(This is just a quick sketch, without proper references. If you're really interested in Teletypes, you should consult some of the much better sources listed in the Bibliography.)
"Teletype" was a trade name for a particular brand of printing telegraph. The encoding and transmission technology it employed became the dominant technology for printing telegraphs, and so the term to some extent also became a generic term for printing telegraphs from other manufacturers (Kleinschmidt, Siemens, Lorenz, etc.) The Teletypesetter was a variation of the Teletype which added capabilities for the control of typesetting/linecasting machinery (Linotypes).
(A 2010 search of the USPTO records indicates that all trademark registrations for "Teletype" are dead. The original registration was by the Morkrum-Kleinschmidt Corp. from 1925 (first use in commerce: November 1921). The familiar "Teletype T" graphic Teletype logo was registered in 1978 (first use in commerce: January 10, 1962.))
As it is in essence a telegraph, the Teletype operates over circuits. That is, while it is in operation there is a single continous connection between all teletypes involved (this could be an actual dedicated wire, a switched circuit, or a radio link). This style of operation may be distinguished from the "store and forward" operating technologies employed by packet switching networks such as the Internet and the ancient Roman postal system.
The teletype was successful in large part because the way in which it used this circuit was extremely robust and required no overall time-standard to synchronize the possibly widely distributed machines. Very briefly: A circuit/line could be in one of two states, called (for deep historical reasons) "marking" and "spacing." An idle line was maintained as marking. Each teletype on the line was equipped with a motor which ran at approximately the same speed (although of course never exactly). Each character was transmitted entirely separately. To start transmitting a character, the sender forced the line into a "space" state for a predetermined period of time. Then during each successive unit of time (5, 6, or 7 units, depending on the model of teletype) the sender forced the line either to "mark" or "space." The receiver, operating at approximately the same speed, could sense for each time-unit whether the line was marking or spacing. It could therefore decode these states into the encoding of a single character. The character was terminated by a longer-than-normal marking state on the line (1.42 times as long as normal). The line was then in a state where it could either be used to transmit another character or remain idle. Because this was serial transmission where each character established its own starting time, the system required no overall synchronization or time-standard. (It was "isochronous" rather than "synchronous.") This is exceedingly reliable.
(I might argue that teletype technology is the third most reliable telecommunications technology ever devised. The most reliable is Morse code (or some equivalent) controlling the on/off state of some arbitrary medium (the current on a telegraph line, or a shining signal light, or an otherwise continuous radio wave, or whatever else might be handy). The second most reliable is Morse over a single physical wire. There are of course issues with making the wire work (just ask any Civil War re-enactor trying to lay a telegraph line using 1860s methods), but it's hard to get a simpler electrical connection than a wire. (The 18th century optical telegraphs were parallel, not serial, devices and had their own reliability issues.) The third most reliable technology is Teletype-style start-stop signalling over such a telegraph wire. Anything fancier than this, while efficient when it works, is just asking for trouble.)
The history of the Teletype (and of compatible equipment by other companies) is fascinating, but far too involved for this brief introduction. Very quickly, its origins go back to the first decade of the 20th century, and are associated most strongly with the engineers Charles and Howard Krum (backed by the Morton family, of Morton Salt fame) and Edward E. Kleinschmidt. The Mortons and the Krums formed the Morkrum Corp. Morkrum developed start-stop isochronous operation around 1919. Morkrum merged in the 1920s with Kleinschmidt's company to form Morkrum-Kleinschmidt. This soon changed its name to that of its product, becoming the Teletype Corp. In 1930 it became a subsidiary of AT&T's Western Electric; it dominated the market, but vanished into the corporate aether in the late 20th century. Kleinschmidt restarted a new Kleinschmidt Labs in 1930, which continues to this day as Kleinschmidt, Inc.
Most Teletype equipment in the early and mid 20th century employed "5 level" encoding (5 bits, in modern terms). Even with a "shift-in"/"shift-out" capability, this restricted printing to uppercase. The 15 Type and 28 Type equipment was 5-level. The requirements of the Teletypesetter led to the introduction of 6-level encoding (giving uppercase and lowercase). The 20 Type printer was a 6-level version of the 15 Type. There were also Teletypesetter-specific machines, such as the TTS Perforators. Later machines employed 7-level code, which permitted full ASCII.
Teletypes used various printing technologies through their evolution. The 15 Type equipment, for example, used a type-basket resembling a standard mechanical typewriter. The 28 Type equipment used an interesting "type box" arrangement with a single striker. The 43 Type equipment (much later) was dot-matrix and not nearly as much fun.
