Telegraphs, Phones and Teleprinters

This is a fairly brief history of the telegraph. For more information see the websites: Porthcurno Telegraph Museum Victorian Internet

Telegraph

The telegraph was the first wave of Information Technology or IT as we now call it.   Subsequent waves might be:

  • Telephones  The first consumer-oriented gadget, expanding from 1876 to the 1990s.
  • Punch cards  and data-processing machines from the 1890 to the 1960s
  • Mainframe computers  from the 1950s to the 1990s.
  • Microprocessors,  Personal Computers and Datacenters from the 1980s on.
  • Entertainment,  the Web, games and TVs made affordable by Microprocessors.
  • Mobile computing  becomming practical now.

Each new wave overlaps and builds on the legacy of the last.

Contents of this page:

Telegraphs are interesting in their own right.   A telegraph is nothing more than a battery, switch and sounder seprated by a long wire, so it is something children can make.   Telegraph machinery is no longer common outside museums - modern network equipment makes better use of the cables. Network equipment deals with the same issues of code formation, duplexing, cable attentuation, noise and reactance that Victorian engineers discovered. The advantage of old equipment is that problem and solution are often visible in copper coils and brass dials. New equipment may be working a billion times as fast, but all that is visible is another plastic chip.

From a computing point of view telegraphs are an ancestor of the communication codes we use, of printers and of the Internet.   A great deal of what we now use computers for was done by comparatively simple electro mechanical machines.   Some of the assumptions we still use arose in the time of the telegraph. Character-codes, text abbreviations, print and business languages are not entirely new, they have histories stretching back to the telegraph.

  • The "Victorian Internet" thesis suggested by Tom Standage is popular . The methods used and the hopes and fears raised by the telegraph find many parallels today. We may actually be less flexible about reorganising ourselves to use machines now than they were a century or more ago when the whole idea of automation was new.
  • Telegraph inventors, like Internet entepreneurs today have a strange variety of backgrounds. Samuel Morse was an artist, other inventors were diplomats and a notable number were medical men. There often is a connection to academia, Cooke would not have succeeded without Wheatstone or Morse without Leonard Gale. The combination of business boom and academia resembles IT today. High speed telegraphy needed the scientific skill of Oliver Heaviside, but he was innitially ignored by bureaucrats - another familiar story.
  • Character coding methods devised for the telegraph are still in use. ASCII was a printing telegraph code and still underlies much of what is written - this web-page for instance. The absence of a few characters from ASCII necessitated markup languages. It turned out that teleprinters made affordable computer terminals. Teletypes were still used this way into the 1980s. Other coding schemes like ‘ Unicode ’ were forgotten, whilst ideas like the Q-codes have yet to reach their full potential in business languages.
  • Simplicity versus ease of use has always been a problem similar to what we have today. Morse code had to be learned but was then fast and efficient even with limited equipment. The Wheatstone ABC was a more complicated instrument and not as fast or flexible, but could be used by anyone. Todays computers can be used by anyone, but the simplifications that make that possible get in the way of better programming.
  • Government has always been involved. Innevitably it is a big customer and has power over wayleaves in the public highway and airwaves. Early telegraph systems competed - unless governments nationalised them as a means of censorship, to avoid civil insurection or as a way to guarantee reliable service. Communication has a natural tendenct towards monopoly which democratic governments usually limit. The telegraph industry had a natural tendency towards monopoly. British Conservative governments nationalised telegraphs in 1869 and telephones in 1912 and they remained that way until 1984.

There are a great many histories of the telegraph, many of them marvelling at what could be done with bits of metal and mahogany. That is part of the theme here.

There are some odd questions as well.

  • The Romans (and Greeks) had semaphore telegraphs two thousand years ago. The idea was forgotten. In the Norman conquest England was covered with new castles on hills but no telegraph -the best they could do was light beacons. The idea of semapore on land wasn't to recurr until 1790.
  • Both the telegraph and the telephone seem to have been re-invented several times before the idea eventually took hold. Social and legal convention has it that there is a single inventor who deserves the honour of a money-making monopoly for several years. The history of telegraphy (and most electronics) suggests in contrast that there are often several near-simultaneous discoveries and who gets the money and the honour is a lottery.
  • Telegraphs aren't possible without batteries, wiremaking and aptitude in metalworking. The jewellers and ‘ toymakers ’ of Birmingham had that, but didnt invent the telegraph. At some point an electric telegraph becomes innevitable but springs from the minds of artists, scientists and surgeons, not the practical men of the day.
  • Telegraphs are quickly displaced as a use of wires by telephones. Telegraphy doesn't dissapear, it is overshadowed by new technology but continues to grow and finds new uses as a more formal but flexible business communication. It is only the fax and email that finally kill telegraphy.
  • Telegraphy seems to have a huge economic impact. Businesses, banks, newspapers and even large parts of government were regional in the early 19th century and national by the 20th century.
  • The "Productivity Paradox" suggests that despite their dramatically greater speed and flexibility today's IT systems aren't necessarily delivering very much more than the telegraph did. Mobile phone use has grown dramatically since the 1980s, when business executives, politicians and journalist would argue that they were essential tools. So whay has the economy not grown rapidly. Could more communication ever be a bad thing?

What follows below is the usual ‘ gallop ’ through a sequence of discoveries, each partly building on the last in making telegraphy more affordable or easier to use. Potted histories of this kind sometimes seem to imply innevitable progress.

Once a copper wire has carried a message from one city to another and the usefullness of the idea has been demonstrated by carrying election results, arresting criminals fleeing justice and enquiries about the sailing of vessels and the price of fish something innevitable seems to set in. If a line can run from Philadelphia to Washington then Pittsburgh, New York and Boston must follow. If the line is busy it must be doubled, then duplexed and quadruplexed. It must be possible to devise printing telegraphs. At some point someone will realize that a telephone, a wireless telegraph and computers are practical and all the technology of today rise from it.



Historical Span

Telegraphs were invented around 1840 and are the oldest electrical devices. There are still some telegraphs in use although they were replaced to some extent by the phone, more particularly the fax and proprietary computer communications and then by the Internet quite rapidly from the 1970s onwards.

Early telegraphs used Morse Code and it remained in use in aviation, shipping and military circles right through to the end of the 20th Century. Morse is still used for some kinds of audio signalling, for instance in air navigation beacons. Some people use Morse as a ring-tone to signal who is calling. It could be use to read texts - and indeed to send them. Morse code could still be useful.

In the 20th Century printing telegraphs became more common. The ASCII code originated as a printing telegraph code and is still the basis for most of the text held in computers.

There are just a few fragments of actual telegraph systems left working. For instance the telegram has a status in Japan so the system is retained, printing greetings on special paper. The UK's telegram service was spun off as "Telegrams Online" in 2003 - it too provides a greeting service. People still talk of getting a "Telegram from the Queen" on their hundredth birthday or Golden wedding. There is a demand for "steam punk" or geek retro ways of doing things - just as there is for valve amplifiers.

