What happened to the Teletype Corporation?
hanc...@bbs.cpcn.com
screens, the Teletype Corporation upated its product line accordingly
with its "Dataspeed" terminals.
But in the breakup of the Bell System, old entities like Western
Electric and Teletype disappeared.
I would've thought given the huge demand for terminals, Teletype
would've been well positioned to supply them, as well as high quality
teleprinters to compete with Selectrics for high grade terminal use
(ie banking, airline ticket counters).
Would anyone know what happened to the corporation and why?
krw
hanc...@bbs.cpcn.com says...
--
Keith
Steve Wolstenholme
>Would anyone know what happened to the corporation and why?
http://www.kekatos.com/teletype/ describes the history.
--
Steve Wolstenholme Neural Planner Software Ltd
EasyNN-plus. The easy way to build neural networks.
Frank McCoy
>As the computer terminal world moved from teleprinters to on-line CRT
>screens, the Teletype Corporation upated its product line accordingly
>with its "Dataspeed" terminals.
>
>But in the breakup of the Bell System, old entities like Western
>Electric and Teletype disappeared.
>
>I would've thought given the huge demand for terminals, Teletype
>would've been well positioned to supply them, as well as high quality
>teleprinters to compete with Selectrics for high grade terminal use
>(ie banking, airline ticket counters).
>
There are more "Silent 700" teletypes sitting around ....
I have one out in the garage still.
>Would anyone know what happened to the corporation and why?
--
_____
/ ' / ™
,-/-, __ __. ____ /_
(_/ / (_(_/|_/ / <_/ <_
John L
>I have one out in the garage still.
You might want to go look at it, particularly the name plate where it
says Texas Instruments.
It's a fine old boat anchor, but it's not a Teletype.
Jim Haynes
Basically, a combination of things.
AT&T owned Teletype because it provided terminals for their TWX
service and private line offerings. That was perhaps 50% of
Teletype's business, the other part being government, Western
Union (primarly as a government contractor), railroads, and lately
the minicomputer business. AT&T was persuaded to sell TWX to
Western Union, which eliminated a portion of AT&T's need for the
company. By then the best days of both TWX and Telex were past,
so it was no longer a vigorous business segment.
Teletype's great prowess was in the ability to design and manufacture
intricate mechanical stuff. This asset was severely devalued practically
overnight by progress in microelectronics. Some of Teletype's management
were slow to realize what was happening.
Teletype was historically weak in marketing, since much of that was
done by the Bell System companies. And marketing was difficult for
the Bell System because it operated as a regulated public utility,
had to file tariffs and have them approved before it could offer a new
product or service.
As the industry moved from hardcopy to video terminals, Teletype was
suddenly in an extremely competitive situation. Teletype had few
competitors in the mechanical printer era, but anybody could make
video terminals with off-the-shelf parts. Even with hard copy printers,
electronics took over with dot matrix and daisywheel and later inkjet and
laser designs. Texas Instruments could make the Silent 700 family
because it took advantage of some aspects of semiconductor technology,
such as the ability to make a thermal print head. Teletype patented a
daisywheel back in 1939, but it wasn't practical without microelectronics.
Teletype's most successful video terminals, I think, were in the form of
a system that emulated an IBM terminal cluster and was supported by the
Bell System. And the microcomputer business produced a bunch of
manufacturers of dot matrix hardcopy printers that were cheaper than
anything Teletype had ever made.
Teletype made some costly bad decisions. Model 37 circa 1967 was
a dinosaur. The Inktronic ink jet printer was a clever idea leading
to a bad product and delaying work on a fast impact printer. (It
had essentially nothing in common with today's inkjet printer technology.)
Model 43 was a decent dot-matrix teleprinter but should have been started
years earlier. A lot of money went into learning to produce MOS
integrated circuits in-house, when going to a silicon foundry would
probably have made more sense.
The 1956 Bell System consent decree was rigidly interpreted and kept
Teletype from moving into computer technology. (basically said that
the Bell System would not engage in data processing. Also that Teletype
couldn't sell anything on the open market that had not been developed
for Bell System use, with an exception for government purchases.)
Following the Bell System breakup AT&T didn't seem to know what it was
doing - witness buying NCR, selling NCR, buying in cell phones and
selling out of cell phones, etc. so didn't know what to do with Teletype.
The timeline is that Teletype became AT&T Teletype about 1989, then
part of AT&T Information Systems. The Skokie complex was torn down
and converted to a shopping center, as was A.B. Dick and some other
major companies in the Skokie area. Some of the Teletype engineers
went to Bell Labs in Naperville. The Little Rock plant of Teletype
continued operation for a while, couldn't really compete with the
rest of the industry making PCs and dot-matrix printers for PCs.
AT&T spun off Western Electric in to Lucent, and then Lucent spun off
a couple of its operations. Last I heard the Little Rock building was
still standing but was for sale, and had wound up with Avaya or another
of the Lucent spinoffs. Meanwhile Lucent is now part of Alcatel.
--
jhhaynes at earthlink dot net
hanc...@bbs.cpcn.com
Jim Haynes wrote:
> Teletype's great prowess was in the ability to design and manufacture
> intricate mechanical stuff. This asset was severely devalued practically
> overnight by progress in microelectronics. Some of Teletype's management
> were slow to realize what was happening.
It's a shame that mechanical expertise wasn't somehow exploited in the
new products coming out.
IBM had to play catch up in electronics although it had strong
mechanical background for its tab days. This expertise helped it make
the 1403 printer, a huge hit (per Campbell-Kelly Aspray "Computer").
It's too bad Teletype use its past expertise to do well in the desktop
printer market. Also, from your description it seems Teletype was
isolated from the skill bases at Western Electric and AT&T which might
have helped.
IBM, while playing catchup in electronic computers, had its expertise
in systems-design as a major advantage. I don't know if Teletype had
an equivalent "expertise" in terminal communications to exploit.
Indeed, from a user's point of view, I didn't care for the Teletype
33. I'm nostalgic over it because it's what I used. But I thought
the keyboard was lousy and not conducive to comfortable fast typing.
> Teletype was historically weak in marketing, since much of that was
> done by the Bell System companies. And marketing was difficult for
> the Bell System because it operated as a regulated public utility,
> had to file tariffs and have them approved before it could offer a new
> product or service.
That was a big problem with the whole AT&T breakup--the entire Bell
System was not used to marketing.
> The 1956 Bell System consent decree was rigidly interpreted and kept
> Teletype from moving into computer technology. (basically said that
> the Bell System would not engage in data processing. Also that Teletype
> couldn't sell anything on the open market that had not been developed
> for Bell System use, with an exception for government purchases.)
In the 1990s IBM quietly sought to abolish its consent decree
restrictions and did so. This enabled it to return to business lines
it had to abandon years ago, such as service bureaus and consulting.
A big part of IBM's revenue today comes from those lines.
It's a shame AT&T remained locked into that consent decree when it no
longer had a monopoly in phone service and equipment. After
Carterphone and certainly when MCI got its green light to break in,
AT&T should've been freed from its Consent Decree.
One of the things that bugs me about the Bell System breakup was the
double standard. The newcomers were free to do whatever was good for
them (under cut rates, new products) but they fought hard to keep the
Bell System locked in place.
> The timeline is that Teletype became AT&T Teletype about 1989, then
> part of AT&T Information Systems. The Skokie complex was torn down
> and converted to a shopping center, as was A.B. Dick and some other
> major companies in the Skokie area.
I think AB Dick is another company that went through tough
technological changes. I think they're offset printing business is
still around, but their mimeo and spirit-ditto duplicator, once big
lines, are gone, replaced by high speed cheap xeroxgraphy copiers.
Many offices limited xerox-type copying machines to 10 copies,
anything more had to be reproduced by other means. All schools had
mimeo and spirit duplicators. Frankly, I don't want to go back to
mimeograph, I have enough ink stained shirts from those days.
Another reproducer, Gestetner, is around, advertising document
handling products. Around 1983 they were still pushing an automatic
sealed printing system but using mimeo technology. Mimeo is cheap but
I guess today it is obsolete.
Thanks for the info!
Ted Nolan <tednolan>
<hanc...@bbs.cpcn.com> wrote:
>
>
>I think AB Dick is another company that went through tough
>technological changes. I think they're offset printing business is
>still around, but their mimeo and spirit-ditto duplicator, once big
>lines, are gone, replaced by high speed cheap xeroxgraphy copiers.
>Many offices limited xerox-type copying machines to 10 copies,
>anything more had to be reproduced by other means. All schools had
>mimeo and spirit duplicators. Frankly, I don't want to go back to
>mimeograph, I have enough ink stained shirts from those days.
>
I think they could make money with a spirit-ditto perfume..
Ted
jmfb...@aol.com
Just another guess; they didn't get the intellectual property
that should have gone with the gear.
/BAH
Tim Shoppa
> The 1956 Bell System consent decree was rigidly interpreted and kept
> Teletype from moving into computer technology. (basically said that
> the Bell System would not engage in data processing. Also that Teletype
> couldn't sell anything on the open market that had not been developed
> for Bell System use, with an exception for government purchases.)
So the reason that the 33 and 43 made it into the mini/microcomputer
world at all was because they were existing variants of Bell System
equipment? Sounds plausible.
For a while in the 70's/80's I read of Teletype doing some interesting
stuff with video terminal bit-blitting, not sure how that would've
been related to Bell System products (maybe it was all internal Bell
System stuff that couldn't have been sold to others?) By that point I
was convinced the Tek 4010 style vector terminal had to be the winner
anyway, heck I still feel that way :-).
Tim.
Jim Haynes
Tim Shoppa <sho...@trailing-edge.com> wrote:
>
>So the reason that the 33 and 43 made it into the mini/microcomputer
>world at all was because they were existing variants of Bell System
>equipment? Sounds plausible.
Yes. The 32/33 were developed for Telex and TWX. There happened to be
the "private line" version that was simple, just a loop for the keyboard
and a loop for the printer. So that was readily adapted to minicomputers.
And for time sharing terminals, the using agencies had the option of
using telephony with modems or hard wiring with simple terminals.
>
>For a while in the 70's/80's I read of Teletype doing some interesting
>stuff with video terminal bit-blitting, not sure how that would've
I got the impression that Bell Labs saw computing developing along the
time-sharing line, with graphics done in the terminal but the major
computing being done centrally. Of course the microcomputers blew that
away; you could buy more computing power for less money in personal
computers than in big time shared machines. Grosch's Law repealed,
at last.
Jim Haynes
>33. I'm nostalgic over it because it's what I used. But I thought
>the keyboard was lousy and not conducive to comfortable fast typing.
>
same principles but for up/low case ASCII was even worse. That would
have sunk the Model 37 if its basic obsolescence had not done so already.
They knew how to make a good keyboard but didn't feel they had time to
develop it and meet the target dates for Model 37 introduction, so
that top-of-the-line machine was produced with the bottom-of-the-line
keyboard.
At the time it cost about $50 to manufacture that keyboard. We never
would have believed that before the year 2000 you'd be able to buy a
104-key electronic keyboard at Wal-Mart for less than $10.
Jim Haynes
<jmfb...@aol.com> wrote:
>
>
>In article <MPG.206b4b319...@news.individual.net>,
> krw <k...@att.bizzzz> wrote:
>>In article <1174502280.4...@o5g2000hsb.googlegroups.com>,
>>hanc...@bbs.cpcn.com says...
>>> would've been well positioned to supply them, as well as high quality
>>> teleprinters to compete with Selectrics for high grade terminal use
Selectric was a terminal made from a typewriter. Teletype had learned
long before that a typewriter is not rugged enough to use as a heavy-
duty teleprinter. I've heard that there were three different versions
of Selectric with three different levels of ruggedness.
There is some monkey business involved in sales prices and rental rates.
