And are they still used today?
Telex is an old way of sending text electronically.
It’s still used, even though it’s only 50 baud. Telex is still used in banking, for fund transfer. (I think it’s what people mean when they talk about getting money wired, but I could be wrong there). Telex is used to communicate with people who still have old telex terminals, but no access to email or fax.
Here’s a link http://www.ulltra.com/html/telex.html
and another http://www.amnesty-usa.org/uan.html#telex
telex’s are still used by investment banks to confirm important trades, although this practice has dropped off a lot.
Just a WAG, but still used because there is a clear legal precident regarding transmission and receipt of message. I don’t believe that email, bloomberg messages, faxes, which are all very common forms of confirming stock market trades, have the same legal weight.
Telex was first developed in the 20s and 30 as an automated form of telegraphy but was given the biggest impulse by WWII. Instead of the morse variable length code, it uses 5 bit constant length code called Baudot transmitted at 110 baud. The telex terminals were strictly electro-mechanical. You can find a History of the teletypewriter written in 1963.
By the 60s and 70’s the telex network exploded as corporations and other entities used it. The network is a switched network just like the telephone: you have a physical pair of wires that go from your terminal to a central switching office. To send a message you dial (type, whatever) the number of the receiver. You can get a busy signal, no response etc. When the receiver responds you send your message. You can type it but most terminals had a way of typing off-line and keeping the message in a “memory” in the form of a punched paper tape. Once connected you can run the paper tape to send the message. Communication was billed by time connected, just like the telephone, so you would want to keep it brief.
Those terminal were very ingenious as they had no electronics. Telex only usex Capital letters, numbers and limited signs. One of the signs was the bell so you could start an important message by sounding the bell ding ding ding ding. You can see a Teletype terminal here. Also http://www.nadcomm.com/33asr.htm That terminal has a stand but you could have the terminal on a table or desk and it looked like a big typewriter. On the photo you can see the paper tape reader-puncher, the keyboard and the printer. Teletype was the biggest manufacturer of terminals and those were the last electromechanical models. You can see a much older model here.
IN the 80s several things happened. One was the advent of microprocessors so now computerised telex terminals came out. I remember telex terminal with a screen rather than a printer and no need for punched paper tape for memory. Now you could buy surplus electromechanical telex terminals for pennies so they were often used as terminals (and printers and punched tape memory!) for the first home computers. They were called TTY and you will ocassionally see TTY to mean a printer with text-only capability. Bill Gates and Paul Allen used TTYs in their first projects.
TTY stood for TeleType which was a brand of telex terminals.
With the advent of microprocessors, fax machines were developed as well as personal computers and modems. These enabled users to communicate over the existing phone lines with no need to use a dedicated telex line so the telex network began to shrink.
For people who did not have an in-house telex line there were always offices from which they could send a telex. One great advantage for global business besides having a hard copy for proof was that (like a fax later) you could send it at any time, even if the recipient’s office was closed, and they would get it the next day.
Anyway, that’s my contribution. I know it’s kind of jumbled but I wrote it as I searched the Net and it’s taken quite a while (and I better send it before I lose all as we are having bad weather and power failures these days).
Wow Sailor! That was a great post and not a single pun! I still remember the “chugga chuggs chugga” of the old TTY terminals from 1977 in the old computer science room when I was an undergraduate freshman at the U of MD. The new computer science room had the uber cool terminals with screens that we used to edit our punch carded programs. The DM and hammerbank line printers were for the upperclassmen.
I’m trying (but failing) to remember the old iron they were attached to. It was pair of late 60’s/early 70’s circa warhorses that some manufacturer had donated to the University. The comp sci grad students were thought to be kissed by the hand of God because they got to use time on some PDP mini-computer the University had snagged.
I still can’t believe we got to the moon with less CPU horsepower on board the LEM than my Palm V has. Ah (wheeze) memories!