The landlines and trans-oceanic copper cables that made up the telegraph and telephone systems have been replaced by optical fiber. Morse keys and sounders were largely replaced by printing telegraphs and then by internet routers and screens.

There are still some big bits of infrastructure from the days of the telegraph in use - the wayleaves. All those ducts and poles carrying cables were re-used for phones and the Internet.


Early Telegraphs.

Before the telegraph, information moved at the speed of a horse on land or a ship at sea.

There were a few ways to signal over distances. The Romans used semaphore flags to signal between forts but the idea was largely forgotten.

News of battles took days to travel across Europe and weeks to cross the Atlantic. Sometimes the first people knew of an invading army was when it was sighted. A thousand men died at the Battle of New Orleans (1815) because neither army knew a peace treaty had been signed weeks earlier.

The idea of a beacon, a fire on a high hill, was around. The problem is that becaons really just communicate alarm, they don't say who is doing what. Refined to smoke signals fire might be a bit of a better messenger. In a windy place like Western Europe smoke signals won't work well.

Around 1790 Claude Chappe invented telegraph towers, a windmill like structure, but with semaphone arms instead of sails. The telegraph proved important to Napoleon in his conquest of Italy. The British Admiralty built a similar system, as did the Prussians and Swedes. The problem with all these schemes was that manned telegraph towers were expensive to maintain and the systems failed on a foggy day. If the Americans had a telegraph, however, those men would presumably have lived.

Several early electrical telegraphs were devised. Stephen Gray built an electrostatic telegraph using thin iron wire, silk thread and a brass electroscope in 1729. de-Salva used paper insulation for the wires in an electrostatic telegraph in 1795.

In 1816 Francis Ronalds demonstrated a scheme using static and clockwork dials on 8 miles of wire in glass tubes. He proposed the Admiralty adopt it, but got the reply Telegraphs of any kind are wholly unnecessary, no other than the one in use will be adopted . Inventors often have their high hopes of government dashed. Ronalds was knighted in 1870.

Baron Pavel L'vovitch Schilling, a Russian Diplomat built a working telegraph in his apartment in St Petersburg in 1832. It seems to have been the first to use binary signalling.

Harrison Gray Dyar and his brother Joseph laid a half mile wire through their home town of Concord Massachusets in 1826 and then along the Long Island racetrack in 1827. They succesfully transmitted messages that were recorded by sparks acting on on spools of dampened litmus paper. However Dyar was apparently threatened with prosecution for "Conspiracy to send secret communications in advance of the mail". Dyar moved to Paris and only returned to America in 1858. Samuel Morse's first wife Lucretia was the sister of one of Dyar's associates, it is not clear whether Morse knew of the experiment. Morse's methods do seem similar.

Cooke & Wheatstone

William Cooke was an ex-army officer and son of a surgeon. He probably saw Ronalds' telegraph, the families were friends. He had also seen Schilling's needle telegraph demonstrated at a lecture. He devised his own mechanism but needed a long wire to test it.

Cooke was introduced to Charles Wheatstone by Peter Mark Roget, secretary to the Royal Society and a noted physician, writer and lexicographer. Roget remains famous, he created "Roget's Thesuarus" which groups words by meaning. He also invented the log-log slide rule allowing fractional powers and roots to be calculated easily.

Wheatstone was an established inventor who had made several new devices for measuring sound and invented the English concertina. His successes lead to him being appointed Chair of Experimental Physics in King's College London in 1834. However his attempts at lecturing were a failure, he became tongue tied when asked to speak in public. Wheatstone had his own ideas for a telegraph - and a four mile circuit. Although he could have pressed ahead with his own ideas he was impressed by Cooke's "zeal, ability and perseverance" so they agreed a partnership and in 1837 took out a joint patent. It is sometimes suggested that Wheatstone was a bit of an academic snob and regarded business as beneath him. That seems unlikey, he and his brother inherited their uncles music shop. It seems that Wheatstone enjoyed the science but had no great throught of commercialising it.

Cooke and Wheatstone developed several telegraphs using one or more wires. Their first patent filed in 1837 on Improvements in Giving Signals and Sounding Alarms in Distant places by means of electric currents transmitted through Metallic Circuits .

As the title suggests, inventors were not sure quite what the aim of a telegraph truly was. Cooke envisaged a system of alarms in streets communicating with fire stations. He intended to lay a brief statement of this before the government, envisaging a 'complete electro-galvanic civic, milito-police system.   Two centuries later people still have Messianic visions of what might be accomplished using the Internet.

After some initial setbacks Cooke demonstrated a Telegraph to the GWR and got the support of Brunel. In May 1838 they laid a line from Paddington to West Drayton, 13.5 miles away. Brunel laid the 6-way cable in iron pipes and this may have made the scheme expensive. This was one of the first underground wires and a section failed, so the line may have ceased working. When it came to extending the line one of the GWR directors, Mr Hayward of Manchester denounced the invention as new fangled and blocked the work. Cooke got permission to extend the line to Slough, then to Windsor if he maintained it.

The five needle (and 5 wire) telegraph would almost spell out alphabetic phrases one letter at a time - but the design allowed for just 20 letters so C,J, Q, V X an Z needed substitutes. Cooke & Wheatstone devoted a lot of effort to making simple telegraphs that didn't need skilled operators. However the five needle system used to West Drayton proved too expensive and Cooke substituted a two-needle scheme instead on the line to Slough.

Rather than burying the cable as Brunel had done Cooke also came up with the idea of a line of poles with the wires hung from pottery insulators. Much the same idea as used by Cornell in the US.

Cooke & Wheatstone patented their first ABC telegraph in 1840 and later versions of this were widely used by the British Post Office. Where the US telegraph normally expected operators skilled in Morse the British telegraph often used more expensive instruments and less expert staff. (below)

Edward Davy

Edward Davy was a surgeon by profession and began experimenting with telegraphs in 1835, unaware of what others were doing. In 1837 he got permission to lay a mile of wire around Regent's park and also demonstrated his invention at the Belgrave Institute and Exeter Hall in London. In some ways Davy's equipment may have been superior to Cooke and Wheatstones, he used 6 wires which allowed over 200 simple and compound signals according to Wheatstone. He also had a printer using electro-chemical marks on a calico strip wound over a revolving copper cylinder and an "electrical renewer" which subsequently became known as a relay.

When Cooke and Wheatstone applied for patents in May 1837 Davy opposed. He had lodged a caveat and deposited a sealed description of his own invention with the Society of Art.

Davy gave up working on the telegraph for personal reasons - he had mounting debts and his wife was trying to divorce him. In April 1839 he set sail for Australia where he edited the Adelaide Examiner for three years, then ran a copper smelter in Yatala and settled down to running a medical practice.