The Bell System as a regulated company was constrained in how monthly
rates could be set, knowing the purchase price of the equipment. IBM
had a lot more flexibility. It appears that at times they made rental
rates low in comparison to selling prices to encourage renting; and at
other times they made rental rates high in comparison to selling prices
to encourage sales. So IBM, if they chose to, could rent a Selectric
terminal at rates competitive with a Teletype even if the machine was
a lot more costly to make. They may not have been more costly to make,
since IBM was making them by the millions as typewriters and IBM was
also outstanding in manufacturing engineering, able to produce equipment
at low cost.
>>>
>>> Would anyone know what happened to the corporation and why?
>>>
>>Just a guess; they didn't control the other end of the wire?
That's certainly part of it - along with the sudden devaluation of
expertise in making mechanical stuff there was a slowness to recognize
that the product was no longer used person-to-person but rather person-
to-computer.
Teletype was an odd duck in the Bell System. Outside of Teletype, Bell
Labs did all the R&D and Western Electric did all the manufacturing.
Teletype had its own R&D, so there was a certain amount of conflict
over which activies were on whose turf, Teletype's or Bell Labs'.
Teletype had relatively greater independence in the high speed product
line where there was no Bell System history. AT&T could decide there
was a need for the Dataspeed product, for example, and Teletype would
design it with no input from Bell Labs. But in the low speed product
line there was the long history going back to the beginning of TWX.
So Bell Labs people had a lot of control over what Teletype did in that
area. And some of them were old geezers who saw TWX and contemporary
private wire systems as the whole universe.
A favorite horrible example is the Teletype 33 and 35 machines made for
Dial TWX. The modem was a huge thing that took up most of the space in
the pedestal under the Teletype. The Call Control Unit built into the
Teletype had the dial or key pad, optionally a card dialer, and pushbuttons
and other pieces of a telephone. The interface between what Teletype
made and the Bell Labs designed modem used 99 wires! and each of them
was painfully negotiated between Teletype and Bell Labs. A little later
some of the Bell operating companies were saving money by using a Bell
103 type modem, which included the telephone, and buying a private-line
set from Teletype, which had an RS-232 interface as an option.
>
>Just another guess; they didn't get the intellectual property
>that should have gone with the gear.
>
I'm not clear on what you are saying here, please elaborate.
CBFalconer
>
... snip ...
>
> At the time it cost about $50 to manufacture that keyboard. We
> never would have believed that before the year 2000 you'd be able
> to buy a 104-key electronic keyboard at Wal-Mart for less than $10.
Never buy those. For $2 you can get a keyboard from the Salvation
Army without the execresence of Windows keys.
--
Chuck F (cbfalconer at maineline dot net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home.att.net>
--
Posted via a free Usenet account from http://www.teranews.com
jmfb...@aol.com
hay...@alumni.uark.edu (Jim Haynes) wrote:
>In article <ettsp9$8qk...@s1007.apx1.sbo.ma.dialup.rcn.com>,
> <jmfb...@aol.com> wrote:
>>
>>
>>In article <MPG.206b4b319...@news.individual.net>,
>> krw <k...@att.bizzzz> wrote:
>>>In article <1174502280.4...@o5g2000hsb.googlegroups.com>,
>>>hanc...@bbs.cpcn.com says...
>>>> I would've thought given the huge demand for terminals, Teletype
>>>> would've been well positioned to supply them, as well as high quality
>>>> teleprinters to compete with Selectrics for high grade terminal use
>
>Selectric was a terminal made from a typewriter.
No. Selectric was the typewriter. For a while my RUNOFF group
used it as an output device to produce excellent print quality.
> Teletype had learned
>long before that a typewriter is not rugged enough to use as a heavy-
>duty teleprinter.
This makes no sense. I broke enough TTY33s to be banned from
using them. All you had to do is type faster than 35 WPM and
it broke. Selectric typewriters could deal with people who
typed over 100 WPM.
> I've heard that there were three different versions
>of Selectric with three different levels of ruggedness.
<snip>
/BAH
Tim Shoppa
> In article <1174580319.138498.323...@d57g2000hsg.googlegroups.com>,
> >For a while in the 70's/80's I read of Teletype doing some interesting
> >stuff with video terminal bit-blitting, not sure how that would've
>
> I got the impression that Bell Labs saw computing developing along the
> time-sharing line, with graphics done in the terminal but the major
> computing being done centrally. Of course the microcomputers blew that
> away; you could buy more computing power for less money in personal
> computers than in big time shared machines. Grosch's Law repealed,
> at last.
And then in the late 90's PC's again became "terminals" (aka "web
browsers") taking data coming in over a character-based datastream and
rendering it into bitmaps on the screen, and then batching up the next
request to send to the host :-).
Someday I'll write that Tek-4014-rendering web browser!
Tim.
krw
hay...@alumni.uark.edu says...
> In article <ettsp9$8qk...@s1007.apx1.sbo.ma.dialup.rcn.com>,
> <jmfb...@aol.com> wrote:
> >
> >
> >In article <MPG.206b4b319...@news.individual.net>,
> > krw <k...@att.bizzzz> wrote:
> >>In article <1174502280.4...@o5g2000hsb.googlegroups.com>,
> >>hanc...@bbs.cpcn.com says...
> >>> I would've thought given the huge demand for terminals, Teletype
> >>> would've been well positioned to supply them, as well as high quality
> >>> teleprinters to compete with Selectrics for high grade terminal use
>
> Selectric was a terminal made from a typewriter. Teletype had learned
> long before that a typewriter is not rugged enough to use as a heavy-
> duty teleprinter. I've heard that there were three different versions
> of Selectric with three different levels of ruggedness.
I know of three off the top of my hear. The Seletric typewriter,
MTST/MCST/Communicating Selecttric, and the 2741 variety. It didn't
take long for a computer to run a Selectric typewriter into the
ground.
--
Keith
Anne & Lynn Wheeler
krw <k...@att.bizzzz> writes:
> I know of three off the top of my head. The Seletric typewriter,
> take long for a computer to run a Selectric typewriter into the
> ground.
not a "typewriter" ... but the golfball mechanism was also used in the
1052. 1052-7 were the operator consoles for majority of the 360s and
were supposedly significantly more robust ... since many of them could
be printing nearly nonstop.
i once ordered an fe toolkit ... it took 3-4 months just to figure out
how to get the order accepted (apparently nobody out of FE org had
ever ordered one before). it still came with some number of pieces for
dealing with springs and other mechanical pieces related to
typewriters. i've even had opportunity to use one of two of the tools
from time-to-time.
hanc...@bbs.cpcn.com
> I got the impression that Bell Labs saw computing developing along the
> time-sharing line, with graphics done in the terminal but the major
> computing being done centrally. Of course the microcomputers blew that
> away; you could buy more computing power for less money in personal
> computers than in big time shared machines. Grosch's Law repealed,
> at last.
I think many in the industry were thinking along those lines, and the
ASCII model 33/35 were developed with that partially in mind.
I believe Bell Labs actually used an earlier teleprinter as a computer
terminal way back in 1939 with one of their relay computers.
As to mini computers, they still needed an I/O device and Teletypes
worked out perfectly for that role.
I'm just guessing, but I suspect far more Teletype Model 33s were
built for data processing terminal use than telegraph use.
hanc...@bbs.cpcn.com
>
> At the time it cost about $50 to manufacture that keyboard. We never
> would have believed that before the year 2000 you'd be able to buy a
> 104-key electronic keyboard at Wal-Mart for less than $10.
I believe in 1978 my Smith Corona electric portable typewriter (high
end) cost about $200-$300. Today a plain typewriter costs only $100
yet has better print quality and a memory with some functions.
Of course at that time an IBM Selectric office grade typewriter would
be about $1,000.
IBM briefly made a home version of the Selectric but I never saw one.
I picked up some Selectrics used for next to nothing. But since I
rarely use them, the lube has dried out and they don't work that well.
Tim Shoppa
> I think many in the industry were thinking along those lines, and the
> ASCII model 33/35 were developed with that partially in mind.
>
> I believe Bell Labs actually used an earlier teleprinter as a computer
> terminal way back in 1939 with one of their relay computers.
I saw some truly amazing SCADA stuff done with stepper relays, pen
plotters, and an occasional Baudot-teletype-based printing/logging
(using paper tape) unit. All this stuff was being disposed of by the
electrical industry in the 70's as they bought new minicomputer-based
SCADA systems.
Tim.
Joe Pfeiffer
Different problem: I'd never heard of actually breaking a Model 33 by
typing too fast, but there were lots of machines that could be jammed
that way. Other than special cases like that, though, a Model 33 could
be left in a corner typing out reams of text for months and years on
end.
Tim Shoppa
> jmfbah...@aol.com writes:
> > In article <rFHMh.16837$Jl.7...@newsread3.news.pas.earthlink.net>,
What I loved about the 33 was the keyboard: if it wasn't ready for you
to type the next letter, you couldn't press the next letter. It was a
mechanical interlock. When you felt the keyboard "take" your press,
then you move on to the next press.
I'm not saying that it wouldn't have been impossible to jam down the
key but that's not the way to type :-).
I'm sure that others hated this "feature" of the keyboard.
I contrast this with electronic keyboards of the same era that
sometimes were completely incapable of 2-key-rollover, producing
complete junk if you have two keys pressed at the same time.
What I do not like are modern mushy computer keyboards. The IBM Model
M from the 80's is pretty non-mushy for a PC-clone keyboard.
Tim.
mensa...@aol.com
But you had to at least occasionally check the paper roll and ribbon.
mensa...@aol.com
> On Mar 22, 10:48 pm, hay...@alumni.uark.edu (Jim Haynes) wrote:
>
>
>
> > At the time it cost about $50 to manufacture that keyboard. We never
> > would have believed that before the year 2000 you'd be able to buy a
> > 104-key electronic keyboard at Wal-Mart for less than $10.
>
> I believe in 1978 my Smith Corona electric portable typewriter (high
> end) cost about $200-$300. Today a plain typewriter costs only $100
> yet has better print quality and a memory with some functions.
>
> Of course at that time an IBM Selectric office grade typewriter would
> be about $1,000.
Once someone broke into the company I worked for and
stole all the Selectric typewriters and left all the
computers.
Tim Shoppa
> Once someone broke into the company I worked for and
> stole all the Selectric typewriters and left all the
> computers.
Selectrics didn't depreciate as fast :-).
Tim.
CBFalconer
>
... snip ...
>
> Different problem: I'd never heard of actually breaking a Model
> 33 by typing too fast, but there were lots of machines that could
> be jammed that way. Other than special cases like that, though,
> a Model 33 could be left in a corner typing out reams of text for
> months and years on end.
No it couldn't. About one year into it's life the clutch etc.
seized.
hanc...@bbs.cpcn.com
> Different problem: I'd never heard of actually breaking a Model 33 by
> typing too fast, but there were lots of machines that could be jammed
> that way. Other than special cases like that, though, a Model 33 could
> be left in a corner typing out reams of text for months and years on
> end.
Actually, the model 33 was designated as "light duty", while its
sister, the model 35, was designated as heavy duty. The console
typewriter for our HP 2000 was a model 35. Of course, many places
used the 33 quite extensively as you describe.
When our school was renting a 33 from the phone company for
timesharing use, the phone company would complain we were overusing
it. I don't know if anything more came of that. Eventually they were
replaced with owned units and an external modem.
Michael Black
> What I loved about the 33 was the keyboard: if it wasn't ready for you
> to type the next letter, you couldn't press the next letter. It was a
> mechanical interlock. When you felt the keyboard "take" your press,
> then you move on to the next press.
>
That is my recollection from a non-Teletype baudot machine. It required a
certain rhythm to get the typing right.
I think I've said it before, I got that machine the year I was taking
touch-typing at school. So towards the end of the year when I got it,
I was actually practicing on the baudot machine. It generally did work,
even though there was that rhythm, and of course no lowercase.
