6 Nov
2020
6 Nov
'20
9:53 p.m.
Moving to COFF.
below.
On Fri, Nov 6, 2020 at 10:40 AM Will Senn <will.senn(a)gmail.com> wrote:
Clem,
It figures. I should have known there was a reason for the shorter lines
other than display. Conventions are sticky and there appears to be a
generation gap. I use single spaces between sentences, but my ancestors
used 2... who knows why? :).
You never use a real typewriter. Double-space allows you to edit
(physically) the document if need be. This was how I did everything
before I had easy computer access.
I went to college with an electric typewriter and all my papers were done
on it in the fall of my freshman year (until I got access to UNIX). I did
have an CS account for the PDP-10 and they had the XGP, but using it for
something like your papers was somewhat frowned upon. However, the UNIX
boxes we often bought 'daisy wheel' typewriters that had RS-232C
interfaces. Using nroff, I could then do my papers and run it off in the
admin's desk at night.
Clem
7 Nov
7 Nov
1:22 a.m.
New subject: [TUHS] The Elements Of Style: UNIX As Literature
On Fri, Nov 06, 2020 at 10:53:59AM -0500, Clem Cole wrote:
I went to college with an electric typewriter and all my papers were done
on it in the fall of my freshman year (until I got access to UNIX). I did
have an CS account for the PDP-10 and they had the XGP, but using it for
something like your papers was somewhat frowned upon. However, the UNIX
boxes we often bought 'daisy wheel' typewriters that had RS-232C
interfaces. Using nroff, I could then do my papers and run it off in the
admin's desk at night.
When I was in high school, we had a box that could be fitted over an
Olivetti electric typewriter's keyboard, which had solenoids to
"type". The other end had a parallel port and it was connected to a
Heathkit H-89 CP/M system, and so rough drafts would be sent to the
dot matrix printer, but for the final copy, it could look like it came
out of a typewriter --- because technically, it did. :-)
- Ted
1:24 a.m.
New subject: [TUHS] The Elements Of Style: UNIX As Literature
Outstanding hack!
On Fri, Nov 6, 2020 at 2:22 PM Theodore Y. Ts'o <tytso(a)mit.edu> wrote:
On Fri, Nov 06, 2020 at 10:53:59AM -0500, Clem Cole
wrote:
I went to college with an electric typewriter and all my papers were done
on it in the fall of my freshman year (until I got access to UNIX). I
did
have an CS account for the PDP-10 and they had
the XGP, but using it for
something like your papers was somewhat frowned upon. However, the
UNIX
boxes we often bought 'daisy wheel'
typewriters that had RS-232C
interfaces. Using nroff, I could then do my papers and run it off in the
admin's desk at night.
When I was in high school, we had a box that could be fitted over an
Olivetti electric typewriter's keyboard, which had solenoids to
"type". The other end had a parallel port and it was connected to a
Heathkit H-89 CP/M system, and so rough drafts would be sent to the
dot matrix printer, but for the final copy, it could look like it came
out of a typewriter --- because technically, it did. :-)
- Ted
4:58 a.m.
New subject: [TUHS] The Elements Of Style: UNIX As Literature
On Friday, 6 November 2020 at 10:53:59 -0500, Clem Cole wrote:
However, the UNIX boxes we often bought 'daisy
wheel' typewriters
that had RS-232C interfaces. Using nroff, I could then do my papers
and run it off in the admin's desk at night.
My memory is hazy, but I thought that the daisy wheel printers I knew
(Qume Sprint\5) also had proportional spacing. I ran into significant
problems with early MS-DOS based formatting software because it made
(frequently incorrect) assumptions about character widths.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
8 Nov
8 Nov
3:04 a.m.
New subject: [TUHS] The Elements Of Style: UNIX As Literature
On Fri, Nov 6, 2020 at 5:58 PM Greg 'groggy' Lehey <grog(a)lemis.com>
wrote:
My memory is hazy, but I thought that the daisy wheel
printers I knew
(Qume Sprint\5) also had proportional spacing.