Relays allow long lines to be worked without excessive voltage. They also allow a small signal to engage some circuits whilst disengaging others. Relays became commonplace in early data processing equipment and phone exchanges used several to monitor each line - thousands in a system. A hundred years later some of the first computers were constructed using relays and they are still commonplace today as switches in electrical equipment. Edward Davy was trying to build a telegraph; he invented one of the cornerstones of electronics.

William_Alexander

William Alexander likewise opposed the Cooke and Wheatstone patent. He had experimented with a telegraph on a four mile circuit at Edinburgh University. He apparently withdrew, acknowledging the superiority of Cooke and Wheatstone's plans. He went on demonstrating his own telegraph apparatus however.

Railways & Telegraphs

Telegraphs were becomming necessary to railways. Railways sometimes needed telegraphs where tunnels and obstructions prevented a clear view of the track. The idea of railway signalling was just emerging as the problems of higher speeds became clear. Railways also had a long wayleave between towns that telegraph inventors could see the value of. In 1837 Carl August von Steinheil demonstrated a telegraph over seven miles of railway track in France. In the same year Samuel Morse gave his first public exhibition of his work.

Railway investors were often not great supporters of telegraphs, seeing them as an unnecessary expense. In 1842 Cooke published "Telegraphic Railways or the Single Way" suggesting that double track working was often unnecessary and the expense of those two tracks could be saved by a system of block working. The whole system of double way, timetables and signals is a vain effort to attain indirectly and very imperfectly, at great cost, safety from collision . On the whole British railways went on constructing double track systems and there was some resistance to the idea of block working. Nevertheless by 1844 Cooke & Wheatstone had about 550 miles of telegraph line in use and were beginning to receive royalties on their patents.


Samuel Morse

Samuel Morse conceived an electric telegraph scheme perhaps earlier than 1832 but a practical demonstration waited for ten years. Morse was an artist of some repute, he painted several prominent Americans like John Adams and James Monroe. In 1825 he was in Washington painting a portrait of the Marquis de Lafayette. A horse messenger bought Morse a one line note from his father "Your dear wife is convalescent". Morse rushed home to New Haven but found that his wife had died before he had even received the message. Morse vowed to find some means of long distance communication.

In 1832 during a sea voyage Morse saw a demonstation of electromagnetism and recognised its potential for communication. Morse's initial scheme involved a sender with metal pieces a bit like printers type arranged in a row to make and break a contact as they were pushed across it. At the receiver the pattern of dots and dashes would print marks by pushing a pencil up and down as a strip moved across. Early ideas were too elaborate.

Like Cooke, Morse found difficulty making a telegraph work over a long line. He found help from Leonard Gale, an academic with more knowledge of electromagnetic research. Like Davy they developed relays.

Alfred Vail, whose family owned an iron works helped make and finance equipment.

Like other inventors Morse had difficulty finding finance for a project that most people did not really understand. Morse demonstrated the idea to Congress in 1838 - they were considering a system of telegraph towers like Chappe's. (above) The Congressmen were apparently incomprehending - the demonstation used a long reel of wire but across a short distance, so they could not see the point.

Four years later in 1842 Morse was more succesful in a demonstration to Congress. The wire ran between committee rooms so the action at a distance was more evident. There had also been several demonstrations over fairly long wires (and considerable lobbying). Congress funded an experimental line from Washington D.C. to Baltimore (although they nearly invested in Mesmerism instead). The line was completed in 1844 but by then its utility had already been proven when it transmitted the results of an election in Baltimore from a halfway point to Washington before any other messenger could arrive.

Ezra Cornell

Once people grasped the idea of a telegraph it seems to have inspired them.

Ezra Cornell purchased some rights in a patent for a plough and the rights to sell it in Maine and Georgia. He came across an aquaintance puzzling over some plans for a "scraper" intended to bury telegraph lines in the ground in a lead pipe. Cornell devised a special plough that would cut a 2½ foot ditch, lay the pipe with the wire and cover it back up. Unfortunately the technique didn't work well at the time because condensation in the pipe tended to ruin the insulation. However Cornell won the trust of Morse, he invented the idea of stringing the line from glass insulators and built the Washington DC to Maryland line. He subsequently made a substantial fortune as one of the founders of Western Union.

The Morse telegraph proved a great success. After puzzling over the problem for a decade he, Vail and Gale got the technology just right, or perhaps the time was right. Within a decade there were lines all over America and Europe as well. In 1861 a telegraph line spanned America to the Pacific. A telegraph line across the Atlantic worked in 1958 but failed a month or so later. Several other attempts were made to replace it and in 1866 a connection finally lasted.

Pirates

Morse became aware of what others were doing as his work proceeded. He was angered by their claims:

"I have been so constantly under the necessity of watching the movements of the most unprincipled set of pirates I have ever known, that all my time has been occupied in defense, in putting evidence into something like legal shape that I am the inventor of the Electro-Magnetic Telegraph!! Would you have believed it ten years ago that a question could be raised on that subject? "

Morse had reason to be rather put out. His demonstration triggered the creation of a big industry but telegraph companies routinely ignored his patent or contested it. In 1853 the case came before the US Supreme Court. The court concluded that Morse had been the first to combine a battery with electromagnetism to make a practical telegraph. European governments recognised that something should be done to recognise the invention and awarded Morse 400,000 French Francs - about $80,000 - for his contribution.

Although Morse's patent won overall recognition the majority felt that the claims to rights on any "use of the motive power of the electric or galvanic current ... for marking or printing intelligible characters, signs or letters, at any distances. ..." were too broad.

Perhaps it is innevitable that an invention which allows information to flow freely makes inventors aware of one another and brings about claim and counterclaim. A good invention is usually obvious, it's just that nobody -much- had thought of it before.

Morse deserves honour for his work, he was certainly a pioneer and over time developed a highly workable communication system. He used a single wire where Cooke & Wheatstone initially used several and the Morse system evolved into an efficient system of dots and dashes where the noise of the receiver proved adequate to recognise what was being sent. Morse designs developed into some very practical machinery.

Alexander Bain

Morse was not alone in fearing that his work would be stolen. Alexander Bain invented the electric clock and together with John Barwise, his employer at the time, patented it. The idea is to keep a pendulum moving by applying electromagnetic pulses. The time at one accurate clock can than be trasmitted to others by the pulses.

Bain's side of the story is that in 1840 he could not afford to develop his ideas and was introduced to Charles Wheatstone in the hope of some help. Bain demonstrated his models and was told "Oh, I shouldn't bother to develop these things any further! There's no future in them." But three months later Wheatstone demonstrated an electric clock to the Royal Society.

Wheatstone's side was that Bain had worked for him briefly and his own ideas were earlier. He tried to block Bain's patent but failed.

As a telegraph engineer Bain installed many systems. In later life, due to poor investments, Bain lost his money and in 1873 Sir William Thomson, Sir William Siemens and Latimer Clark were amongst those petitioning Gladstone to give him a Civil List pension which was awarded at £80 per year.