Micahel
Jim Haynes
<jmfb...@aol.com> wrote:
>> Teletype had learned
>>long before that a typewriter is not rugged enough to use as a heavy-
>>duty teleprinter.
>
>This makes no sense. I broke enough TTY33s to be banned from
>using them. All you had to do is type faster than 35 WPM and
>it broke. Selectric typewriters could deal with people who
>typed over 100 WPM.
>
You missed my "heavy-duty" in the previous paragraph. The 33 was
specifically designed as a light-duty machine, consequently selling
for a lot less than its heavy-duty counterpart. The original goal
for the 32/33 line was in TWX installations in offices where they
would be used about 2 hours per day, max. In fact the 33s used as
computer terminals and minicomputer consoles got used a lot more than
that, and they broke down a lot. Although there were improvements
through the production life of the machine that made them last
somewhat longer. There were also anti-improvements. The 32/33 line
was never intended to handle sprocket feed forms, but along came some
customer important enough to get that feature added. It was not very
reliable.
I know there were some reliability problems with IBM 2741 terminals
(Selectric based) in heavy-duty service.
hanc...@bbs.cpcn.com
> The 32/33 line
> was never intended to handle sprocket feed forms, but along came some
> customer important enough to get that feature added. It was not very
> reliable.
I remember seeing (using?) one of those at some company. IIRC, CTRL-L
was form feed, and if you typed that or had it sent to that machine, a
whole page would fly out quickly. (Well, to us kids at the time, it
was neat.)
Indeed, I remember control functions were marked in different ways on
different keyboards even if they were basically identical. Some
keyboards had many functions in red lettering. Others only had a few
functions in white letters.
I think CTRL-G was the bell, a real one, that also sounded as you got
close to the end of the line, just like a typewriter. As a joke, we'd
write a program to issue a bunch of bells to simulate sounding of an
alarm, but the machine noise of each character and weakness of the
bell didn't sound very convincing. Newspaper people say the AP used
three bells to announce a special flash news item coming through.
Maybe it was more audible on the older (pre-33) model machines.
As an aside, the typing sound of the earlier models was often in the
background of radio news. They continued this with a recording after
the machines themselves were gone. I think the older machines sounded
a bit more pleasant.
Frank McCoy
>Joe Pfeiffer wrote:
>>
>... snip ...
>>
>> Different problem: I'd never heard of actually breaking a Model
>> 33 by typing too fast, but there were lots of machines that could
>> be jammed that way. Other than special cases like that, though,
>> a Model 33 could be left in a corner typing out reams of text for
>> months and years on end.
>
>No it couldn't. About one year into it's life the clutch etc.
>seized.
>
For that kind of use, you needed a 35.
Now *there* was a rugged machine.
The type-quality was about 1000% better than a 35, too.
In fact, it was better than Selectric.
(I *know*; having owned both and run them side-by-side. In fact, the
TTY is still out in the garage ... The museum hasn't picked it up yet.)
Where it fell down was that the type-box was only upper-case.
(I heard rumors of an ASR-35 with upper/lower case; but never saw one.)
Some of the daisy-wheel printers were better; but generally only if you
used the single-use film ribbons.
--
_____
/ ' / ™
,-/-, __ __. ____ /_
(_/ / (_(_/|_/ / <_/ <_
krw
sho...@trailing-edge.com says...
> On Mar 23, 12:45 pm, Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
> > jmfbah...@aol.com writes:
> > > In article <rFHMh.16837$Jl.7...@newsread3.news.pas.earthlink.net>,
> > > hay...@alumni.uark.edu (Jim Haynes) wrote:
> >
> > > > Teletype had learned
> > > >long before that a typewriter is not rugged enough to use as a heavy-
> > > >duty teleprinter.
> >
> > > This makes no sense. I broke enough TTY33s to be banned from
> > > using them. All you had to do is type faster than 35 WPM and
> > > it broke. Selectric typewriters could deal with people who
> > > typed over 100 WPM.
> >
> > Different problem: I'd never heard of actually breaking a Model 33 by
> > typing too fast, but there were lots of machines that could be jammed
> > that way. Other than special cases like that, though, a Model 33 could
> > be left in a corner typing out reams of text for months and years on
> > end.
>
> What I loved about the 33 was the keyboard: if it wasn't ready for you
> to type the next letter, you couldn't press the next letter. It was a
> mechanical interlock. When you felt the keyboard "take" your press,
> then you move on to the next press.
I didn't care for the 33 keyboard at all, likely because I first used
one after 029s and Selectrics. The slope of the keyboard was strange
and the stroke too long, causing finger fatigue.
> I'm not saying that it wouldn't have been impossible to jam down the
> key but that's not the way to type :-).
>
> I'm sure that others hated this "feature" of the keyboard.
>
> I contrast this with electronic keyboards of the same era that
> sometimes were completely incapable of 2-key-rollover, producing
> complete junk if you have two keys pressed at the same time.
The 3270 keyboards were quite nice. The "klunker" inside gave a nice
solid keypunch feel. Of course they were heavy enough to be
weapons.
> What I do not like are modern mushy computer keyboards. The IBM Model
> M from the 80's is pretty non-mushy for a PC-clone keyboard.
Model-Ms, or the same thing, are still made. I bought one with a USB
interface to go on my ThinkPad (doesn't like PS/2 for some reason).
--
Keith
Gerard Schildberger
|> shoppa wrote:
|>> Joe Pfeiffer wrote:
|>>> jmfbah wrote:
If not weapons, then boat anchors, or ballast. ______________Gerard S.
|> What I do not like are modern mushy computer keyboards. The IBM Model
|> M from the 80's is pretty non-mushy for a PC-clone keyboard.
Yes, that is why I got two of the IBM Model M keyboards, they are the
best thing next to the 327x keyboards. _______________________Gerard S.
| Model-Ms, or the same thing, are still made. I bought one with a USB
| interface to go on my ThinkPad (doesn't like PS/2 for some reason).
Yuppers, I couldn't live without 'em. For a touch typist, they are
the only way to go. __________________________________________Gerard S.
jmfb...@aol.com
"Tim Shoppa" <sho...@trailing-edge.com> wrote:
>On Mar 23, 12:45 pm, Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
>> jmfbah...@aol.com writes:
>> > In article <rFHMh.16837$Jl.7...@newsread3.news.pas.earthlink.net>,
>> > hay...@alumni.uark.edu (Jim Haynes) wrote:
>>
>> > > Teletype had learned
>> > >long before that a typewriter is not rugged enough to use as a heavy-
>> > >duty teleprinter.
>>
>> > This makes no sense. I broke enough TTY33s to be banned from
>> > using them. All you had to do is type faster than 35 WPM and
>> > it broke. Selectric typewriters could deal with people who
>> > typed over 100 WPM.
>>
>> Different problem: I'd never heard of actually breaking a Model 33 by
>> typing too fast, but there were lots of machines that could be jammed
>> that way. Other than special cases like that, though, a Model 33 could
>> be left in a corner typing out reams of text for months and years on
>> end.
>
>What I loved about the 33 was the keyboard: if it wasn't ready for you
>to type the next letter, you couldn't press the next letter. It was a
>mechanical interlock. When you felt the keyboard "take" your press,
>then you move on to the next press.
Yea, but if you timed your keystrokes just right, you could make it
go too fast. That's what I was told by the hardware type.
>
>I'm not saying that it wouldn't have been impossible to jam down the
>key but that's not the way to type :-).
>
>I'm sure that others hated this "feature" of the keyboard.
>
>I contrast this with electronic keyboards of the same era that
>sometimes were completely incapable of 2-key-rollover, producing
>complete junk if you have two keys pressed at the same time.
Like ^C?
>
>What I do not like are modern mushy computer keyboards. The IBM Model
>M from the 80's is pretty non-mushy for a PC-clone keyboard.
I haven't tried modern keyboards for a while. I'll try to remember
to do that the next time I meet one.
/BAH
jmfb...@aol.com
jmfb...@aol.com
Frank McCoy <mcc...@millcomm.com> wrote:
>In alt.folklore.computers CBFalconer <cbfal...@yahoo.com> wrote:
>
>>Joe Pfeiffer wrote:
>>>
>>... snip ...
>>>
>>> Different problem: I'd never heard of actually breaking a Model
>>> 33 by typing too fast, but there were lots of machines that could
>>> be jammed that way. Other than special cases like that, though,
>>> a Model 33 could be left in a corner typing out reams of text for
>>> months and years on end.
>>
>>No it couldn't. About one year into it's life the clutch etc.
>>seized.
>>
>For that kind of use, you needed a 35.
>Now *there* was a rugged machine.
>The type-quality was about 1000% better than a 35, too.
>In fact, it was better than Selectric.
Huh? You have typos and I can't correct them because I keep
forgetting what the TTYs we used for grey consoles were called.
The type quality of these were never better than a Selectric.
>(I *know*; having owned both and run them side-by-side. In fact, the
>TTY is still out in the garage ... The museum hasn't picked it up yet.)
<snip>
/BAH
jmfb...@aol.com
hanc...@bbs.cpcn.com wrote:
>On Mar 22, 10:48 pm, hay...@alumni.uark.edu (Jim Haynes) wrote:
>>
>> At the time it cost about $50 to manufacture that keyboard. We never
>> would have believed that before the year 2000 you'd be able to buy a
>> 104-key electronic keyboard at Wal-Mart for less than $10.
>
>I believe in 1978 my Smith Corona electric portable typewriter (high
>end) cost about $200-$300. Today a plain typewriter costs only $100
>yet has better print quality and a memory with some functions.
>
>Of course at that time an IBM Selectric office grade typewriter would
>be about $1,000.
Oh! It got cheap. Maybe you people are talking about much
later and worse-built Selctric models.
>
>IBM briefly made a home version of the Selectric but I never saw one.
>I picked up some Selectrics used for next to nothing. But since I
>rarely use them, the lube has dried out and they don't work that well.
/BAH
Frank McCoy
>In article <dji803p1qaqjfksar...@4ax.com>,
> Frank McCoy <mcc...@millcomm.com> wrote:
>>In alt.folklore.computers CBFalconer <cbfal...@yahoo.com> wrote:
>>
>>>Joe Pfeiffer wrote:
>>>>
>>>... snip ...
>>>>
>>>> Different problem: I'd never heard of actually breaking a Model
>>>> 33 by typing too fast, but there were lots of machines that could
>>>> be jammed that way. Other than special cases like that, though,
>>>> a Model 33 could be left in a corner typing out reams of text for
>>>> months and years on end.
>>>
>>>No it couldn't. About one year into it's life the clutch etc.
>>>seized.
>>>
>>For that kind of use, you needed a 35.
>>Now *there* was a rugged machine.
>>The type-quality was about 1000% better than a 35, too.
>>In fact, it was better than Selectric.
>
>Huh? You have typos and I can't correct them because I keep
>forgetting what the TTYs we used for grey consoles were called.
>The type quality of these were never better than a Selectric.
>
The type-quality of a 33 was abysmal.
The type-quality of a 35 was excellent.
Slow, but good.
The big advantages when I went to daisy-wheel were:
A. Lower case.
B. Speed.
C. Incremental positioning. (You could do graphics!)
Selectrics, in comparison, were crappy.
While not nearly as bad as the type-drum on a TTY-33, that ball had four
big defects:
1. It was still moving when slammed against the ribbon.
2. It was *plastic* and wore out quickly.
3. It had a lot of inertia.
4. Being a ball, the registration was never perfect.
These are all the same problems essentially that the TTY-33 had; except
the type-cylinder there was metal ... *cheap* metal. ;-{
The solid-metal type-box on the 35 put the type in precisely the right
position before the hammer pounded it. The same thing with daisy-wheel
printers, except much faster.
Daisy-wheels, being made of plastic also, generally *broke* much easier;
but were so much cheaper and easier to change that you forgave them.
They rarely wore out; usually breaking a letter first.
