn <will.senn(a)gmail.com> wrote:
On 11/8/17 1:22 PM, Random832 wrote:
Some of these are for V7 or V6 contemporary
PDP-11 Unixes, and most of
them are far simpler than more so you should have little difficulty
getting them working anyway.
Links:
http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2BSD/src/cr3.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/src/ucb/more.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=32V/usr/src/cmd/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=AUSAM/source/S/pg.c
(apparently unrelated to the 32V version)
http://minnie.tuhs.org/cgi-bin/utree.pl?file=SRI-NOSC/s2/pg.c
http://minnie.tuhs.org/cgi-bin/utree.pl?file=OpenSolaris_b135/cmd/pg/pg.c
https://danhalbert.org/more.html - mentions cr3
It makes sense that it used to be paper based and slow on screen.
Regarding the links and old bsd's. The binary cr3 on 1bsd works in v7.
Also, the book I'm reading has a c program that does paging. But, I'm
always off down the rabbit hole... I tried to compile the cr3.c source and
I get this error:
# cc cr3.c
Undefined:
_fout
_flush
_getc
_end
My understanding is that cc includes libc by default, so these must not be
in libc. But getc is standard lib, so what am I missing?
Thanks,
Will
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
10:18 p.m.
Woohoo! I actually got it working :). Thanks for the assist. The 2bsd
version includes a retrofit.h file for it to work on v6. I removed the
include and it compiled and linked just fine.
Will
On 11/8/17 4:00 PM, Clem Cole wrote:
fout and flush -- Looks like the V6 version of cr3 -
which is using
the portable C library which was done before the standard C library.
I bet if you grab the one from 2BSD that should have been updated to V7
On Wed, Nov 8, 2017 at 9:18 PM, Will Senn <will.senn(a)gmail.com
<mailto:will.senn@gmail.com>> wrote:
On 11/8/17 1:22 PM, Random832 wrote:
Some of these are for V7 or V6 contemporary PDP-11 Unixes, and
most of
them are far simpler than more so you should have little
difficulty
getting them working anyway.
Links:
http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/cr3.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/cr3.c>
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2BSD/src/cr3.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=2BSD/src/cr3.c>
http://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/src/ucb/more.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/src/ucb/more.c>
http://minnie.tuhs.org/cgi-bin/utree.pl?file=32V/usr/src/cmd/pg.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=32V/usr/src/cmd/pg.c>
http://minnie.tuhs.org/cgi-bin/utree.pl?file=AUSAM/source/S/pg.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=AUSAM/source/S/pg.c>
(apparently unrelated to the 32V version)
http://minnie.tuhs.org/cgi-bin/utree.pl?file=SRI-NOSC/s2/pg.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=SRI-NOSC/s2/pg.c>
http://minnie.tuhs.org/cgi-bin/utree.pl?file=OpenSolaris_b135/cmd/pg/pg.c
<http://minnie.tuhs.org/cgi-bin/utree.pl?file=OpenSolaris_b135/cmd/pg/pg.c>
https://danhalbert.org/more.html
<https://danhalbert.org/more.html> - mentions cr3
It makes sense that it used to be paper based and slow on screen.
Regarding the links and old bsd's. The binary cr3 on 1bsd works in
v7. Also, the book I'm reading has a c program that does paging.
But, I'm always off down the rabbit hole... I tried to compile the
cr3.c source and I get this error:
# cc cr3.c
Undefined:
_fout
_flush
_getc
_end
My understanding is that cc includes libc by default, so these
must not be in libc. But getc is standard lib, so what am I missing?
Thanks,
Will
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
--
GPG Fingerprint: 68F4 B3BD 1730 555A 4462 7D45 3EAA 5B6D A982 BAAF
9 Nov
9 Nov
2:01 a.m.
"Well, in the beginning, there was paper. You didn't need to "page"
because you had actual pages you could hold in your hand (if not the
manual typeset as an actual book, then the printed output from the
teletype)."
Paper indeed... I had a model 37 teletype at home, and would
bring home boxes of paper and carefully arrange the box so that the
blank paper feeding the terminal did not get tangled up with the
"used" paper spewing out of the teletype. A second box for the used
paper was handy.
In the early Unix days, the file system had two kinds of tiles --
small files and "large" files. As a file grew, there was a moment
when the inodes were rewritten to go from small to large. And if the
system crashed during this process, a lot of stuff could get messed up
or deleted. It usually took a few minutes for the system to get back
online, so I kept a highlighter by the terminal, and when the system
crashed I would reel back the used paper and highlight those lines
that I needed to reenter. When the system came back, I copied the
lines that had been lost and pressed on.
One day, I had spent a couple of hours in an intense debugging session
when my wife called me to lunch. I didn't save my stuff, and was
very unhappy to return an hour later and discover the system had
crashed. I was even more unhappy to discover that, while I was at
lunch, our cat had found the box of used paper and used it as a litter
box. After the shock wore off, I picked up my highlighter anyway and
marked the soggy lines I needed to retype...
It was never more true that debugging was a smelly job...
Steve
2:16 a.m.