Facsimile Machines

Bain is also credited with the first invention of a facsimile machine, which he worked on from 1843 to 1846. Bain used the pendulum action of a clock to read a patern of conducting blocks at one end and write the pattern at the other. However Bain's idea wasn't very practical.

The first working fax-like device was the Pantelegraph invented by Giovanni Caselli.

The Caselli Pantelegraph involves writing the message on tin-foil in non-conductive ink. It is then scanned in lines back and forth by a light weight stylus, connected to an electrical circuit. At the receiving end a second stylus passes current through paper impregnated with potassium ferricyanide which turns dark blue when a current passes through it. The key trick which Caselli solved is to maintain synchronisation between the two ends. Although the pantelegraph worked quite well it had a relatively small take-up. It was used in France and Russia; and sometimes by banks to verify signatures. It was too slow to make good use of the wires.

There is sometimes debate as to who invented the raster scanning principle used in television - Nipkow, Baird, Farnsworth or Zworykin. The answer seems to be non of them; the principle had been discovered by Alexander Bain.


Morse Code

Cooke and Wheatstone's early telegraphs were too expensive, needing 5 wires and a return to work five needles and spell out the letters. The later ABC transmitted pulses using a magneto, but used the pulses to move a needle round a clock-face so it could need a train of many pulses to reach a particular letter.

Morse experimented with moveable metal blocks holding the code, but only needed one wire. It turned out that people didn't need the metal blocks, they could learn to tap the codes directly into the wire quite quickly. Morse code developed over time but quickly became well adapted to the task, with short codes for frequently used letters.

Telegraph inventors tried to convey what would have been written. Actually printing the message wasn't impossible, Wheatstone was demonstrating printing telegraphs by 1844). However printing letters was more expensive and error prone than recognising the rhythms in a code. A non-obvious human ability which could be developed with a bit of training was there to be used.

Morse code has some similarities with a modern computer code. It takes ordinary written language and turns it into a series of voltage pulses on the line. The Morse key basically connects and disconnects a battery using a keyswitch. At the other end of a long line a coil senses the change and works a printer or just a sounder.

Sending by just making and breaking the circuit is less firm than sending by actually reversing polarity so that a dot goes to one polarity whilst a dash goes the other. The needles on Cooke and Wheatstone telegraphs could be driven either way. Later telegraphs used a similar idea. For instance, the key might drive one way for a dot, the other for a dash and the receiver might have a steel post on one side and brass on the other to give a distinctive sound. A lot of modern computing code schemes take meaning from polarity reversal, not merely from a voltage level.

Morse does have similarities with a computer coding scheme called Frequency Modulation (FM). Each bit period starts with a clock bit and if it is a "1" is immediately followed by another bit. Morse could be seen as always having a clock followed by "0" for a dot and "11" for a dash.

The code has changed several times. As suggested Morse originally envisaged arranging metal blocks like type on a rod, then he and Alfred Vail conceived the idea of dots and dashes. Originally there were apparently dots, dashes and long dashes but three symbols proved unnecessary.

Many character codes are fixed length but Morse uses a variable length, minimum redundancy scheme that would later be called a Huffman code. For instance the letter "E" and "T" are common in English text so they are assigned a single dot and dash respectively. "X" and "Z" are not common so they have a sequence of four dots and dashes. Alfred Vail is said to have counted the frequency of letters in a local newspaper and assigned the original codes.

The code looks a bit like computer binary but there isn't a direct correspondence with the kind of zero or one signals between logic chips. As well as the dot and the dash there is also a space where there is no transmission. The gaps in transmission might signal the end of the message, or they might indicate a fault. Telegraphs sometimes used "mark" (on) being the rest state so that it was clear if the circuit was working. Some kind of verification that the circuit is live is still common today.

Morse Code Speeds

To the untrained ear Morse code is meaningless. People might pick out the odd phrase like dit-dit-dit dah-dah-dah dit-dit-dit for S.O.S which became the International distress code. A trained person can send Morse very fast and apparently hear it faster still.

Codes can vary in the length a sender gives to the dot and the dash; a dash is supposed to be three dots long but there is no specific length to a dot - its just "brief". A good speed with Morse code is 30 words per minute where a word has 5 characters like "Paris" - so 150 characters per minute. Although that speed is probably comparable with a lot of modern typing it is half the speed of a trained typist who should achieve about 60 words per minute. It is comparable with fast handwriting speeds.

A dash is three dots. An inter character gap is also 3 dots long. Inter word gaps should be 5 to 7 dots long.

People will vary in speed and accuracy. Morse code is not easy, the operator has to remember short bursts of dots and dashes and convert to and from text. Most people can use Morse code a bit (with practice) and the codes are short and seem more friendly than computer binary. Some people can adapt to the code with training and work with it at fairly high speed. Proficient operators are very fast - but there aren't a great many of them. In Eastern Europe high speed Morse has become a "radiosport", speeds of 300 characters / 60wpm are reportedly common.

The speeds achieved by skilled operators imply that they aren't consciously memorising and translating characters as an amateur would but that handling the code has become a sensorimotor skill.

The coding was not well suited to electromechanical printers so most printing telegraphs adopted a different scheme. Apparently early Creed machines did use Morse (below).

Code Complexity

Morse also devised a scheme to transmit a numeric code and look up the meanings in a table. This proved unnecessary for the ordinary Latin alphabet.

For Chinese telegraphy a lookup-table was in effect what was done. Chinese script has thousands of ideograms and it would be difficult to assign a direct code to each of them. Instead the ideograms were arranged in a book and given four figure numbers. The telegraphist sent the numbers and these were looked up at the other end in the code book. This involves an extra layer of skill because looking up the character to number translation is a skill in itself, since there are several thousand common symbols. Some Chinese telegraphers could remeber sufficient of the codebook to work from memory. The "four corner method" was developed in the 1920s

Costs

Telegraphs were a public good. Not only could people telegraph seasons greetings or reserve a room, the telegraph had serious economic use. Perishable goods could be shipped to the markets that would pay best. Merchants could enquire about the price of goods from suppliers in nearby towns and buy from the cheapest. Businessmen could also get their banks to transfer funds, and they could monitor how stock markets were performing. Ship's sailings were gathered by Lloyds hailing stations.

For a few months a vessel was placed off Land's End in the mid English Channel so that ships could exchange messages, take on last minute provisions or drop stowaways. That experiment proved unsustainable, the cable wore out with constant movement and the people on board became ill, but the attempt showed how valuable communication had become.

Newspapers took on a new life, with steam presses allowing mass production and correspondents around the country to fill the space. Department stores could order the latest fashions from Paris and London and advertise them in a local paper. Government administration took on a new directness; Civil Servants coulds actually know what was happening without having to wait three days for a messenger.