I never used that brand. Xerox (which was the main USA supplier as a pure
typewriter to compete with IBM's Selectric 'ball' units) were definitely
fixed width. People started to hack the Xerox units to add access to
serial interface and Xerox made it standard or maybe an option as
somepoint. IIRC it was somebody like Ollivetti that originally did the
daisywheel and Xerox licensed it and they definitely were the primary
player here. But by the late 70s, early 80's, there were a number of
manufacturers of them.
My memory was with the maybe circa 74/75 timeframe, Xerox unit (but it
might have been one of the others) was that the original unit had a serial
port for diagnostics/maintenance which allowed access to the on-board
microprocessor (which might have been a 4-bit TI chip IIRC - same used in
some early 'Simon' games). Somebody figured out how to hack it and the
schematics/description was available. I remember we hacked one of the
units in the EE dept. But by the late 1970's the serial interface was a
first class part of the unit, which made them different from IBM Selectrics
which did not have an easy to access serial interface, even though IBM used
the printer mechanism from the Selectric as the guts of the console for the
360 which I think was called a 2150 but the bits in my brain on that are
extremely stale.
I ran into significant
problems with early MS-DOS based formatting software because it made
(frequently incorrect) assumptions about character widths.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
5:05 a.m.
New subject: [TUHS] The Elements Of Style: UNIX As Literature
On Sat, 7 Nov 2020, Clem Cole wrote:
The daisywheel that I used (Fujitsu?) definitely was fixed-width
(Courier); I used it a lot with WordStar on CP/M.
-- Dave
My memory is
hazy, but I thought that the daisy wheel printers I knew
(Qume Sprint\5) also had proportional spacing.
I never used that brand. [...]
9 Nov
9 Nov
10:36 a.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
On Saturday, 7 November 2020 at 16:04:16 -0500, Clem Cole wrote:
On Fri, Nov 6, 2020 at 5:58 PM Greg 'groggy'
Lehey <grog(a)lemis.com> wrote:
The Qume printers seemed to have been the best round 1980 when we used
them in our applications. In particular, a large choice of wheels and
fine-grained spacing. I forget how the spacing worked.
My memory is hazy, but I thought that the daisy
wheel printers I knew
(Qume Sprint\5) also had proportional spacing.
I never used that brand. Xerox (which was the main USA supplier as a pure
typewriter to compete with IBM's Selectric 'ball' units) were definitely
fixed width. People started to hack the Xerox units to add access to
serial interface and Xerox made it standard or maybe an option as
somepoint. IIRC it was somebody like Ollivetti that originally did the
daisywheel and Xerox licensed it and they definitely were the primary
player here. But by the late 70s, early 80's, there were a number of
manufacturers of them. But by the late 1970's the serial interface was a
first class part
of the unit, which made them different from IBM Selectrics which did
not have an easy to access serial interface, even though IBM used
the printer mechanism from the Selectric as the guts of the console
for the 360 which I think was called a 2150 but the bits in my brain
on that are extremely stale.
The golfball console for the /360 was much earlier than that, like the
/360 itself. The model numbers I recall were 735, and the newer
generation 2731/2735. The last digit related to the carriage width
(11"/15").
Round the time in question I bought a second-hand 735 machine. It had
an arcane interface that directly talked to the magnets. I built an
interface for it to a parallel port, but it never worked well. Not
the interface: the 735 was second-hand and basically worn out, and it
kept coming out of adjustment. The Qume machines were *so* much
easier to use.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
8:26 p.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
On Sun, Nov 8, 2020 at 11:36 PM Greg 'groggy' Lehey <grog(a)lemis.com>
wrote:
The golfball console for the /360 was much earlier
than that, like the /360
itself.
Hmmm, I think what I said is correct. The S/360 system was released in
1964. My friend Russ Roebling (360/50 chief designer ) once told me the
console came from the office products (typewriter) division. I wish I
could remember the story he told me, but IIRC it was something WRT to
politics inside of IBM and ensuring the console device and the 360's launch
between the divisions. [Just like every large firm I have worked, I'm not
really surprised to hear that divisional fiefdoms were rampant at IBM in
those days, too].