Some people would put that last as a plus for Selectrics.
>>(I *know*; having owned both and run them side-by-side. In fact, the
>>TTY is still out in the garage ... The museum hasn't picked it up yet.)
><snip>
>
>/BAH
--
Anne & Lynn Wheeler
> I know there were some reliability problems with IBM 2741 terminals
> (Selectric based) in heavy-duty service.
i.e. selectric typewriters, 2741s, and 1052s all used the golfball
mechanism, however the 1052-7 used as 360 "consoles" were
significantly more robust/rugged than the others; recent post
http://www.garlic.com/2007f.html#62 What happened to the Teletype Corporation?
Joe Pfeiffer
> >
> >What I do not like are modern mushy computer keyboards. The IBM Model
> >M from the 80's is pretty non-mushy for a PC-clone keyboard.
>
> I haven't tried modern keyboards for a while. I'll try to remember
> to do that the next time I meet one.
My Logitech wireless keyboard at home feels pretty good to me. My
Belkin at work is a piece of mushy crap.
krw
pfei...@cs.nmsu.edu says...
dump. I always had a dozen or so Model-Ms stashed away. I didn't go
through too many though. I've only had one fail in almost 20 years
and that was old enough to drive.
--
Keith
Jim Haynes
>(I heard rumors of an ASR-35 with upper/lower case; but never saw one.)
The up/low machine was the Model 37, which used a very different mechanism
from the 35. It was designed for 150 wpm. Only a few were made. One
problem was that it came to market too late, as a result of stretched-out
development cycle. And that, some of us believe, was the result of
mechanical people insisting they could do mechanically something that
would have been a lot easier to do electronically. Also the 37 had a
really bad keyboard, based on the Model 33 and made worse by having a
lot more keys for up/low ASCII. Teletype knew how to make a good keyboard
but felt there wasn't time to develop it for the 37 - and then the printer
was so late that they would have had time after all, but they had not
started.
Teletype's last successful all-mechanical product was a stock ticker
based on the Model 37 technology and introduced about 1965. I wouldn't
consider the Model 37 page printer machine successful, in view of the
small number produced. Model 38 was a 33 adapted to upper and lower
case, and was also too late to market to be worth doing. Somebody decided
it should be wide enough to handle 14" computer forms. Widening made
some of the internal parts flex more than they did in the 33, and this
caused some reliability problems. So only a few of those were made, too.
And then the next product of that sort to come along was Model 43, with
electronic keyboard and dot-matrix printer and lots of electronics inside.
Frank McCoy
>In article <dji803p1qaqjfksar...@4ax.com>,
>Frank McCoy <mcc...@millcomm.com> wrote:
>>Where it fell down was that the type-box was only upper-case.
>>(I heard rumors of an ASR-35 with upper/lower case; but never saw one.)
>
>The up/low machine was the Model 37, which used a very different mechanism
>from the 35. It was designed for 150 wpm.
That would suck.
>Only a few were made. One
>problem was that it came to market too late, as a result of stretched-out
>development cycle. And that, some of us believe, was the result of
>mechanical people insisting they could do mechanically something that
>would have been a lot easier to do electronically. Also the 37 had a
>really bad keyboard, based on the Model 33 and made worse by having a
>lot more keys for up/low ASCII.
That would *really* suck.
The 35 was good. The 33 ....
Well, they always seemed to be falling apart.
>Teletype knew how to make a good keyboard
>but felt there wasn't time to develop it for the 37 - and then the printer
>was so late that they would have had time after all, but they had not
>started.
>
>Teletype's last successful all-mechanical product was a stock ticker
>based on the Model 37 technology and introduced about 1965. I wouldn't
>consider the Model 37 page printer machine successful, in view of the
>small number produced. Model 38 was a 33 adapted to upper and lower
>case, and was also too late to market to be worth doing. Somebody decided
>it should be wide enough to handle 14" computer forms. Widening made
>some of the internal parts flex more than they did in the 33, and this
>caused some reliability problems.
As if the 33 didn't have enough already. ;-{
>So only a few of those were made, too.
>And then the next product of that sort to come along was Model 43, with
>electronic keyboard and dot-matrix printer and lots of electronics inside.
Which I presume was WAY too late to compete with all of the other
dot-matrix machines printers and good keyboards connected to computers.
Gerard Schildberger
|> pfeiffer wrote:
|>> jmfbahciv wrote:
|>>> What I do not like are modern mushy computer keyboards. The IBM Model
|>>> M from the 80's is pretty non-mushy for a PC-clone keyboard.
|>> I haven't tried modern keyboards for a while. I'll try to remember
|>> to do that the next time I meet one.
|> My Logitech wireless keyboard at home feels pretty good to me. My
|> Belkin at work is a piece of mushy crap.
| At work I'd liberate the keyboards off systems on their way to the
| dump. I always had a dozen or so Model-Ms stashed away. I didn't go
| through too many though. I've only had one fail in almost 20 years
| and that was old enough to drive.
I never had an IBM keyboard (model M) or 327x fail on me, ... ever.
However, my boss used to have a channel connected IBM 3278 terminal
at home (yeah, the company hired a trencher and laid a coax to
his house). His cat liked the warmth on top of the terminal and,
of course, was up there a lot, and one of his jobs was to give
himself a bath by licking himself all over ... with the usual
results: barfing up furballs (and other stuff).
When my boss called the IBM CE (the terminal was rented, of
course, and therefore under warranty), he was very apologetic
and was begging forgiveness for the cat's up-chucking, and
the CE said, "that's nothing, I've cleaned out a lot worse".
I always meant to talk to that CE and find out what the heck
was was worse than that, but never did. _________________Gerard S.
CJT
I'm typing on one right now. They just keep going. Plus, they've got
the best feel of any keyboard I've ever used.
--
The e-mail address in our reply-to line is reversed in an attempt to
minimize spam. Our true address is of the form che...@prodigy.net.
CBFalconer
>
... snip ...
>
> Teletype's last successful all-mechanical product was a stock
> ticker based on the Model 37 technology and introduced about 1965.
> I wouldn't consider the Model 37 page printer machine successful,
> in view of the small number produced. Model 38 was a 33 adapted
> to upper and lower case, and was also too late to market to be
> worth doing. Somebody decided it should be wide enough to handle
> 14" computer forms. Widening made some of the internal parts flex
> more than they did in the 33, and this caused some reliability
> problems. So only a few of those were made, too. And then the
> next product of that sort to come along was Model 43, with
> electronic keyboard and dot-matrix printer and lots of electronics
> inside.
Once the Epsom MX80 became available, at around 300 to 500 USD, we
never considered a Teletype for printed output again. The problem
was getting a 20mA loop interface for them. They were available,
but hard to get. I had 40 odd of those things spread out over
miles of leased copper.
Frank McCoy
>Jim Haynes wrote:
>>
>... snip ...
>>
>> Teletype's last successful all-mechanical product was a stock
>> ticker based on the Model 37 technology and introduced about 1965.
>> I wouldn't consider the Model 37 page printer machine successful,
>> in view of the small number produced. Model 38 was a 33 adapted
>> to upper and lower case, and was also too late to market to be
>> worth doing. Somebody decided it should be wide enough to handle
>> 14" computer forms. Widening made some of the internal parts flex
>> more than they did in the 33, and this caused some reliability
>> problems. So only a few of those were made, too. And then the
>> next product of that sort to come along was Model 43, with
>> electronic keyboard and dot-matrix printer and lots of electronics
>> inside.
>
>Once the Epsom MX80 became available, at around 300 to 500 USD, we
>never considered a Teletype for printed output again. The problem
>was getting a 20mA loop interface for them. They were available,
>but hard to get. I had 40 odd of those things spread out over
>miles of leased copper.
>
With my ASR-35, since all my *other* terminals were RS-232, and I liked
to swap terminals all the time, I made a home-brew (2-transistor)
converter: One transistor going each way, from RS-232 to 20ma
current-loop. Worked great until one day I was moving the big beast and
carved the cable half in two when running over it with a dolly. The
converter is still there, buried inside th back-panel; and I assume it
still works; but I have long since lost the schematic; so I'd have to
puzzle out the wiring connections to that RS-232 connector if I wanted
to use it that way again.
Of course, since *both* computers I used to hook it up to got sold to a
collector, why bother?
I always figured that what I could whip up in an afternoon, others could
as well.
Richard Steiner
I've always thought so, at least when thinking about the WORLDFLIGHT
system I worked on at NWA (which used 6-bit FIELDATA, though).
For many types of data input, case might serve an aesthetic purpose,
but it really doesn't matter in terms of the actual MEANING of the data
in question. So flight plans, weather reports, etc., are just as good
in UPPER CASE as in MiXeD CasE, at least most of the time.
And on the 36-bit word-oriented machine it ran on, using FIELDATA for
storage saved quite a buit of space (six characters per word instead of
four!).
--
-Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Mableton, GA USA
Mainframe/Unix bit twiddler by day, OS/2+Linux+DOS hobbyist by night.
WARNING: I've seen FIELDATA FORTRAN V and I know how to use it!
The Theorem Theorem: If If, Then Then.
jmfb...@aol.com
rste...@visi.com (Richard Steiner) wrote:
>Here in alt.folklore.computers, jmfb...@aol.com spake unto us, saying:
>
>>In article <eu19b4$m07$1...@theodyn.ncf.ca>,
>> et...@FreeNet.Carleton.CA (Michael Black) wrote:
>>
>>>I think I've said it before, I got that machine the year I was taking
>>>touch-typing at school. So towards the end of the year when I got it,
>>>I was actually practicing on the baudot machine. It generally did work,
>>>even though there was that rhythm, and of course no lowercase.
>>
>>lOOKING BACK, THAT WAS A FEATURE! (nO LOWERCASE).
>
>I've always thought so, at least when thinking about the WORLDFLIGHT
>system I worked on at NWA (which used 6-bit FIELDATA, though).
>
>For many types of data input, case might serve an aesthetic purpose,
>but it really doesn't matter in terms of the actual MEANING of the data
>in question. So flight plans, weather reports, etc., are just as good
>in UPPER CASE as in MiXeD CasE, at least most of the time.
This is the most important area to keep the kiddies from being
CUTE. I hate cute. There is no place for cute when debugging
code using symbols. Or trying to find a file.
>
>And on the 36-bit word-oriented machine it ran on, using FIELDATA for
>storage saved quite a buit of space (six characters per word instead of
>four!).
I didn't care about space (that was a guy thing). I did care
about user throughput and efficiency in getting a computer job
done.
/BAH
CBFalconer
My systems couldn't use such 'whip-ups'. They were transmitting
medical results, and a dropped bit could have serious treatment
consequences. The environment involved miles of wire with
elevators, x-rays, diathermy, whatnot in various vicinities. So
everything was carefully optically isolated, and grounds strictly
controlled. The central driver had a large array of isolated
current sources.
Frank McCoy
You make it sound as if:
A. I didn't know what I was doing.
B. The result was more likely to drop or add bits than the original
current-loop. (Actually, it was *less* likely to do so.)
C. Only by a team of engineers spending months on a design is it
worthwhile or reliable.
Reasonable care and *understanding* both interfaces was all that are
needed to make a good handshake between the two; since the *timing* was
identical. A good scope to check the result helps too.
If isolation was needed, then using an optical isolator instead of the
transistor (with perhaps an additional transistor as driver) on each
side could do the job easily. The current-loop side might need the
addition of a small power-supply for proper isolation; as you couldn't
rob from the RS-232 power-pin like I did.
That a quick "whip up" only takes a good engineer hours instead of
several months of corporate approvals and specifications, does NOT mean
that it's any less correct or reliable ... assuming the engineer knows
his job and the thing he's working on.
I did.
The system survived use and abuse for decades without ever dropping a
bit or getting a parity-error. It was neatly done, on a homemade
pc-board, installed with proper mounting and strain-relief. Fully as
rugged and reliable as any other part of the entire TTY.