There is more on the costs of providing telegraphs below. In principle the telegraph companies might charge what they liked for a service that had never existed before. It looked as though it could be a very propfitable activity. Certainly some people who invented the right thing or bought shares at the right time did make fortunes. A telegraph business also had to balance charges against use - too high a charge would mean equipment and staff sitting idle. Hopes of being wildly profitable weren't always met, perhaps because government used but mistrusted telegraphs. In most of Europe the telegraphs were government owned or supervised. In the UK telegraphs were nationalised in 1870 establishing a government monopoly over communication that lasted until 1984. Unfortunately, however, the telegraph and then the phone system never made much profit and they would blame government interference in their affairs for this.

Telegraph companies generally charged per word. Telegrams were not cheap, around the turn of the nineteenth century a short telegram in the UK with 12 words including the address cost sixpence. A telegram from South Africa to London started at 10 shillings. Telegrams were charged per word so brief communications cost less.

To put these prices in perspective wages at the time might be 40 shillings per week, so a telegram from South Africa was more than a days wage. If your brother emigrated you might expect to exchange letters but only the wealthy would send telegrams. The telegraph was basically a business tool and even then a message would be carefully thought through.

Codes &s; Privacy

Telegraphy was done by people reading the messages and translating them onto the line. This is a problem if you have anything private to say and particularly undesireable for trade and financial information.

There were a few private telegram networks, in the US it was possible to get a leased line from the telegraph companies. In the UK this was less common and not usually allowed. It would be an expensive option as a pair of skilled people would be needed. Where the UK allowed private lines they usually had Wheatstone ABC instruments which spelled the words out on a dial.

One obvious solution was to encrypt the telegram. This wasn't usually allowed by telegram companies. It is more difficult to transmit a seemingly senseless cypher than it is to send natural language words. Words have some inbuilt redundancy so slip-ups in the coding are often correctable. With a cypher a slight irregularity in typing will turn "dot dash" meaning "A" into "ET". The 19th century was also a time of war and spy scandals, so some countries refused any encrypted telegram.

Coded Telegrams and Commercial Codes

Code telegrams used regular words but with changed meanings. The telegram companies chose to do little about this, perhaps because steganography is difficult to detect. "MISSING YOU COME SOON AGATHA" looks innocuous, might indeed be innocent but could have a covert meaning know to the recipient. Commercial codes were common, for instance:

The Great Western Railways used codes for years. They published a book “ 1939 Telegraph Message Code ” with more than 900 code words - many of them also used by other railways. The code words were also stencilled on the side of rolling stock, so a MOGO is a covered motor car wagon, a MACAW is a timber truck and a TOAD is a goods brake van. CHICORY means ‘Cannot trace delivery’ and STORK ‘We have no trace of your invoice ’ - terse messages that remain relevant.

The A.B.C. Telegraphic Code

Bentley's Complete Phrase Code

Slater's Telegraphy Code (1916)

Western Union Universal Codebook (1907)

Unicode (1889) uses Latin words to communicate phrases. A user would look in the book for a topic, pick the appropriate phrase and use the latin word given alongside. The words used were brief and usually unambiguous but largely meaningless in themselves. The recipient would look at the alphabetic arrangement of the Latin words and get the phrase. The Unicode book largely provides verb-phrases and adjectives. Users supplied the nouns.

Many specialist telegraphic codes were used, creating prototype business languages.

Q code is a list of three-letter codes not easily mistaken for natural language. The letter ‘Q’ is only occasionally the start of a natural-language word (hence it's high value in Scrabble). Sending a Q-code implies a question, so that is appropriate as well. Words that start ‘Q’ are therefore used for call-signs (nouns) and Q-codes (verbs and adjectives) in a telegraphic scheme.

Q-codes were created by the British GPO for use by ships and coastal stations around 1909. The codes were particularly useful for ships wireless operators speaking different languages - all they needed was to understand the code - not interpret a foreign language.

Q-codes include:

  • QRA - What ship or coast station is that ? - This is __.
  • QRB - What is your distance ? - My distance is __.
  • QRC - What is your true bearing? - My true bearing is __ degrees.
  • QRD - Where are you bound for? - I am bound for __.
  • QRF - Where are you bound from? - I am bound from __.
  • QRG - What line do you belong to? - I belong to the __ line.
  • QRH - What is your wavelength in meters? - My wavelength is __ meters.
  • QRJ - How many words do you have to send? - I have __ Words to send.
  • QRK - How do you receive me? - I am receiving well.
  • QRL - Are you busy? - I am busy.
  • QRM - Are you being interfered with? - I am being interfered with.
  • QRN - Are your atmosperics strong? - Atmospherics are very strong.

Q -codes were intended for wireless telegraphy between ship and shore stations. Questions elicit a response. If the response is the same code it has a supplementary or confirmatory meaning.

The list of Q-codes grew to over 100 by the 1970s. By this time they were being supplemented by the Z code use by Cable & wireless and then with a different set of military meanings by NATO. The NATO Combined Communications-Electronics Board (CCEB) Publishes the Communications Instructions Operating Signals - ACP 131 which details the codes in use)

In their limited field of radio operation Q-codes have been more succesful than Esperanto.

The need for telegraph codes rather died away when telephones became commonplace - which is a pity because the idea of business languages has considerable relevance now.


Costly and Difficult Work

Early Cable

Telegraph lines were not cheap to construct; hence the early inventors interest in government funding. Then as now the main difficulty was often securing a wayleave - a route for the cable.

The lines needed a right of way. In Europe much of the land was private property and negotiating with farmers and householders would be difficult. The same was true of the Eastern Seabord United states. Roads might be used, but they were often winding and the permission of a parish vestry would be needed. Railways were an obvious public good so had special acts of parliament. Telegraphs tended to use railway lines as they had the straightest routes and there was considerable common interest because railways increasingly used telegraphs. In towns the number of telegraph wires alongside roads and over rooftops became a nuisance.

Cable is a familiar thing today, houses and offices have hundreds of them. However the notion of a cable had to be invented for the telegraph, and really took off with the invention of the telephone and electric light. (below)

Wire has been used in jewelry since ancient times. Wire is made by drawing; pulling a rod through one or more dies, usually cold.

Brass and iron wires became important in England because of their use in the carding of wool. Edward IV banned imports in 1463. After a dispute with the Duke of Northumberland regarding mining of gold Elizabeth 1st created a monopoly for the Company of Mineral and Battery Works who established a wireworks at Tintern in Monmouthshire. However by the late 17th century the monopoly seems to have been ignored, although the wireworks survived until 1895.

Birmingham in the English midlands was a world-renowned centre for the making of fine metal items. Metalworking using using iron ore from Warwikshire and coal from Shropshire was established by the mid 16th century. By 1683 records for hearth tax returns show 202 forges in the town being worked by hiltmakers, bucklemakers, scalemakers, pewterers, wiredrawers, locksmiths, swordmakers, and workers in solder and lead. Between 1760 and 1850 Birmingham people registered over three times as many patents as any other town or city. Oddly, although Birmingham was a center of enterprise it doesn't seem to have become a particular center of expertise in telegraphy.