I'm fairly sure that the Selectric (I) was early1960s (I think 61/62). I
just don't remember the model number of the S/360's console (every device
at IBM had numeric names), your memory is likely that the number was 7xy.
But as I said, I'm fair sure that the guts of the console were based on
the Selectric's design.
10 Nov
10 Nov
6:10 a.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
On Monday, 9 November 2020 at 9:26:05 -0500, Clem Cole wrote:
On Sun, Nov 8, 2020 at 11:36 PM Greg
'groggy' Lehey <grog(a)lemis.com> wrote:
Thanks for the interesting details. Yes, that all matches my
recollection. Originally you were talking about mid- to late 1970s,
and that's what my "much earlier" referred to.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
The golfball console for the /360 was much
earlier than that, like the /360
itself.
Hmmm, I think what I said is correct. The S/360 system was released in
1964. My friend Russ Roebling (360/50 chief designer ) once told me the
console came from the office products (typewriter) division. I wish I
could remember the story he told me, but IIRC it was something WRT to
politics inside of IBM and ensuring the console device and the 360's launch
between the divisions. [Just like every large firm I have worked, I'm not
really surprised to hear that divisional fiefdoms were rampant at IBM in
those days, too].
I'm fairly sure that the Selectric (I) was early1960s (I think 61/62). I
just don't remember the model number of the S/360's console (every device
at IBM had numeric names), your memory is likely that the number was 7xy.
But as I said, I'm fair sure that the guts of the console were based on
the Selectric's design.
8:48 p.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
Fair enough, sorry to be confusing. It is interesting that a piece of IBM
early 60s mechanical design (the electric), lasted as long as it did. I
don't know if there was a Selectric IV, there certainly was a Selectric III
that was sold through the 70s and early 1980s. Wang created what they
called word processing and only then did the Selectrics and Daisy Wheels
start to slowly diminish[1]. By the 80s, when we created Stellar
Computers, all of the admin's had a PC/AT and a copy of Wordperfect and
used our LaserWriters in Engineering, but we still had one Selectric for
times when a typewriter was easier.
Clem
1] A fun side story. One of many sisters is/was a professional concert
harpist (she has incredible manual dexterity). Tough to feed yourself as
a concert harpist, so she got a job at MIT working as Ron's admin. Her
terminal was an ITS connection and so they taught her to edit documents
using EMACS/Tex (she actually typed the RSA papers for Ron so many years
ago using the same). At one point, she was thinking of leaving MIT, and
when she would interview different firms, they usually would ask her if she
knew 'Wang.' It's interesting that her MIT skills were the ones that
lasted. For the last few years, she has worked as a technical editor/book
index creator *etc*.. for a number of research orgs and technical book
publishers -- she can handle EMACS and LaTex of course, not just MS-Word ;-)
On Mon, Nov 9, 2020 at 7:10 PM Greg 'groggy' Lehey <grog(a)lemis.com>
wrote:
On Monday, 9 November 2020 at 9:26:05 -0500, Clem
Cole wrote:
launch
On Sun, Nov 8, 2020 at 11:36 PM Greg
'groggy' Lehey <grog(a)lemis.com>
wrote:
> The golfball console for the /360 was much earlier than that, like the
/360
itself.
Hmmm, I think what I said is correct. The S/360 system was released in
1964. My friend Russ Roebling (360/50 chief designer ) once told me the
console came from the office products (typewriter) division. I wish I
could remember the story he told me, but IIRC it was something WRT to
politics inside of IBM and ensuring the console device and the 360's between the divisions. [Just like every large
firm I have worked, I'm
not
really surprised to hear that divisional fiefdoms
were rampant at IBM in
those days, too].
I'm fairly sure that the Selectric (I) was early1960s (I think 61/62).
I
just don't remember the model number of the
S/360's console (every device
at IBM had numeric names), your memory is likely that the number was 7xy.
But as I said, I'm fair sure that the guts of the console were based on
the Selectric's design.