British Telecom are now using their now mainly surplus telex network as a low cost and lost bit-rate data network for such things as remote telemetry systems and for connecting ATMs to centaral computers. This is ok when you do not need high speed data transfer. It is a good way of using existing infrastructure and equipment.
Thanks astro. Yes, it is amazing that they sent a man to the moon using things like TTYs. I also remember the chucca chucca of the machines. They were extensively used as computer terminals and the paper tape as storage. As programs got larger, the paper tapes were getting to be very long and slow to read. I used a portable, optical paper reader which could read something like 20 times faster than the electromechanical reader. It could read a 70K program in 6 minutes which would take 2 hrs on the TTY. It took the paper tape from a spool and shot it out and you would just let it pile loosely in a big box as it had no receiving spool. Once finished you would rewind it. Paper tape was just not practical for such large “files” as any kink meant the entire tape was useless. I remember audio cassettes being used for a while to store data and programs.
I did look into one of those TTY machines once and it was very ingenious. Pressing a key started the whole sequence. A first “start” bit was sent which put the receiver in motion. Then the five data bits which controlled the print head which was in the form of a cylinder. The head would rise more or less, rotate more or less etc and then strike. Truly ingenious. Then you had a “stop” bit which was just dead rest time. The next start bit would initiate the process all over again. This system allowed the different machines to have a wide tolerance for synchronism as the were re-synchronized at the start of every character. This is was is known as asyncronous transmission. It is slow but it works when you have wide timing tolerances. If I remember correctly the two voltages used were +12 and -12. The paper or carriage did not move. It was the print head that moved across. Changing languages was as easy as changing print heads.
I believe the famous “red telephone” between the White House and the Kremlin never existed as such and is really a Telex link.
I remember in the later years you would have chemical multicopy paper. I happen to have right here on my desk a partially used roll of the original paper and it measures 8.5" wide, just like a normal sheet. It has become totally yellow over the years and don’t ask me why I keep it because I haven’t a clue. It’s good I did not buy a TTY when they were going cheap or I’d still have that too.
Some further explanation for those too young to ever have seen one in operation: If you look at the photo the thing in yellow, to the right, is the operator, clearly distinguishable from the machine itself.
On the left of the machine itself you can see two paper tape units. The larger one in the rear is the puncher. If you activated it, every time you pressed a key it would punch the code for that character on the tape. There were five holes across for the five bits plus one for pulling the tape along. te data were 3 on one side and 2 on the other.
After you had finished composing the message you would cut the paper tape and position it on the reader which is the smaller unit nearer to the operator. Then you would dial the number of the terminal you wanted.
Every terminal had an answerback which it would send automatically when it received an answerback request code. So I would dial xyz12345 and when I had it on line I would send the answerback request. Once I had received the correct answerback, I knew I had the correct tereminal at the other end, then I could activate the paper tape reader and the thing would start chugging along non-stop until the end of the tape.
Then you’d do another answerback just to be sure and then disconnect.
If you wanted to chat with someone in real time you could use the bell to call someone’s attention. (Excuse me, are you the guangding bycicle factory or the guangding jackhammer factory?) It was fun to see most of the operators typed even much slower than I did.
The expense of having a separate data network from the telephone network is what made telex expensive and speeled its doom. The modem killed the telex. Without modems there would be no fax or data over voice lines and the telex network would have probably grown exponentially in extension, speed etc.
And, finally, a true anecdote of my childhood: I was probably about 9 when one of the boys in class came to us and said he had been to his father’s office. He said he had seen a machine like a typewriter which, when you pressed the keys, it would type in a similar machine somewhere else and the person at the other machine could do the same and you could communicate. We all laughed so hard at him Did he really expect us to believe that? Did he really think we were so stupid? We all knew it was impossible and, even if it were possible, why would grown ups want a toy like that anyway? We made fun of him on that account for a long time. Poor kid. Sometimes it is better to shut up even if you are right.