Wires could be copper or galvanised iron. A telegraph can use a single wire and an ‘ earth return ’ although if there are several telegraphs close together the earth return may not work well. Within a few years there was a good idea of resistance - which is what was overcome by higher voltages or preferably by repeaters.

Wiremaking was an established art, even if there had not previously been much call for miles of very fine wire. But a cable needs both wire and an insulator.

The insulators readily available to Victorian engineers were fabric woven around the wire, rosin based varnish and gutta-percha, a newly discovered natural plastic produced by the Palaquium gutta tree. An electromagnet would use varnished wire, the connectors cotton insulated wires and connections between equipment and outdoors might use gutta-percha or later vulcanised rubber sheath.

Cooke's very first cable used a wooden form with copper strips buried in its side. This was expensive and not very effective. Later outdoor underground cables might use copper wire in an iron or lead pipe. Filling the pipe with bitumen was one way to keep damp from the conductors. Digging the ditch and putting a cable in it was an expensive exercise then as it is now.

Most overland lines were bare wire, held up by glass insulators on tall telegraph poles. Putting a crosspiece at the top of the pole gave space for several wires. Both Cooke in the UK and Cornell in the US came up with similar ideas.

From the 1850s to the 1970s each railway line was accompanied by a line of telegraph poles, with the number of crosspieces declining between town as various lines branched off into the countryside. The importance of a place could be judged by the number of wires it had.

There was the beginning of an idea that wires had inductance and capacitance and that high voltages did nothing to overcome that.

Techniques such as twisted pairs to cancel interference and impedance matching and termination to avoid crosstalk and reflections in the wire were partly developed using loading coils on phone wires. These ideas began to develop with submarine cables, where reactance slowed operation. The development of the telephone was about contemporary with that of lighting and trams, which interfered badly. Development was pushed further by wireless. They only fully developed for network cables in the 1980s.

The most difficult overland route was probably that attempted by the Western Union Telegraph Expedition which intended to run a cable from San Francisco to Moscow and then connect to the rest of Europe. A line had already been run from Washington to San Francisco, and all Western European cities were interconnected. The Line had to cross territory unknown to Europeans in British Columbia and Russian America (Alaska) and the project was abandoned in 1867 when the Transatlantic cable was completed as the idea now seemed obsolete. Politically it may have had considerable importance, making Americans more aware of the nature of their continent. Expeditions were needed to explore territory, and the teams had to construct their own roads.

Submarine Cables

Telegraphs across the ocean would link nation to nation, an obvious objective. Morse tried the idea in 1842, using a wire insulated with India rubber and hemp in New York Harbour and was able to send a telegraph. Charles Wheatstone tried the idea from a boat in Swansea Bay in 1843.

Wayleave was the main problem for a cable largely over land. For a cable across the ocean the problems are technology, materials and the odd behaviour of long cables in close proximity to a return circuit.

A submarine cable is not unlike a cable intended for underground use, just better insulated and rather more robust. Considerable strength is needed because a great length of cable will be suspeded from the rear of the cable-laying ship. There are a couple of dozen ships these days that specialise in ocean cable laying. Once on the sea floor the cable might be snagged by fishing nets or ships anchors and nobody was quite sure what kinds of events happened on the sea floor. these days cables are often buried in trenches despite being 2 kilometers down - accidental damage and sabotage have been all too common.

Channel Cable

The first cable between England and France had a brief history. Brothers Jacob and John Watkins Brett secured a concession from both the French and British governments for the Anglo-French Telegraph Company to lay a cable. In London the Gutta Percha company made 25 nautical miles of cable made fron No 14 gauge copper covered in gutta percha to half an inch diameter. Short lengths had to be joined with solder and covered with more gutta-percha then tested by immersing them and checking with a battery and galvanometer. On 28th August 1850 a small steam tug the ‘Goliath’ set off from Dover, accompanied by HM Surveying boat ‘Widgeon’. The ship proceeded at about 4 miles per hour. The cable was little heavier than water so at intervals it was clamped in lead weights. Tests were made along the way and the cable seemed to work. However when the cable was landed it was equipped with an early printing telegraph. The problem was that the letters became jumbled. The issue was reactance in the cable, causing signals to become mis-timed, At the time the problem might have been solved using an ordinary Morse key and sounder and being patient.

At either side of the channel the operators blamed one another, thinking that too much champaign had addled their wits, and retired at midnight. Next morning the cable was completely dead. Apparently a French fisherman had accidentally hooked the cable and thought it was a strange seaweed with gold in it, so he hauled in as much as he could get and cut it with his knife.

The Brett's attempt to link England and France is comical in its innocence. They simply pay out an insulated wire and try it - and it almost works.

In 1851 Thomas Crampton, a railway engineer, tried again. His cable was altogether better, four conductors of No 16 Birmigham Gauge copper with interstices filled by tarred Russian hemp covered with two layers of gutta percha and then covered with 10 galvanised iron wires wound around as an armour. This cable weighed 7 tonnes per mile and was proof against fishermen.

Several more submarine telegraph cables were run from France to England and between Wales and Ireland. On the US side a cable was laid to Newfoundland.

Atlantic Cables

The Atlantic Telegraph Company was formed in 1856 and it's cable from Ireland to Newfoundland was laid in 1857/58. The cable failed after a few weeks of operation. The cable had been slow in operation - the same problem seen in shorter submarine cables but on a greater scale. The idea of cable reactance and impedance was just beginning to form.

The company electrician Edward Orange Wildman Whitehouse had previously been a succesful surgeon, but gave it up to pursue his interest in electricity. Whitehouse had contacts on the board of the ‘ Atlantic ’company and was appointed its electrician. Whitehouse was convinced that the answer to longer range telegraphs was more batteries and more voltage. He operated the cable at 500 volts with coils capable of delivering 2000 volts. Whitehouse aimed to overcome obstacles by sheer pressure (he may have been working to an innapropriate plumbing metaphor). He was later blamed for damaging the cable. Investor confidence in the idea of a transatlantic cable flagged and it was several years before the finance could be raised for another attempt.

The 1866 Transatlantic cable was laid by the SS Great Eastern, a huge ship for the time, fitted with three large tanks for the 4,300km of cable to be stored under water. Gutta percha decays in air but is biologically innert in water. Exposure to air had probably contributed to the failure of the previous attempt. The decks of the Great Eastern were converted for the paying out gear. Companies making the cable were reorganised to form the Telegraph Construction and Maintenance Company "Telcon" - under it's later name "BICC" the company survived to the end of the 20th Century (and is now Balfour Beatty".

The new cable was of higher specification with very pure copper conductors and multiple coatings of insulator.