BTW: I was saying that if *I* could do such a job easily, the engineers
of any company worth working for could do the same. That a particular
job is EASY and FAST to do, doesn't mean it's not reliable!
That's *engineering* for you, not just kludging something that might
work. Knowing the design, knowing the parts and what they can do, and
putting them together in a planned manner ... then TESTING the result to
see that it meets (and preferably far exceeds) specification, does *NOT*
mean you have to spend days, weeks, and months getting corporate
approval for the design to be good.
The difference between a real engineer and somebody who graduates from
school with an engineering degree.
Joe Pfeiffer
>
> For many types of data input, case might serve an aesthetic purpose,
> but it really doesn't matter in terms of the actual MEANING of the data
> in question. So flight plans, weather reports, etc., are just as good
> in UPPER CASE as in MiXeD CasE, at least most of the time.
Except that either lower case or mixed case can be read more
accurately and more quickly, because of the greater variation in
letter shape. Lower-case-only printer would have been a better
choice than upper-case-only.
mensa...@aol.com
Everybody knows this. Any truth to the rumor that the
system was deliberately made inefficient just so they
wouldn't have to spell "God" in lower case?
CBFalconer
>>
... snip ...
>>>
>>> I always figured that what I could whip up in an afternoon, others
>>> could as well.
>>
>> My systems couldn't use such 'whip-ups'. They were transmitting
>> medical results, and a dropped bit could have serious treatment
>> consequences. The environment involved miles of wire with
>> elevators, x-rays, diathermy, whatnot in various vicinities. So
>> everything was carefully optically isolated, and grounds strictly
>> controlled. The central driver had a large array of isolated
>> current sources.
>
> You make it sound as if:
> A. I didn't know what I was doing.
> B. The result was more likely to drop or add bits than the original
> current-loop. (Actually, it was *less* likely to do so.)
> C. Only by a team of engineers spending months on a design is it
> worthwhile or reliable.
That was not my intent. There is a great difference between
something knocked up to drive something 6 feet away, and a
transmission system.
Richard Steiner
Joe Pfeiffer <pfei...@cs.nmsu.edu> spake unto us, saying:
>rste...@visi.com (Richard Steiner) writes:
>
>> For many types of data input, case might serve an aesthetic purpose,
>> but it really doesn't matter in terms of the actual MEANING of the data
>> in question. So flight plans, weather reports, etc., are just as good
>> in UPPER CASE as in MiXeD CasE, at least most of the time.
>
>Except that either lower case or mixed case can be read more
>accurately and more quickly, because of the greater variation in
>letter shape.
Yes, that might be true for textual data.
Most of the data I recall on our application screen displays tended to
be numeric values with short alpha headers or prefixes, and those alpha
headers derived minimal benefit from the addition of case.
>Lower-case-only printer would have been a better choice than upper-case-
>only.
I would guess that teletype practice was probably derived from radio or
telegraph practice. <??>
jmfb...@aol.com
Sure. However, you are talking about human-readable format.
I'm talking about the cases where machine-readable format
was more important than human-readable.
If you are a code^Wdeveloper, having properly written
variables hampers the job; code is not meant to be human-readable;
only the error messages need to have that kind of formatting.
/BAH
jmfb...@aol.com
"mensa...@aol.com" <mensa...@aol.com> wrote:
I missed this in Joe's comment. Lower case was lousy
print. Upper case gave you a chance of decyphering the
blurred character you saw on the page. Upper case
block-formatted letters were always easier for humans
to read. Ask any keypuncher or data enterer worth her salt.
/BAH
jmfb...@aol.com
rste...@visi.com (Richard Steiner) wrote:
>Here in alt.folklore.computers,
>Joe Pfeiffer <pfei...@cs.nmsu.edu> spake unto us, saying:
>
>>rste...@visi.com (Richard Steiner) writes:
>>
>>> For many types of data input, case might serve an aesthetic purpose,
>>> but it really doesn't matter in terms of the actual MEANING of the data
>>> in question. So flight plans, weather reports, etc., are just as good
>>> in UPPER CASE as in MiXeD CasE, at least most of the time.
>>
>>Except that either lower case or mixed case can be read more
>>accurately and more quickly, because of the greater variation in
>>letter shape.
>
>Yes, that might be true for textual data.
>
>Most of the data I recall on our application screen displays tended to
>be numeric values with short alpha headers or prefixes, and those alpha
>headers derived minimal benefit from the addition of case.
>
>>Lower-case-only printer would have been a better choice than upper-case-
>>only.
>
>I would guess that teletype practice was probably derived from radio or
>telegraph practice. <??>
Nuts. Think about lower case a and lower case o. Think about
lower case m and n and r all mushed together. You all seem
to have forgotten that printer output was lousy. It sometimes
took a TECO command to "see" if that character was a comma or
a period.
I have the very same problems with these webbing TTY screens
where on pixel is missing between the rn. I am always interpreting
it as m.
/BAH
Philip Nasadowski
"Tim Shoppa" <sho...@trailing-edge.com> wrote:
> I saw some truly amazing SCADA stuff done with stepper relays, pen
> plotters, and an occasional Baudot-teletype-based printing/logging
> (using paper tape) unit. All this stuff was being disposed of by the
> electrical industry in the 70's as they bought new minicomputer-based
> SCADA systems.
Most.
I still run into electromechanical systems (mostly pulse duration
transmitters) and Quindar stuff a lot. Heck, I know a few water
companies that still have racks of Quindar stuff all over.
Haven't seen a Teletype, though :( FDNY used to have 'em when I was a
kid, though...
Tim Shoppa
> In article <1174667407.459056.181...@l75g2000hse.googlegroups.com>,
> "TimShoppa" <sho...@trailing-edge.com> wrote:
>
> > I saw some truly amazing SCADA stuff done with stepper relays, pen
> > plotters, and an occasional Baudot-teletype-based printing/logging
> > (using paper tape) unit. All this stuff was being disposed of by the
> > electrical industry in the 70's as they bought new minicomputer-based
> > SCADA systems.
>
> Most.
>
> I still run into electromechanical systems (mostly pulse duration
> transmitters) and Quindar stuff a lot. Heck, I know a few water
> companies that still have racks of Quindar stuff all over.
Heck, for my day job I got to meet with the current president of
Quindar's successor last Friday :-). We have mostly upgraded from the
first generation!
Tim .
Tim Shoppa
> In article <1174667407.459056.181...@l75g2000hse.googlegroups.com>,
Now that I stretch my brain, I believe that those teletype-logging
SCADA systems were by Western Electric. I will scrounge for what books
I have on those systems (I think post-WWII). I definitely do remember
the power-line-carrier modems, racks filled with 6SN7-based generators
and balanced modulators and 807's for output drivers, being WE-
branded.
Tim.
mensa...@aol.com
> In article <1174855655.011315.120...@n59g2000hsh.googlegroups.com>,
And I doubt the decision makers were keypunchers.
Whether lower case was _actually_ easier to read
wasn't the question I asked. What I had heard was
that it was percieved to be better and a decision
was made (based purely on religous issues) not to
use it.
Now if that's just an urban legend, fine. I'm simply
asking if anyone has any information along those lines.
>
> /BAH
mensa...@aol.com
> In article <UV2BGpHpvSzc09...@visi.com>,> >Joe Pfeiffer <pfeif...@cs.nmsu.edu> spake unto us, saying:
>
> >>rstei...@visi.com (Richard Steiner) writes:
>
> >>> For many types of data input, case might serve an aesthetic purpose,
> >>> but it really doesn't matter in terms of the actual MEANING of the data
> >>> in question. So flight plans, weather reports, etc., are just as good
> >>> in UPPER CASE as in MiXeD CasE, at least most of the time.
>
> >>Except that either lower case or mixed case can be read more
> >>accurately and more quickly, because of the greater variation in
> >>letter shape.
>
> >Yes, that might be true for textual data.
>
> >Most of the data I recall on our application screen displays tended to
> >be numeric values with short alpha headers or prefixes, and those alpha
> >headers derived minimal benefit from the addition of case.
>
> >>Lower-case-only printer would have been a better choice than upper-case-
> >>only.
>
> >I would guess that teletype practice was probably derived from radio or
> >telegraph practice. <??>
>
> Nuts. Think about lower case a and lower case o. Think about
> lower case m and n and r all mushed together.
Not as bad as when you use an "Arial" font. I often do web pages
in <tty> font for that very reason (still using lower case). Try
looking at an all upper case web page. PITA.
> You all seem
> to have forgotten that printer output was lousy.
Who uses a character printer anymore? What's important now
are spreadsheets and their ilk. Bill Gates deserves to be
kicked in the ass for a lot of things, but certainly for
MS defaulting to Arial in technical applications like Excel
and Access instead of the more sensible choice of Courier.
> It sometimes
> took a TECO command to "see" if that character was a comma or
> a period.
Well, at least now you can temporarily switch magnification
to 200%. At least in Excel. For some strange reason, that feature
isn't available in Access.
>
> I have the very same problems with these webbing TTY screens
> where on pixel is missing between the rn. I am always interpreting
> it as m.
Oh, one last thing, I'm the modern equivalent of a keypuncher,
a database manager. So when I say lower case Courier is the prefered
font, I know what I'm talking about.
>
> /BAH
Jim Haynes
Richard Steiner <rste...@visi.com> wrote:
>>Lower-case-only printer would have been a better choice than upper-case-
>>only.
>
>I would guess that teletype practice was probably derived from radio or
>telegraph practice. <??>
>
days the ancestor of Teletype was trying to sell to Western Union and its
competitor Postal Telegraph. This was so long ago that Teletype was working
with office typewriters with solenoids to push the keys down, before
they started making their own machinery. Postal Telegraph wanted a machine
with up/low case and Western Union wanted a single-case machine.
Eventually the single case won out (until we get to Teletypesetter) and
was upper case. This makes a lot of sense considering that Morse telegraphy
was still in wide use at the time and is single case. It also makes sense
from saving transmission time, since having two cases means more bits
to transmit; and in the early days bandwidth was hard to come by. Altho
IBM's approach (much later) was to use the shift key mechanism of a
typewriter, and to send a shift character before upper case and an
unshift character before lower case. (Still takes more bits and more
time.)
I can only guess why Morse telegraph practice was upper case rather than
lower. Probably because people object to lower case in "god" and it
proper names. Morse operators in the telegraph industry were taught
to write in longhand, that being faster than printing in capitals or
in lower case. Military operators were taught to print in capitals.
With typewriters available for copy I suppose they would, for speed,
rather type in all upper case than have to shift cases when proper names
came along in the text.
Someone at Teletype, in the mid 1960s, had a design firm design a type
font that was monocase but was intended to overcome objections to lower
case in "god" (that was specifically noted in the report) and in proper
names. Probably the motive was that Teletype's up/low printer was taking
a long time to develop and the market was demanding up/low case, so
this was an attempt to fool the eye and get by with single case a while
longer. Nothing ever came of it. To my eye it was rather bizarre
looking and hard to read, and didn't fool anybody about being single case.
hanc...@bbs.cpcn.com
> What's important now
> are spreadsheets and their ilk. Bill Gates deserves to be
> kicked in the ass for a lot of things, but certainly for
> MS defaulting to Arial in technical applications like Excel
> and Access instead of the more sensible choice of Courier. ...
When I started in DP everything was in upper case--line printers,
printing terminals, CRT screens, keypunch labels, program terminals.
Indeed, as things (e.g. program code) switched to upper and lower
case it took some getting used to. Keep in mind we have substantial
green-on-glass 3270-type applications that are mostly or all upper
case.
As to what looks and works best, it depends on the application and can
be somewhat subjective.
For traditional mainframe work (ie COBOL), I like all caps in a plain
sans-serif fixed-space typeface.