Thanks for the interesting details. Yes, that all matches my
recollection. Originally you were talking about mid- to late 1970s,
and that's what my "much earlier" referred to.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
9:10 p.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
Speaking of the Selectric
Another old story about printers. Back in 1974 (ish) I was an undergrad at MIT working at
the Architecture Machine Group, which was the predecessor to the Media Lab. We had a
home-grown OS for the Interdata 16-bit minicomputers, whose instruction sets were very
much like 16-bit IBM 360’s.
There was an IBM 2741 there for talking to the institute mainframes, and somehow I got the
job of writing a device driver for it. It was quite an adventure getting the tilt-rotate
codes and so forth to fit in the 160 hex bytes available… I recall having to chain short
branches together if the condition codes were right.
The success of that made me a go-to guy for printing, unfortunately, so later it was my
job to patch the line printer to print capital O after the 0 wore out*.
* A real thing, way before Dilbert got hold of it.
4:08 a.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
On Mon, 9 Nov 2020, Greg 'groggy' Lehey wrote:
The Qume printers seemed to have been the best round
1980 when we used
them in our applications. In particular, a large choice of wheels and
fine-grained spacing. I forget how the spacing worked.
Presumably some sort of a table lookup, based on which character is about
to be hit? Or are you referring to the micro-spacing itself?
The golfball console for the /360 was much earlier
than that, like the
/360 itself. The model numbers I recall were 735, and the newer
generation 2731/2735. The last digit related to the carriage width
(11"/15").
I once had a fine collection of goofballs (as we called them); sadly lost
in a house move :-(
Round the time in question I bought a second-hand 735
machine. It had
an arcane interface that directly talked to the magnets. I built an
interface for it to a parallel port [...]
I'd like to know a bit more about that interface... You'd have to control
the carriage, roller, swivel/tilt/hit etc. How did you detect the BREAK
key to get the 360's attention and unlock the keyboard?
-- Dave
6:48 a.m.
New subject: Daisy wheel printers (was: [TUHS] The Elements Of Style: UNIX As Literature)
On Tuesday, 10 November 2020 at 9:08:44 +1100, Dave Horsfall wrote:
I was going to say "ditto", but I think I actually sold them along
with the 735.
Yes, I'm trying to recall that too. The ball itself was controlled by
6 signals: Up 1, up 2 (for the 4 rows of characters), left 1, left 2,
left 2 (yes, twice) and right 5, for a total of 11 columns. But my
recollection was that I only had about 10 power transistors driving
the thing. I wish I had kept more details. Maybe there's something
amongst the useless junk in the shed.
On Mon, 9 Nov 2020, Greg 'groggy' Lehey
wrote:
"Yes". As I said, I forget. I have a feeling that it must have been
explicit micro-spacing, since the machine didn't know anything about
the kind of daisy wheel that was fitted.
The Qume printers seemed to have been the best
round 1980 when we used
them in our applications. In particular, a large choice of wheels and
fine-grained spacing. I forget how the spacing worked.
Presumably some sort of a table lookup, based on which character is about
to be hit? Or are you referring to the micro-spacing itself? The golfball
console for the /360 was much earlier than that, like the
/360 itself. The model numbers I recall were 735, and the newer
generation 2731/2735. The last digit related to the carriage width
(11"/15").
I once had a fine collection of goofballs (as we called them); sadly lost
in a house move :-( Round the time
in question I bought a second-hand 735 machine. It had
an arcane interface that directly talked to the magnets. I built an
interface for it to a parallel port [...]
I'd like to know a bit more about that interface... You'd have to
control the carriage, roller, swivel/tilt/hit etc. How did you detect the BREAK key to get the 360's
attention and
unlock the keyboard?
I didn't. The 735 doesn't have a BREAK key. It was a typewriter, not
a teletype. I used it as a printer in addition to a normal glass TTY.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
1502
days inactive
1506
days old
12 comments
5 participants
participants (5)
-
Clem Cole -
Dave Horsfall -
Greg 'groggy' Lehey -
Lawrence Stewart -
Theodore Y. Ts'o