A few days ago, I rented Thirteen Days – the movie with Kevin Costner (I think, I always get that actor mixed up with Nicolas Cage, but that’s really neither here nor there …) about the Cuban Missile Crisis. At one point, the folks in the White House are reading a communication from Moscow, and even though it is clearly NOT a fax, my mind just translated the image into “fax” so I could move along with the plot. It looked almost as if they were assembling it – short snips of a message put together on a regular size sheet of paper.
If anyone has seen the movie, AND knows what a Telex message looks like, was that an accurate way to show it?
Also, those stock tickers … when they were in remote locations, were they using Telex technology? Lord help me, the only example I can think of is when Gomez on the Addams Family would check his stocks, and you would see the machine spitting out thin paper strips that would pile up at his feet.
delphica, I have not seen the movie so I cannot comment but telex was printed on a paper roll the same width of a letter and just cut to whatever length so it could be very short if it was just one line, about the size of a page, or very long as the only limit was the length of the roll of paper. I have handled lists which were very long (I kind of miss continous paper printers as you could do certain things).
So, the paper format was not very characteristic, what was more characteristic was the printing itself by the font, spacing etc. I believed they printed 10cpi or 80 char/line
By the way, I forgot to mention the later telex machines printed what you sent in black and what was received in red so it was easy to distinguish.
Gosh, if I dig I can probably find some original telex messages somewhere. I’ll see if I can find any
Many thanks for these excellent, detailed responses!
I worked in the Telex industry (for a discount Telex forwarding service) for about a year in the '80s, so I’m somewhat familiar with it. Sailor’s response is pretty much on the money, although I believe that the Telex network ran at 50 baud (as mentioned by Tansu at the top of this thread). There was another similar network called TWX that ran at 110 baud; I’m actually not sure how these networks differed, although TWX numbers looked just like phone numbers (i.e. 10 digits). I believe TWX may have been restricted to the US.
Since it was the 80s when I worked there, the equipment was quite a bit more sophisticated than the ancient ASR-33s talked about above. We had a VAX 11/750 connected to specialized communication hardware that allowed us to dial out on hundreds (I don’t remember how many) of lines simultaneously; we also had a sister office in the UK and another in Suriname where we batched our overseas telexes.
Telex, as opposed to Teletype, was Western Union. I worked as a graveyard shift hotel person in Austin, Texas in the 1970s and our hotel became the WU office when the real one shut down for the night. And I had a Telex machine to use (we also filled in for WU on weekends, so I occasionally got to wire $20,000 to someplace in Mexico on behalf of Margaritaville sorts - we had some cash limit that I don’t remember).
The one I used did not use the paper strip codes that sailor mentions. It looked like the 1958 Cadillac version of a typewriter, complete with chrome doo-dahs. I did have to type up messages without seeing the output, and then I sent them and learned my mistakes later. Connection was hit-or-miss. One message could occupy me for hours, until my employer (transplanted Brit) showed me the good old kick it on its back “reboot.”
One thing I did learn at the time was that WU messages were universally regarded (in the absence of anticipated money orders) as harbingers of bad news. Very poor selection of a method to ask a co-ed out to dinner.
Telexes are still widely used in commercial shipping, particularly amongst chartering brokers. We still have one in our office, but we are getting to the point it is not worth doing so because we only get a couple of telexes a year, often from ships at sea.
The advantage of telex for chartering brokers is the answerback code described by Sailor. When you are doing a deal worth millions on the strength of a few quick messages, you really want to be sure you are talking to the person you think you are talking to.
Because the answerback codes are controlled centrally by the telecoms coy, you can’t falsify who you are. And because you can ask for an answerback at the beginning and end of transmission, you can prove not only that a message was sent, but also that it was received by whoever you sent it to.
Unlike the reply codes on email, or a fax headers, which are easily faked.