On Nov 8, 2017, at 6:01 PM, Steve Johnson
<scj(a)yaccman.com> wrote:
I was even more unhappy to discover that, while I was at lunch, our cat had found the
box of used paper and used it as a litter box. After the shock wore off, I picked up my
highlighter anyway and marked the soggy lines I needed to retype...
Reminds me of a few lines in one of my favourite _Arrogant Worms_ songs (okay, they all
are):
.
.
.
William Shakespeare's in my cat
He rarely ever talks
He makes his loudest statements
Standing in the litter box
.
.
.
Perhaps s/he was doing code review :-)
https://arrogantworms1.bandcamp.com/track/william-shakespeares-in-my-cat
8 Nov
8 Nov
7:29 p.m.
On 11/8/2017 10:54 AM, Will Senn wrote:
Hi,
In the meantime, how did folks page through text like
man sh and such
before more? I know how to view sections of text using sed and ed's ok for
paging file text (painful, but workable). I just can't seem to locate the
idiomatic way of keeping everything from constantly scrolling out of view!
Obviously, this isn't a problem on my mac as terminal works fine, but I
like to try to stay in character as a 1970 time traveling unix user :).
In the early days of Unix I was told that it wasn't practical to write
a pager because such a thing would have to run in raw mode in order
to process single letter commands, such as the space character for
going on to the next page. Since raw mode introduced a significant amount
of overhead on already overtaxed machines, it was considered an anti-social
thing to do.
In 1977 and 1978 I worked at Ford Aerospace in the group that produced
KSOS, which was a "secure" version of Unix (I didn't actually work on
KSOS myself since I didn't have, nor want, a security clearance). Anyway,
that group used some incredibly expensive HP terminals that contained
enough local memory to contain most text files, the way we overcame the
lack of a pager was to cat the file, and then page around in the local
memory.
IIRC later versions of Unix added the ability to respond to a specific
list of single characters without going into raw mode. Of course, that
didn't help when full-screen editors like vi and the Rand editor came out.
Cordially,
Jon Forrest
7:43 p.m.
I'm not sure why raw mode would be particularly loading on a system unless it is
receiving a lot of data. In the situation described, it's doing nothing other than
waiting on the user. The load should be the same as if it is waiting in cooked mode.
Ah, those old HP terminals. I had one that had many pages of memory and would even
spool off the older stuff on to one of its two data tape drives.
The other ones we had too little memory. They couldn't even buffer 24 x 80
characters, so if you had long lines, the thing would start scrolling even before it hit
the bottom of the screen.
We had some terminals (I think VT100's had this option) that would automatically send
a ^S to the host when the screen filled and then you could release it.
I still remember the day I was working at RU. One of the young female operators says
"hey, someone is catting /usr/dict/words to my terminal." I told her to hit ^S
immediately to freeze the sending process and then ran a PS to find the culprit.
This reminds me of the dichotomy of "write" versions. The early ones did just
rely on being able to open the remote user's terminal. We found this rather
antisocial, so at JHU we always left the terminals protected against world write. We
used the x bit as the "want to receive writes" flag and made write setuid (this
also allowed us to avoid fun and games like people redirecting the input to things that
weren't typewriters).
9:03 p.m.
If I recall correctly there was a program called p on v6 unix that served as a pager. I
think it was part of the standard distribution and not a local Toronto mod.
Sent from my iPhone
On Nov 8, 2017, at 2:29 PM, Forrest, Jon
<nobozo(a)gmail.com> wrote:
On 11/8/2017 10:54 AM, Will Senn wrote:
Hi,
In the meantime, how did folks page through text
like man sh and such
before more? I know how to view sections of text using sed and ed's ok for
paging file text (painful, but workable). I just can't seem to locate the
idiomatic way of keeping everything from constantly scrolling out of view!
Obviously, this isn't a problem on my mac as terminal works fine, but I
like to try to stay in character as a 1970 time traveling unix user :).
In the early days of Unix I was told that it wasn't practical to write
a pager because such a thing would have to run in raw mode in order
to process single letter commands, such as the space character for
going on to the next page. Since raw mode introduced a significant amount
of overhead on already overtaxed machines, it was considered an anti-social
thing to do.
In 1977 and 1978 I worked at Ford Aerospace in the group that produced
KSOS, which was a "secure" version of Unix (I didn't actually work on
KSOS myself since I didn't have, nor want, a security clearance). Anyway,
that group used some incredibly expensive HP terminals that contained
enough local memory to contain most text files, the way we overcame the
lack of a pager was to cat the file, and then page around in the local
memory.
IIRC later versions of Unix added the ability to respond to a specific
list of single characters without going into raw mode. Of course, that
didn't help when full-screen editors like vi and the Rand editor came out.
Cordially,
Jon Forrest
2358
days inactive
2360
days old
23 comments
14 participants
participants (14)
-
Arthur Krewat -
Bakul Shah -
Clem Cole -
Dan Cross -
Dave Horsfall -
Forrest, Jon -
Jeremy C. Reed -
Lyndon Nerenberg -
Mark Green -
Ralph Corderoy -
Random832 -
Ron Natalie -
Steve Johnson -
Will Senn