For newer work, it varies by application. Generally upper/lower is
ok. If column alignment is necessary, then fixed font Courier is
good. However, for header and row literals, I like a contrasting
typeface, such as Times New Roman or Arial, just to set the literals
off from the data.
On spreadsheets, I don't mind Arial.
Indeed, there was a popular Selectric typeface that was sans serif (I
forgot the name of the font) but it was popular for financial work
like invoices.
For certain documents, I wish they used numbers styled after old
typewriter fonts, where the numbers were very distinct; that is, no
way a 3 or 6 be turned into an 8 because the ends stuck out far.
Historically, I believe the IBM 407 tab machine had a more modern
looking font while the 402 tab, which came out at the same time, had a
more traditional "typewriter" style font.
As to Teletypes, I didn't really care for the font style, especially
in reading long text in all caps. But for what we used it for--
computer statements--it was fine. There was a distinctive to the
Teletype font so you know where it came from, it was not the same as a
regular typewriter. I believe the A had slightly rounded lines.
Somewhere someone put up a TrueType teletype font.
As an aside, one thing that drove me crazy was inconsistent slashing
of O and O. Some people, including me and my employers, slashed the
zero, but many others slashed the letter. I believe Teletypes were
available in both styles. On some printers and screen the letter O is
made almost square and the numeral 0 is made very slim to distinct
them; I like that look. But many devices have them look almost the
same.
hanc...@bbs.cpcn.com
> It also makes sense
> from saving transmission time, since having two cases means more bits
> to transmit; and in the early days bandwidth was hard to come by.
WU wrote a piece on the introduction ASCII for the Technical Review
around 1962. My impression of their feelings (not overtly stated in
the article) was that WU saw many disadvantages but few advantages in
going to ASCII. I think they were quite happy with their 5 bit Baudot
and wanted to stick with it. I am guessing they had little interest
in going to upper/lower case printers because (1) the printer would
cost more to buy and maintain, (2) the bandwidth would be more, and
(3) they didn't see much of a need for it back then. AFAIK, the bulk
of the computer world in those days was 6 bits per character leaving
little room for lower case. Text communications back then tended to
be extremely brief.
I am not aware of any IBM keypunches that could punch lower case in
cards. Would anyone know of any?
Anne & Lynn Wheeler
> I am not aware of any IBM keypunches that could punch lower case in
> cards. Would anyone know of any?
I don't know of either 026 or 029 card punch with upper/lower case
shift key, however you could punch any punch hole combination with the
use of "multi-punch" ... you just didn't have a key for directly
selecting upper/lower case; i.e. just memorize punch hole combos from
the green card (which gave both bcdic and ebcdic mappings for
character to hex as well as punch hole combinations).
recent post mentioning green cards
http://www.garlic.com/~lynn/2007f.html#65 History - Early Green Card
and a crude converstion of old gcard ios3270 file to html
(although doesn't include any punch hole mappings)
http://www.garlic.com/~lynn/gcard.html
ebcdic to hex table
http://www.garlic.com/~lynn/gcard.html#1.3
ascii to hex table
http://www.garlic.com/~lynn/gcard.html#1.4
front & back scan of a "real" yellow "green card"
http://www.planetmvs.com/greencard/x20-1703-3.jpg
This has a table of EBCDIC to both hex and card punch holes
http://www.cs.uiowa.edu/~jones/cards/codes.html
upper-case A-I : 12-(1-9) punch holes
lower-case A-I : 12-10-(1-9) punch holes
upper-case J-R : 11-(1-9) punch holes
lower-case J-R : 12-11-(1-9) punch holes
upper-case S-Z : 10-(2-9) punch holes
lower-case S-Z : 11-10-(2-9) punch holes
i.e.
turn upper case A-I to lower-case by multi-punch "10" hole
turn upper case J-R to lower-case by multi-punch "12" hole
turn upper case S-Z to lower-case by multi-punch "11" hole
Richard Steiner
"mensa...@aol.com" <mensa...@aol.com> spake unto us, saying:
>Not as bad as when you use an "Arial" font. I often do web pages
>in <tty> font for that very reason (still using lower case). Try
>looking at an all upper case web page. PITA.
It doesn't take long to get used to reading all-uppercase output; I had
to do that when moving from a college ASCII environment to a production
FIELDATA environment in 1988. It wasn't really that hard.
BTW -- to me, Arial is much cleaner and easier to read then Courier,
which is why I have Firefox set to use Arial as the default sans serif
font and use that font by default on all web pages.
>> I have the very same problems with these webbing TTY screens
>> where on pixel is missing between the rn. I am always interpreting
>> it as m.
>
>Oh, one last thing, I'm the modern equivalent of a keypuncher,
>a database manager. So when I say lower case Courier is the prefered
>font, I know what I'm talking about.
In many computing contexts, things like "fonts" have no meaning.
We send vanilla ASCII data on the line from one machine to another, and
the presentation of such data is the responsibility of the client box
on the other end, not ours.
Richard Steiner
saying:
>Keep in mind we have substantial green-on-glass 3270-type applications
>that are mostly or all upper case.
Yup -- the Uniscope/UTS applications we're maintaining right now (and
which manu airlines maintain) are often all upper case, even when the
option to use lower case exists.
>As to what looks and works best, it depends on the application and can
>be somewhat subjective.
It can be VERY subjective. :-)
>As an aside, one thing that drove me crazy was inconsistent slashing
>of O and O. Some people, including me and my employers, slashed the
>zero, but many others slashed the letter. I believe Teletypes were
>available in both styles. On some printers and screen the letter O is
>made almost square and the numeral 0 is made very slim to distinct
>them; I like that look. But many devices have them look almost the
>same.
I prefer slashed or dotted zero characters myself. The screen font I
am using right now under OS/2 does that, as does the font I tend to use
on my UTS emulator at work.
Richard Steiner
>In article <UV2BGpHp...@visi.com>,
> rste...@visi.com (Richard Steiner) wrote:
We had high-speed band printers in college and at Unisys printing on
green-bar paper, as well as relatively high-tech DECWriters and such
which could be used interactively. Neither was too bad as long as the
printer ribbons were changed with enough frequency.
I don't remember seeing a lot of character mushing, but the band
printers did do a good job of making words dance due to vertical
alignment issues...
>I have the very same problems with these webbing TTY screens
>where on pixel is missing between the rn. I am always interpreting
>it as m.
Sounds like you need a better font. :-)
One of the reasons I tend to use a text web browser such as Links 1.x
instead of Firefox for reading news sites (and text apps for reading
forums such as this one) is clarity of text presentation.
It's much easier on my middle-aged eyes when I'm using a large glowing
cyan/green/yellow/white font on a black background.
Philip Nasadowski
"Tim Shoppa" <sho...@trailing-edge.com> wrote:
> Heck, for my day job I got to meet with the current president of
> Quindar's successor last Friday :-). We have mostly upgraded from the
> first generation!
Heh. I'm used to the plug in 19 inch rack stuff. I think the later
stuff was QEI . We've got tons of it around the office that's been
pulled out.
Interestingly, Bristol's PLC equipment besides being somewhat slow* and
programmed in that weird ACCOL language, have built in support for pulse
duration signals, which was how they did things before fast a/d
converters. The transmitting ends I've seen have all been mechanical
devices. I'm guessing the receive ends were at one time too, but I've
only seen electronic ones.
The Phone Company must have loved this years ago. Of course, today,
it's all radios...
*But man, that stuff never breaks. I've seen plenty of 3300 series
stuff that's been going for years and years. The boards warped from the
heat, but still running.
Philip Nasadowski
hanc...@bbs.cpcn.com wrote:
> As an aside, the typing sound of the earlier models was often in the
> background of radio news. They continued this with a recording after
> the machines themselves were gone. I think the older machines sounded
> a bit more pleasant.
WINS in NYC still has it. I'm sure 20 years from now, they'll still
have it, and people will wonder what the heck that noise is and why it's
there....
Frank McCoy
>Here in alt.folklore.computers,
>"mensa...@aol.com" <mensa...@aol.com> spake unto us, saying:
>
>>Not as bad as when you use an "Arial" font. I often do web pages
>>in <tty> font for that very reason (still using lower case). Try
>>looking at an all upper case web page. PITA.
>
>It doesn't take long to get used to reading all-uppercase output; I had
>to do that when moving from a college ASCII environment to a production
>FIELDATA environment in 1988. It wasn't really that hard.
>
>BTW -- to me, Arial is much cleaner and easier to read then Courier,
>which is why I have Firefox set to use Arial as the default sans serif
>font and use that font by default on all web pages.
I used to use "Courier New" for everything, to make certain of a fixed
format font. Lately though, I've been switching to the Lucida fonts for
such use ... Lucida Console is good for lots of things.
Lucida Typewriter for ... typewriter looks, instead of Courier.
;-}
>
>>> I have the very same problems with these webbing TTY screens
>>> where on pixel is missing between the rn. I am always interpreting
>>> it as m.
>>
>>Oh, one last thing, I'm the modern equivalent of a keypuncher,
>>a database manager. So when I say lower case Courier is the prefered
>>font, I know what I'm talking about.
>
>In many computing contexts, things like "fonts" have no meaning.
>
>We send vanilla ASCII data on the line from one machine to another, and
>the presentation of such data is the responsibility of the client box
>on the other end, not ours.
--
Stan Barr
>Eventually the single case won out (until we get to Teletypesetter) and
>was upper case. This makes a lot of sense considering that Morse telegraphy
>was still in wide use at the time and is single case.
5-bit Baudot/Murray code was/is u/c only too, of course.
Still in use on the wireless in RTTY and TOR forms...
--
Cheers,
Stan Barr stanb .at. dial .dot. pipex .dot. com
(Remove any digits from the addresses when mailing me.)
The future was never like this!
Joe Pfeiffer
> >
> > I missed this in Joe's comment. Lower case was lousy
> > print. Upper case gave you a chance of decyphering the
> > blurred character you saw on the page. Upper case
> > block-formatted letters were always easier for humans
> > to read. Ask any keypuncher or data enterer worth her salt.
OK... a legible lower case is better than a legible upper case. A
legible *anything* is better than illegible (just as the poor students
whoa re subjected to my handwriting...).
> And I doubt the decision makers were keypunchers.
>
> Whether lower case was _actually_ easier to read
> wasn't the question I asked. What I had heard was
> that it was percieved to be better and a decision
> was made (based purely on religous issues) not to
> use it.
>
> Now if that's just an urban legend, fine. I'm simply
> asking if anyone has any information along those lines.
I've never seen the actual studies (showing lower case is easier to
read than upper case), but I've heard the claim they exist. And based
on what I see on my terminal, it certainly matches my sample-of-one
results.
I've got no idea why upper case was used so universally on early ttys
and punchcards.
Joe Pfeiffer
> On Mar 26, 6:35 am, jmfbah...@aol.com wrote:
> >
> > >Yes, that might be true for textual data.
> >
> > >Most of the data I recall on our application screen displays tended to
> > >be numeric values with short alpha headers or prefixes, and those alpha
> > >headers derived minimal benefit from the addition of case.
Of course. I've never seen either an upper or a lower case 0. It's
only letters that are relevant.
> > >>Lower-case-only printer would have been a better choice than upper-case-
> > >>only.
> >
> > >I would guess that teletype practice was probably derived from radio or
> > >telegraph practice. <??>
> >
> > Nuts. Think about lower case a and lower case o. Think about
> > lower case m and n and r all mushed together.
With the font I use (terminus) I have no trouble distinguishing them.
And in my hazy memories of upper-case-only, M vs. N, S vs. 5, O vs. D
were at least as obnoxious as any of the examples you've presented.
The thing is, with lower case you've at least got a *chance* - you've
got ascenders and descenders. Upper case doesn't.
> Not as bad as when you use an "Arial" font. I often do web pages
> in <tty> font for that very reason (still using lower case). Try
> looking at an all upper case web page. PITA.