A telex gave a certain level of proof in its day but today it would be trivial to fake. A fax also is still considered to be quite reliable and it is also quite trivial to fake by anyone with a computer. Photos are also something which were considered good proof and will probably soon be considered quite unreliable as manipulating them is also becoming quite trivial.
The trend will go to electronic signatures.
The USA was the only country with several telex networks as in other countries they were part of the government’s PTT. I remember TWX (which I believe was ATT), Western Union, and I think there were another one. Anybody remember?
I also remember telex directories which were books just like phone books.
In the late 80s, before email as we know it, some private electronic messaging services appeared like Compuserve and MCI mail. Obviously you could send messages to others on the same network but you could also send and receive to telex terminals. I still have old business cards with a “telex” number listed as MCI-650-337-4537. A telex sent there would go to my mailbox and I would retrieve it as an email and I could send messages to telex machines. This was extremely useful.
For a while these electronic messages had to be compatible with older technologies. Using them I could send messages to a telex terminal and I could send a letter by mail. Rather than mail the letter from where I am I could send it via MCI who would print it and mail it by regular mail from their closest center. That would shorten the delivery time. You could also send to a fax machine. All these different electronic messaging systems have either disappeared (MCI mail) or become regular email (Compuserve).
How? I’m not saying it can’t be done, but how would you fool your victim into thinking that a telex they had received was from Party X when it was actually sent by you, without hacking into the relevant telex company’s system? Given that as I understand it the answerback codes are centrally controlled (or is this wrong?), how would you make Party X’s answerback show up on the victim’s machine when it sent the request code at the beginning and end of the message?
Compare that to a fax, where all you have to do to fool the victim is buy a fax machine the same make as Party X’s machine, then set up the header to be the same as Party X’s.
Obviously encryption and electronic signatures are even more secure, but unless my understanding is entirely wrong, telexes are much more secure than fax.
I said telex used one pair of wires but now I am in doubt and I am not sure if it used two pairs.
THIS page has instructions (near the bottom) on how to send a telex using MCI mail. It was published in 1987 and provides some insight into the telex networks (including the old scam of sending fake bills like they do with the telephone yellow pages).
When sending a telex abroad you had to dial first the country code, then the telex number. When sending a telex from abroad to the USA (IIRC) you dialed the country code, then the network code, then the telex number… I am a bit hazy on this…
By chance I found Report on CCITT Meetings on New Data Networks 2327 November 1970 at Geneva. An interesting document where countries discuss the transmission of data and ARPAnet is mentioned
In Baudot, characters are expressed using five bits. Baudot uses two code sub-sets, the “letter set” (LTRS), and the “figure set” (FIGS). The FIGS character (11011) signals that the following code is to be interpreted as being in the FIGS set, until this is reset by the LTRS (11111) character. WRU (who are you? get it?) is the request for the answerback transmission.
This reminds me that, unlike a typewriter, in telex you had to send separately a CR and LF. It would not do it automatically and if you did not do it, the machine would just overtype the last character on the line. A LF would just advance the paper but not return the print head home and a CR would return the print head but not advance the paper. This is the origin of the separate CR and LF commands which later appeared on compatible screen terminals like VT100.
This page has a good history written in 1990 when telex was still quite prevalent.
It seems there were more modes and standards than I was aware of. While Western Union used a dedicated data network, ATT used modems and phone lines. This is turning to be quite a broad topic.
Princhester, I was actually thinking in terms of genrating the printed piece of paper you could show as proof that the other party said X.
As for your scenario, ie. sending a telex to someone and inducing them to believe it came from someone else, I do not remember well but I believe the only identification came from the answerback programed into the machine, not from the switching equipment. If this is true, then all you had to do was change your answerback to whatever you wanted, just like a fax terminal.
Yes, I am quite sure the answerback was sent by the terminal itslef as you had a key you could press to send your answerback to identify yourself. From that point of view faking a telex should be about as difficult as faking a fax. I suppose banks and other users who needed security had another level of codes and encryption to assure the needed level of security.