Especially with the early equipment, the system was modular. Thus, although it is common to speak of a "Model 15," it is really necessary to distinguish what capabilities this particular Model 15 had. For example, it could be a Model 15 KSR (with a keyboard, capable of sending and receiving) or a Model 15 RO (no keyboard, receive only). The keyboard on a Model 15 KSR was easily removable, transforming the KSR unit into a RO unit. One might also have a quite different unit, such as a Model 15 TD (that is, a standalone tape reader which functioned as a Transmittor-Distributor to read tape and send it out over the wire). The Teletype Corp. documentation often refers therefore to "N Type" (e.g., "15 Type") equipment, a usage which seems awkward at first but which makes sense as one realizes just how modular it all was.
From an operational point of view, Teletypes may run over a dedicated or leased line, or they may run over a switched line which is (for the duration of use) switched so that it is used entirely by the Teletypes. You could lay your own line (railroads tended to do this, as did the military) or you could lease lines or time on switched lines from telegraph/telephone companies. (The AT&T "Teletypewriter Exchange" (TWX) service was provided using switched lines.) You could also, of course, run the "line" over the radio (RTTY).
Users included pretty much anyone who needed to send printed text: the major wire services (AP, UPI, etc.), railroads, the military, private companies in-house, etc. To the best of my knowledge, however, all commercial operation of teletypes ceased in the late 20th century.
During the later years of their operation, teletypes were used in two important transitional applications (often by amateurs). First, they were used by radio operators for "Radio Teletype" use (RTTY). This became a strong subculture of the amateur radio community. So many TTYs which left commercial service went first into amateur radio RTTY service. Second, they were often used as input devices for early computers, both commercially and by hobbyists. (The second computer I ever used was a PDP-8 through a TTY-33 (uppercase ASCII) and I spend a lot of time at TTY-43 (upper and lower case ASCII) on a VAX in college.) If you were building a computer in, say, the early 1970s then a TTY was probably your best bet for a terminal.
(2) ITTY. Internet TTY News. George Hutchison, W7TTY, has devised a 7/24 Internet-based news service designed for TTY output. So you can hook your TTY up to your computer and copy today's news. See: http://www.rtty.com/itty/index.htm Also, search YouTube for "teletype itty" to see video of various Teletypes copying ITTY news.
(6) Re-create historical operations. Just as US Civil War re-enactors routinely re-create 1860s telegraph field operations, so also various museum and re-enactment groups have to a greater or lesser degree re-created particular historical uses. The efforts of Vietnam-era veterans to simulate TTY-based communications centers of that era are particularly impressive (although I haven't followed these activities closely enough to know the degree to which they are employing physical equipment as opposed to virtual simulations).
(7) Provide Teletype messaging services today. John Nagle of The Aetheric Message Machine Company, Ltd. is doing just this with a "Steampunk"-ified Model 15. "... to bring text messaging to the Victorian era." Splendid.
Teletypes are a secondary interest of mine. Linotypes (and related machines) are my primary interest. So I started collecting Teletype equipment in support of my Linotypes. Since Teletypes were used historically/commercially to support Linotypes in two ways, I'm doing this in two ways, as well.
First, I'm trying to set up a "private line" telegraph/teletype circuit so that I can send (fictitious) news stories in from the "wire service" (well, down in my office) up to the receiving newspaper operation (well, the Linotypes up in the garage). Right now my Linotypes are not equipped for Teletypesetter operation. So to do this I'm using "ordinary" 5-level Teletype equipment. The stories at the receiving end are printed on the Teletype there, from which point they would be edited, retyped, and then typeset on the Linotype.
This will also require setting up a conventional manual telegraph line with a sounder in the printing shop, as manual telegraphy was used to a surprisingly late date in the newspaper industry - until the 1930s commonly, and in some cases until the 1940s. This is not a problem, since I'm wiring the house for telegraph anyway.
All portions of this document not noted otherwise are Copyright © 2010 by David M. MacMillan and Rollande Krandall.
Circuitous Root is a Registered Trademark of David M. MacMillan and Rollande Krandall.
This work is licensed under the Creative Commons "Attribution - ShareAlike" license. See http://creativecommons.org/licenses/by-sa/3.0/ for its terms.
Presented originally by Circuitous Root®
Select Resolution: 0 [other resolutions temporarily disabled due to lack of disk space]