The istruments at either end operated with sensitive mirror galvanometers. The cable was still slow - it could handle just 8 words per minute.

Transatlantic cables had a problem with fast operation. The cable was just several thousand kilometers of copper. There were no repeaters because there was no viable way to power them. Telegraph companies were aware of the risk that the cables wouldn't work well.

William Thomson

William Thomson, later to be Lord Kelvin was amongst the luminaries who attempted to solve the problem of long cable.

Thomson's answer was the Curb Sender. This curbs the prolongation of the signals due to reactance by following each positive or negative pulse by it's reverse. By today's standards this might be thought to make echoes and interference worse, but apparently this aftercurrent did reduce the problem.

Thomson also invented the siphon recorder. A roll of paper passes under a glass siphon which oscillates slightly with polarity of the signal. A static charge on the paper attracts or repels so that ink makes dot and dash patterns. The siphon recorder is something like an inkjet printer - but well over 100 years ahead of its time.

Reactance

Short lines carrying a slowly changing signal can be characterised as resistances. Telegraph lines were rarely more than ten kilometers long without there being some sort of repeater to regenerate the signal. Even if the line was long it was usually carried on a pole, some distance away from the wire or ground-path used for the return signal. Transatlantic cables were different, a long cable and its return in one tight cable enclosed in the sea. The cable acted pretty largely as a capacitor but because of it's great length it also imposed a fractional time delay. The idea of cable termination was unknown so that a signal emitted at one end would reflect from the other.

In the 1880s Oliver Heaviside, a nephew of Wheatstones who had become a telegrapher and then an enthusiast for the newly suggested Clerk Maxwell equations worked out solutions to the problem. Heaviside produced the "telegraphers equations" which showed that increased inductance would reduce distortion and attenuation. Heaviside suggested either a uniformly distributed inductance using iron or load coils to counteract the problem.

Corporate politics had begun to dominate in telegraphy. By this time officials in the British Post Office had their own theory that inductance was to be avoided, so for many years they resisted the idea. Some years later the idea was rediscovered and came into use in the US and Germany. Heaviside patented coaxial cable in 1880 but made little or no money from his inventions. Despite living in some poverty he seems to have been motivated purely by science.

British companies dominated in the 19th century submarine cable market. British entrepreneurs had the wealth to subscribe to new cable ventures - and they wanted news of their foreign investments. Shipowners and the Lloyds insurance market were particualrly keen on telegraphy. Many of the countries to which cables ran had been settled by people of British origin, so their activities had a sympathetic interest for newspaper readers. Britain was then vitally dependent on shipping (it still is) and a shipowner could read news of events and telegraph new orders as vessels arrived. Furthermore the supply of gutta-percha,came from the British colony in Malasia. Gutta-percha is a natural plastic and was considered the best insulator (and used until polythene was invented). And of course the government was interested, they had the Empire to consider.

More Return on Investment

The more traffic a cable can carry the more it is worth. Double the traffic might be double the revenue (within limits). In the early 20th century speeds on transatlantic cables finally rose to 120 words per minute. The same need for speed also applied to landlines.

One of the keys to getting more from a cable was the invention of duplexing. A cable can carry current in either direction. Julius Gintl invented duplexing and Joseph Barker Stearns devised the first practical way to duplex two signals onto a line by using different polarities. The idea was taken up within a couple of years on the Atlantic cables.

Thomas Edison

Thomas Edison was a telegraph operator and developed the idea of quadruplex working. To send two signals in the same direction at the same time one varies polarity of the line (we would now call this phase modulation) and the other varies the voltage (known as amplitude modulation).

Edison is possibly the most famous inventor in history. Edison did not invent electric power, but his making of a practical light-bulb made electric lighting and the widespread use of electric power possible. Before Edison telegraphs and telephones were battery powered - almost everything electrical was. The reason for power stations, pylons and cables was essentially electric light. General Electric was started by Edison, one of several large companies originating in the telegraph age.

Edison was a poor student and was homeschooled by his mother. He had an entrepreneurial spirit and sold candy and newspapers on trains. He saved a child from being run over and in gratitude the father trained him as a telegrapher so he got a job with Western Union. He worked for a time at the job, experimenting when he wasn't busy. Eventually he got the sack when acid used in one of his experiments ran onto his bosses desk.

Edison patented many ideas relating to telegraphy including a stock-ticker printer in 1869.

The Quadruplex telegraph was something of an innovation, he had thought of selling the idea for $5,000 but asked Western Union for their valuation - and they said $10,000.

With the money from the quadruplex Edison started a research laboratory "Menlo Park" in 1876. This is generally regarded as the first industrial research campus and perhaps it's most succesful. In a few years it produced the carbon microphone that proved vital to telephones, the phonograph (the first music player) and the first practical electric light bulbs. Edison created a complete system for generating and distributing electricity.

The Edison model of a big central research lab attacking a problem has been recreated several times, very succesfully with Bell Labs. There is some questioning of this model now. Pfizer is in the process of closing several of it's large research facilities in 2011.

The town of Edison, New Jersey, which is where the Menlo Park Lab was, calls itself "Birthplace of the Modern World" and has the motto "Let There Be Light" with some justification.

Edison may have started patents-as-a-game. He had 1,093 (?) US patents to his name. Until about 1870 patents explain what they are about clearly and seem to be advertisements asking for manufacturer interest. Sometime in the 20th century patents became a series of incoherent claims primarily aimed as traps for unwary manufacturers.

As so often happens Edison was not alone inventing the ligtbulb. Joseph Swan invented electric light, and demonstrated it several times including to a meeting of several hundred people at the Newcastle Litterary and Philosophical Society a year before Edison's patent. They agreed to share patents and set up ‘EdiSwan’.

Growth in Traffic

Morse telegraphy was widely adopted and an extensive network of wires developed. However no technology has an infinite life.

Since Morse was a cost effective system at the time it came to be widely adopted as a standard, despite the inconvenience of needing specialist operators. It may not necessarily have been the best conceivable system - but it was the system we had.

Morse is a good way for a skilled human to send information across an expensive communications line. Some proportion of the population seem able to develop peculiar skills to the highest pitch - musicians for instance do a job similar to telegraphy in manipulating a small instrument quickly. Not everyone can master a musical instrument. There are still competitions in high speed Morse.

In his early inventions Morse had tried to automate the sending process by assembling conductive type elements on a form in order to transmit the message more quickly. As traffic built up there was a growing need to get more information onto a line beyond what even quadruplex operation and skilled operators can do.

Wheatstone was ever inventive. In 1858 he patented an automatic sender based on a paper tape to send messages in Morse code. The Wheatstone Automatic telegraph used a paper tape with two channels and a sprocket wheel. The ‘ Automatic ’ could send ten times faster than a human operator - at up to 400 words per minute. The message could be received by a Morse printer and the dots and dashes decoded in the usual way. Messages could be prepared offline, then spliced together. The Automatic was particularly suited to sending news reports quickly. Following the introduction of Prime Minister Gladstone's bill for Home Rule in Ireland in 1886 1,500,000 words were despatched from the central telegraph station in London by 100 of Wheatstones Automatic transmitters. Messages sent by the Automatic were charged by the yard, not the word.