Major PITA. But for all the evils of modern web pages, I really can't
remember the last time I saw one in upper case.
> > You all seem
> > to have forgotten that printer output was lousy.
>
> Who uses a character printer anymore? What's important now
> are spreadsheets and their ilk. Bill Gates deserves to be
> kicked in the ass for a lot of things, but certainly for
> MS defaulting to Arial in technical applications like Excel
> and Access instead of the more sensible choice of Courier.
Was. Keyword was. But no, I don't remember it as being as lousy as
you do. I don't remember what printer I used back in my card-deck
upper-case-only days as a University of Washington student, but the
300LPM Printronix I used for several years down here (damn, that was a
great printer. If their 600LPM had only not sucked so bad) wasn't bad
at all.
> > It sometimes
> > took a TECO command to "see" if that character was a comma or
> > a period.
That's a problem today! On the reservations page at
http://lcctnm.org, I've had to put in a filter for people who make that
mistake.
> Well, at least now you can temporarily switch magnification
> to 200%. At least in Excel. For some strange reason, that feature
> isn't available in Access.
Excel? Access? Must be MS stuff. I don't know anything about them.
> >
> > I have the very same problems with these webbing TTY screens
> > where on pixel is missing between the rn. I am always interpreting
> > it as m.
That's kerning. In most cases, it enhances readability.... but not
in all.
> Oh, one last thing, I'm the modern equivalent of a keypuncher,
> a database manager. So when I say lower case Courier is the prefered
> font, I know what I'm talking about.
I like mixed case terminus. But courier isn't bad at all.
Joe Pfeiffer
> On Mon, 26 Mar 2007 19:04:40 GMT, Jim Haynes <hay...@alumni.uark.edu> wrote:
>
> >Eventually the single case won out (until we get to Teletypesetter) and
> >was upper case. This makes a lot of sense considering that Morse telegraphy
> >was still in wide use at the time and is single case.
>
> 5-bit Baudot/Murray code was/is u/c only too, of course.
> Still in use on the wireless in RTTY and TOR forms...
That's one of the subtlties -- it wasn't. It had a representation for
the first letter of the english alphabet, but whether that was
rendered as 'A', 'a', or that weird Egyption glyph that Prince used to
use for his name was a matter of convention.
Unless somewhere in the spec that I've never seen it says "thou shalt
render the code with the following bit-pattern", of course.
jmfb...@aol.com
"mensa...@aol.com" <mensa...@aol.com> wrote:
>On Mar 26, 6:33 am, jmfbah...@aol.com wrote:
>> In article <1174855655.011315.120...@n59g2000hsh.googlegroups.com>,
>> "mensana...@aol.com" <mensana...@aol.com> wrote:
>>
>>
>>
>>
>>
>> >On Mar 25, 2:37�pm, Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
>> >> rstei...@visi.com (Richard Steiner) writes:
>>
>> >> > For many types of data input, case might serve an aesthetic purpose,
>> >> > but it really doesn't matter in terms of the actual MEANING of the
data
>> >> > in question. So flight plans, weather reports, etc., are just as good
>> >> > in UPPER CASE as in MiXeD CasE, at least most of the time.
>>
>> >> Except that either lower case or mixed case can be read more
>> >> accurately and more quickly, because of the greater variation in
>> >> letter shape. Lower-case-only printer would have been a better
>> >> choice than upper-case-only.
>>
>> >Everybody knows this. Any truth to the rumor that the
>> >system was deliberately made inefficient just so they
>> >wouldn't have to spell "God" in lower case?
>>
>> I missed this in Joe's comment. Lower case was lousy
>> print. Upper case gave you a chance of decyphering the
>> blurred character you saw on the page. Upper case
>> block-formatted letters were always easier for humans
>> to read. Ask any keypuncher or data enterer worth her salt.
>
>And I doubt the decision makers were keypunchers.
If you produced a product which made it difficult for a keypuncher
to look at, the "decision makers" would be listening.
>
>Whether lower case was _actually_ easier to read
>wasn't the question I asked. What I had heard was
>that it was percieved to be better and a decision
>was made (based purely on religous issues) not to
>use it.
>
>Now if that's just an urban legend, fine. I'm simply
>asking if anyone has any information along those lines.
Take a good look at a card that has been keypunched. Notice
the dots that were used to print the characters. I don't
remember every seeing a typewriter ribbon that was not
a strip of material soaked in ink.
/BAH
jmfb...@aol.com
Characters were intended to be _machine_-readable, not
human-readable. Machines didn't need lower case.
/BAH
jmfb...@aol.com
rste...@visi.com (Richard Steiner) wrote:
>Here in alt.folklore.computers, jmfb...@aol.com spake unto us, saying:
>
>>In article <UV2BGpHp...@visi.com>,
>> rste...@visi.com (Richard Steiner) wrote:
>
>>>I would guess that teletype practice was probably derived from radio or
>>>telegraph practice. <??>
>>
>>Nuts. Think about lower case a and lower case o. Think about
>>lower case m and n and r all mushed together. You all seem
>>to have forgotten that printer output was lousy. It sometimes
>>took a TECO command to "see" if that character was a comma or
>>a period.
>
>We had high-speed band printers in college and at Unisys printing on
>green-bar paper, as well as relatively high-tech DECWriters
And I'm talking about an earlier era ;-). Much earlier, w.r.t.
print quality technology.
>and such
>which could be used interactively. Neither was too bad as long as the
>printer ribbons were changed with enough frequency.
>
>I don't remember seeing a lot of character mushing,
Son, that's because you are too young :-)).
> but the band
>printers did do a good job of making words dance due to vertical
>alignment issues...
>
>>I have the very same problems with these webbing TTY screens
>>where on pixel is missing between the rn. I am always interpreting
>>it as m.
>
>Sounds like you need a better font. :-)
Nope. It's everywhere. Newspapers, webs, books, etc.
>
>One of the reasons I tend to use a text web browser such as Links 1.x
>instead of Firefox for reading news sites (and text apps for reading
>forums such as this one) is clarity of text presentation.
>
>It's much easier on my middle-aged eyes when I'm using a large glowing
>cyan/green/yellow/white font on a black background.
The best viewing was our VT06. I can never remember the real name
for it.
/BAH
Michael Black
> On Mon, 26 Mar 2007 19:04:40 GMT, Jim Haynes <hay...@alumni.uark.edu> wrote:
>
>>Eventually the single case won out (until we get to Teletypesetter) and
>>was upper case. This makes a lot of sense considering that Morse telegraphy
>>was still in wide use at the time and is single case.
>
> 5-bit Baudot/Murray code was/is u/c only too, of course.
> Still in use on the wireless in RTTY and TOR forms...
But what's intriguing is that neither baudot or morse code is "uppercase".
They are both just letters, numerals and some punctuation. I'd never
given it thought before this thread, but certainly in morse code there is
no upper- or lower-case specified.
It's purely how it's written down at the other end that defines whether
it's UC or LC.
And ultimately, it's the same with baudot, though of course since it
was traditionally locked to mechanical machines, the printout mechanism
tends to define it, and that has been UC.
This is quite different from ASCII, where early Teletype machines obviously
only displayed upper case, but the code itself was capable of both upper and
lower case.
Michael
Joe Pfeiffer
> >I've got no idea why upper case was used so universally on early ttys
> >and punchcards.
>
> Characters were intended to be _machine_-readable, not
> human-readable. Machines didn't need lower case.
That gets back to a comment I made yesterday: all the encoding in the
holes meant was 'the first letter of the English alphabet'. Printing
it as either upper- or lower-case was just a convention to help humans
read it.
Frank McCoy
There were the "film ribbons": Essentially a strip of plastic with "ink"
powder painted on one surface.
hanc...@bbs.cpcn.com
> I don't know of either 026 or 029 card punch with upper/lower case
> shift key, however you could punch any punch hole combination with the
> use of "multi-punch" ...
I was never good at using that feature, I always lost track of what I
already punched or the column. Some fields have up to five punches in
them.
Years ago I understand it was common for shops to have an IBM #1
keypunch, which was a small portable numeric only punch. Useful for
emergency fixes. Everywhere I was had a spare 029 for emergency use.
Funny thing, although I am nostalgic for punch cards, I really found
terminals to be a huge improvement because of the ease of correction
and higher speed typing. I did not like 029s. I liked the 129
because I could backspace and fix errors.
When we got newer CRT terminals we could use lower case on them.
Gradually (very gradually) we used lower case in output where there
was a large block of English, such as instructions to the customers,
and was printed on a laser printer that handled lower case.
On a Univac machine, we had a chain for lower case and a carbon film
ribbon. But the printer was still lousy with registration and no
matter how much we tinkered the letters never printed quite right for
an attractive reproduction. We had hoped to use print directories.
> the green card (which gave both bcdic and ebcdic mappings for
> character to hex as well as punch hole combinations).
My yellow 370 card has punches, but the white S/390 does not have
punch codes anymore. The S/390 has columns for "Bookmaster symbol",
five EBCDIC codes, ASCII, ISO-8, and IBM-PC.
To this day, in a COBOL or DYL program, using the DISPLAY command with
a signed numeric field may result in an "overpunch", as in printing
36A or 36J.
hanc...@bbs.cpcn.com
> Interestingly, Bristol's PLC equipment besides being somewhat slow* and
> programmed in that weird ACCOL language, have built in support for pulse
> duration signals, which was how they did things before fast a/d
> converters. The transmitting ends I've seen have all been mechanical
> devices. I'm guessing the receive ends were at one time too, but I've
> only seen electronic ones.
>
> The Phone Company must have loved this years ago.
Did the stuff go through a modem or transmit pulses directly?
Years ago the phone company supported telegraph lines in which pulses
themselves, not voice frequency, were transmitted. I don't know much
about them, I think they were known as 20mA and 60mA. These had the
advantage of being very low bandwidth, 50-100 baud. I don't know what
the tarffis were on them, hopefully cheaper than voice grade lines.
In the early days of long distance, they could multiplex a telegraph
line over a voice line.
In the 1960s, Western Union had a national network of private lines
linking businesses together. While we think of them of the public
telegram, that was a small part of their business.
Tim Shoppa
> On Mar 26, 6:48 pm, Philip Nasadowski <nasado...@usermale.com> wrote:
>
> > Interestingly, Bristol's PLC equipment besides being somewhat slow* and
> > programmed in that weird ACCOL language, have built in support for pulse
> > duration signals, which was how they did things before fast a/d
> > converters. The transmitting ends I've seen have all been mechanical
> > devices. I'm guessing the receive ends were at one time too, but I've
> > only seen electronic ones.
>
> > The Phone Company must have loved this years ago.
>
> Did the stuff go through a modem or transmit pulses directly?
>
> Years ago the phone company supported telegraph lines in which pulses
> themselves, not voice frequency, were transmitted. I don't know much
> about them, I think they were known as 20mA and 60mA. These had the
> advantage of being very low bandwidth, 50-100 baud. I don't know what
> the tarffis were on them, hopefully cheaper than voice grade lines.
> In the early days of long distance, they could multiplex a telegraph
> line over a voice line.
In telco language these were known as "dry pairs" (they did not carry
dialtone but were simply wire pairs between point A and point B).
One trick in the industry I was in, was ordering a dry pair for fire
alarm signaling (by tariff far far cheaper) and then using them for
data.
Many cities used electromechanical fire pull boxes which encoded in
bell rings the exact location of the pull box, for details see:
http://www.foxhall.org/preservation/CallBoxHistory.htm
This signalling allowed many call boxes to share the same copper pair.
Tim.
Stan Barr
>Stan Barr (sta...@dial.pipex.com) writes:
>> On Mon, 26 Mar 2007 19:04:40 GMT, Jim Haynes <hay...@alumni.uark.edu> wrote:
>>
>>>Eventually the single case won out (until we get to Teletypesetter) and
>>>was upper case. This makes a lot of sense considering that Morse telegraphy
>>>was still in wide use at the time and is single case.
>>
>> 5-bit Baudot/Murray code was/is u/c only too, of course.
>> Still in use on the wireless in RTTY and TOR forms...
>
>But what's intriguing is that neither baudot or morse code is "uppercase".
>They are both just letters, numerals and some punctuation. I'd never
>given it thought before this thread, but certainly in morse code there is
>no upper- or lower-case specified.
You make a valid point, as does Mr Pfeiffer. I hadn't though of it in
quite that way.
Mr Morse simply specified that "dot-dash" shall represent the letter A,
etc. He did not specify the case. In fact, on the surviving copy of the
"What hath God wrought..." tape of 1844 the letters written under the code
are in mixed case, written by whom, I don't know.
Philip Nasadowski
hanc...@bbs.cpcn.com wrote:
> Did the stuff go through a modem or transmit pulses directly?
Quindar was typically sent over leased phone lines. It's properly
called 'tone telemetry', each discrete signal is given a tone of
frequency X. At the well house, you'd have a rack of transmitter and
recievers all at various phone-friendly frequencies (audio), and ditto
at the opposite end (i.e., the water company's HQ). Needless to say,
you couldn't send too many signals this way. Well flows, etc would be
sent via pulse duration (I think 13.3333 secs was that standard
timebase), and the devices were mechanical in nature.
The most famous application of Quindar is NASA - that's what those odd
beeps you hear at times are for.
> Years ago the phone company supported telegraph lines in which pulses
> themselves, not voice frequency, were transmitted.
These were voice. You could use a dial up line for it.
> In the 1960s, Western Union had a national network of private lines
> linking businesses together. While we think of them of the public
> telegram, that was a small part of their business.
Too bad they didn't embrace data and fiber optics and become a modern
data bandwidth provider. Not like that's overly profitable, but it's
gotta be better than telex these days (do people even USE telex anymore?)
Most water systems, etc, use radios and work at 1200 to 19.2k. Modbus,
etc are very forgiving protocalls. I'm dealing right now with a
multilayered and decent sized BSAP system, and it ocasionaly goes south.
I'm thinking radios, but BSAP is weird, so who knows?
CBFalconer
>
... snip ...
>
> Years ago the phone company supported telegraph lines in which pulses
> themselves, not voice frequency, were transmitted. I don't know much
> about them, I think they were known as 20mA and 60mA. These had the
> advantage of being very low bandwidth, 50-100 baud. I don't know what
> the tarffis were on them, hopefully cheaper than voice grade lines.
> In the early days of long distance, they could multiplex a telegraph
> line over a voice line.
I've run 9600 baud over several miles of 20 mA current loop. The
wave forms were solid and error free. I was nowhere near the
limit.
--
Chuck F (cbfalconer at maineline dot net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home.att.net>
--
Posted via a free Usenet account from http://www.teranews.com
Nico de Jong
"Philip Nasadowski" <nasa...@usermale.com> skrev i en meddelelse
news:nasadowsk-96D8F...@news.verizon.net...
>
> Too bad they didn't embrace data and fiber optics and become a modern
> data bandwidth provider. Not like that's overly profitable, but it's
> gotta be better than telex these days (do people even USE telex anymore?)
>
Nico
jmfb...@aol.com
Could it be more of the case of block letters? Block letters are
never lower case script.
/BAH
jmfb...@aol.com
If you do a traceback, I'll bet the lower case bit flows always
were meant to end up as human-readable such as newspapers or
books or specs or formulas.
I remember people back then having extreme difficulty remembering
the difference between human- and machine-readable w.r.t.
stating product plan goals. Machines can't read human-readable
and visa versa.
/BAH
Charles Shannon Hendrix
jmfb...@aol.com wrote:
> In article <1174855655.0...@n59g2000hsh.googlegroups.com>,
> "mensa...@aol.com" <mensa...@aol.com> wrote:
> >On Mar 25, 2:37�pm, Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
> >> rstei...@visi.com (Richard Steiner) writes:
> >>
> >> > For many types of data input, case might serve an aesthetic purpose,
> >> > but it really doesn't matter in terms of the actual MEANING of the data
> >> > in question. So flight plans, weather reports, etc., are just as good
> >> > in UPPER CASE as in MiXeD CasE, at least most of the time.
> >>
> >> Except that either lower case or mixed case can be read more
> >> accurately and more quickly, because of the greater variation in
> >> letter shape. Lower-case-only printer would have been a better
> >> choice than upper-case-only.
> >
> >Everybody knows this. Any truth to the rumor that the
> >system was deliberately made inefficient just so they
> >wouldn't have to spell "God" in lower case?
>
> I missed this in Joe's comment. Lower case was lousy
> print. Upper case gave you a chance of decyphering the
> blurred character you saw on the page.
That depends entirely on the font used. Plenty of fonts are *terrible* when
they get their uppercase parts blurred, but OK with blurred lowercase.
> Upper case
> block-formatted letters were always easier for humans
> to read. Ask any keypuncher or data enterer worth her salt.
A great many people who had to sit and use those printouts for hours every
day would disagree with you. Depending on the data and what other work it
was being used with, all uppercase could be a nightmare for them.
...and again, a lot depends on the fonts used.
--
shannon | Meddle not in the affairs of Wizards, for thou art
| crunchy, and taste good with ketchup.
| -- stolen from Tom Manos' signature file
Charles Shannon Hendrix
jmfb...@aol.com wrote:
> In article <UV2BGpHp...@visi.com>,
> rste...@visi.com (Richard Steiner) wrote:
> >Here in alt.folklore.computers,
> >Joe Pfeiffer <pfei...@cs.nmsu.edu> spake unto us, saying:
> >
> >>rste...@visi.com (Richard Steiner) writes:
> >>
> >>> For many types of data input, case might serve an aesthetic purpose,
> >>> but it really doesn't matter in terms of the actual MEANING of the data
> >>> in question. So flight plans, weather reports, etc., are just as good
> >>> in UPPER CASE as in MiXeD CasE, at least most of the time.
> >>
> >>Except that either lower case or mixed case can be read more
> >>accurately and more quickly, because of the greater variation in
> >>letter shape.
> >
> >Yes, that might be true for textual data.
> >
> >Most of the data I recall on our application screen displays tended to
> >be numeric values with short alpha headers or prefixes, and those alpha
> >headers derived minimal benefit from the addition of case.
> >
> >>Lower-case-only printer would have been a better choice than upper-case-
> >>only.
> >
> >I would guess that teletype practice was probably derived from radio or
> >telegraph practice. <??>
>
> Nuts. Think about lower case a and lower case o. Think about
> lower case m and n and r all mushed together. You all seem
> to have forgotten that printer output was lousy. It sometimes
> took a TECO command to "see" if that character was a comma or
> a period.
True, a lot of my favorite printers really are bad at lowercase, including
some I used at work as last as 1995.
However, keep in mind sometimes people here aren't assuming a particular time
period.
There are still uppercase only systems today, and you have to wonder why.
> I have the very same problems with these webbing TTY screens
> where on pixel is missing between the rn. I am always interpreting
> it as m.
Unreadable fonts are a pet peeve of mine too.
I've never understood why people who know nothing about fonts want to spend
so much time playing with them.
They almost always end up choosing one that is inappropriate for the
particular medium.
It actually takes some sense and training to pick good fonts and layout when
creating documents, and most people obviously don't have it.
That's why I often question the utility of modern software that puts so much
publishing power in the hands of people who don't know how to use it.
At some point, a lot of software stopped focusing on the job and put too much
effort into presentation.
Charles Shannon Hendrix
"mensa...@aol.com" <mensa...@aol.com> wrote:
> On Mar 26, 6:35 am, jmfbah...@aol.com wrote:
> > In article <UV2BGpHpvSzc09...@visi.com>,
> > rstei...@visi.com (Richard Steiner) wrote:
> >
> >
> >
> >
> >
> > >Here in alt.folklore.computers,
> > >Joe Pfeiffer <pfeif...@cs.nmsu.edu> spake unto us, saying:
> >
> > >>rstei...@visi.com (Richard Steiner) writes:
> >
> > >>> For many types of data input, case might serve an aesthetic purpose,
> > >>> but it really doesn't matter in terms of the actual MEANING of the
> > >>> data in question. So flight plans, weather reports, etc., are just
> > >>> as good in UPPER CASE as in MiXeD CasE, at least most of the time.
> >
> > >>Except that either lower case or mixed case can be read more
> > >>accurately and more quickly, because of the greater variation in
> > >>letter shape.
> >
> > >Yes, that might be true for textual data.
> >
> > >Most of the data I recall on our application screen displays tended to
> > >be numeric values with short alpha headers or prefixes, and those alpha
> > >headers derived minimal benefit from the addition of case.
> >
> > >>Lower-case-only printer would have been a better choice than upper-case-
> > >>only.
> >
> > >I would guess that teletype practice was probably derived from radio or
> > >telegraph practice. <??>
> >
> > Nuts. Think about lower case a and lower case o. Think about
> > lower case m and n and r all mushed together.
>
> Not as bad as when you use an "Arial" font. I often do web pages
> in <tty> font for that very reason (still using lower case). Try
> looking at an all upper case web page. PITA.
>
> > You all seem
> > to have forgotten that printer output was lousy.
>
> Who uses a character printer anymore?
Um... thousands of corporations and an equal or greater number of individual
users?
I'm looking for one right now, because I prefer them for code listings to
laser printers.
> What's important now
> are spreadsheets and their ilk. Bill Gates deserves to be
> kicked in the ass for a lot of things, but certainly for
> MS defaulting to Arial in technical applications like Excel
> and Access instead of the more sensible choice of Courier.
What is Windows?
> Well, at least now you can temporarily switch magnification
> to 200%. At least in Excel. For some strange reason, that feature
> isn't available in Access.
What is Excel and Access?
> Oh, one last thing, I'm the modern equivalent of a keypuncher,
> a database manager. So when I say lower case Courier is the prefered
> font, I know what I'm talking about.
One last thing, I'm the modern equivalent of a keypuncher, because I enter
data and code all day for a living. So when is say Courier sucks, I know
what I'm talking about.
Learn anything?
Charles Shannon Hendrix
hanc...@bbs.cpcn.com wrote:
> As to what looks and works best, it depends on the application and can
> be somewhat subjective.
Finally, someone sees that... :)
> For traditional mainframe work (ie COBOL), I like all caps in a plain
> sans-serif fixed-space typeface.
It's funny, but so do I. I'm not sure because modern compilers don't care.
The last time I edited FORTRAN code, I used uppercase because the programs I
was debugging/changing were all uppercase. The compiler didn't care, but I
found that using it the way I learned it worked better for me, at least at
the time.
> For certain documents, I wish they used numbers styled after old
> typewriter fonts, where the numbers were very distinct; that is, no
> way a 3 or 6 be turned into an 8 because the ends stuck out far.
...and zeroes. I hate zeroes that look like odd shapped ohs (O).
I prefer zero with a slash, but at least give me a dot in it.
> As an aside, one thing that drove me crazy was inconsistent slashing
> of O and O. Some people, including me and my employers, slashed the
> zero, but many others slashed the letter.
To me, that sounds like an error.
I never saw anyone or any typeface that slashed the letter oh.
hanc...@bbs.cpcn.com
> I used to use "Courier New" for everything, to make certain of a fixed
> format font. Lately though, I've been switching to theLucidafonts for
Yes, defnitely, LucidaConsole is great. I use that for documentation--
the basic text is Times New Roman, but things that are directly typed
I use Lucida. I use uppercase for fixed commands and lowercase
italics for the variable operands.
mensa...@aol.com
wrote:
That you don't use Windows.
mensa...@aol.com
It also struck me that someone who is ignorant of Windows,
Excel and Access in fact _doesn't_ know what he's talking about.