The ‘ automatic ’ did have some defects. It sent messages quickly but was tedious to operate requiring a good knowledge of the code. Punching the tape was also a bit difficult requiring some energy from the person working the punch hitting three keys with rubber mallets.

Simpler Telegraphs

Morse apparatus is simple to make, it is essentially just a battery, a switch and a coil which works a sounder. It is easy to use if both sender and receiver have all the time in the world to look up codes. Using it professionally and fast requires skill.

Although most of the world adopted Morse Code there were always a few lines where an instument that was simpler to use was preferred. The Wheatstone ABC provided this. The original instrument was credited to both Cooke and Wheatstone but the latter went on refining it.

The generator consisted of a rotating coil between the poles of a permanent magnet. Rotating the armature generated a series of positive and negative pulses - just as later magnetos would. Wheatstone would later realised that he had invented an electrical generator, not just a telegraph sender.

The communicator had a circular dial marked out with letters of the alphabet and equipped with a pointer. Next to each letter was a key. The user pressed the key corresponding to the letter to be sent and wound the generator handle. The needle on their own intrument turned with the generator. When the correct number of pulses had been emitted the communicator disconnected the generator from the line. The communicator controlled transmission of the pulses sent to the line.

At the receiving end the indicator also had a dial which moved with the pulses.

The communicator and indicator were synchronised at the start of a connection. The communicator is effectively a synchronous motor. Speeds of about 15 words per minute were possible. ABC telegraphs were used on less busy lines in Britain and the Empire, avoiding the need for a telegrapher familiar with Morse code.

The generator in the "ABC" is a hand cranked magneto-generator. However there was no great need of generators, since neither electric motors or lighting had been invented.

Charles Wheatstone and Dr. Werner Siemens actually announced the invention of electrical generators on January 17th 1867. Siemens company had originally been founded to make telegraphs to the Cooke & Wheatstone patent.

If the Wheatstone ABC moved a star-wheel carrying type instead of a needle then it could be made into a printing device.

Printing Telegraphs

Modern computer printers are descendants of printing telegraphs.

Telegraph lines were expensive to install and the number of messages grew quickly so that faster transmission became very desireable. In the later 19th century several inventors developed printing telegraphs that could use paper tapes so that messages typed off line could be transmitted at high speed.

Wheatstone developed an early printing telegraph by 1844 and was demonstrating it at Paddington station on the line to Slough. Gertrude Sullivan recorded:

.... Another prints the messages it brings, so that if no-one attended to the bell,....the message would not be lost. This is effected by the electrical fluid causing a little hammer to strike the letter which presents itself, the letter which is raised hits some manifold writing paper (a new invention, black paper which, if pressed, leaves an indelible black mark), by which means the impression is left on white paper beneath. This was the most ingenious of all, and apparently Mr. Wheatstone's favourite; he was very good-natured in explaining but understands it so well himself that he cannot feel how little we know about it, and goes too fast for such ignorant folk to follow him in everything.

Wheatstone had a printing telegraph very early on. The problem seems to have been that it didn't make very good use of the line and perhaps it was expensive to make.

Hughes Printer

One of the first printing telegraphs to be widely accepted was devised by David Hughes. Hughes was working as a professor of music in Kentucky so perhaps naturally his keyboard looked like a piano with black and white keys. The Hughes keyboard was easier and quicker than a Morse key, anyone could use it. The messages printed along a paper tape so there was no need for complicated paper motions. Telegrams printed onto tape were still common into the mid 20th century. The Hughes "Type Printing Telegraph" was adopted by Western Union and it became an early standard in Europe.

David Hughes was an early experimenter and developer in many fields. He devised a carbon microphone, but rather than patent it gave the idea freely to the world. He developed early wireless technology as well, but he met with demands for scientific methods and proofs. Rather than battle through the claims he seems to have been satisfied with the fortune he had made from printing telegraphs.

Stock-Ticker

Stock Ticker machines were used from about 1870 through to 1970. The machines were typically used to give stock-market prices and were located in dealers offices and sometimes even in hotel lobbies. The machine is known as a ticker-tape because of the continual ticking noise the little print mechanism makes as it works.

Thomas Edison invented a stock ticker in 1869 (but there seem to have been others around the same time?)

The print mechanism is often a solenoid driving a print wheel via a ratchet mechanism. The wheel typically has 32 print positions - "0" and "1" are subsitituted by "O" and "I" and some letters omitted. It will take an average of 15 pulses for it to get into positiom to print. This gave a stock ticker a speed of about 1 character per second so it wasn't fast but all it had to do was cycle through the abbreviations and prices of the stocks. The ticker symbols were usually three letter abbreviations of company names.

The ticker just cycles through all the names printing their three-letter abbreviated name and the current stock price. The ‘ memory ’ until the next cycle is the tape, which is always in the same order so that it can be quickly searched.

Giving companies NYSE fixed names is another contribution to the idea of a business language. In today's world that isn't quite good enough - we need all companies in all countries to be part of a naming system.

Ticker tape was used in huge quantities in financial centers and once printed it had no further use, so it came to be used as confetti in ticker-tape parades.

Baudot Printing Telegraph

Émile Baudot's 5 bit printing telegraph code is particularly interesting from the point of view of computer development because it uses a 5 bit binary code. Cooke & Wheatstone had used 5 wires in early 5-needle telegraphs but with positive and negative signals to select letters. Baudot uses on-off signals which can also be recorded on tape. Baudot code gives rise to Murray code and then ASCII and UTF-8.

The Baudot system used a distributor - a rapidly rotating electric switch. The distributor does two jobs:

  • It multiplexes the apparatus, selecting between four or six key and printer sets at each end.
  • It then reads (keyboard) or sets (printer) a succession of relays that determine the character.

It also used a "chord keyboard" - another invention which could make a comeback as a way to enter data on the move. (see Microwriter) .

The operator types a character using two fingers of their left hand and three fingers of the right, the keys lock until a distributor reads them, unlocks them with an audible click and then the next character is pressed. Operators were expected to maintain a steady rythm and achieve 30 words per minute. Later version of the Baudot system prepared the message offline onto tape which was then read onto the line.

Baudot code was adopted as International Telegraph Alphabet No. 1. The Baudot code itself is no longer used, it was replaced by ITA Alphabet No 2 designed by Donald Murray.

Murray used a keyboard performator to prepare a tape using a typewriter-like keyboard and a transmitter to send the tape. This new design seems to have been the first use of "format effectors" for line-feed and carriage-return.

During the 20th century landline telegraphy was increasingly taken over by these printing telegraph machines. A printing telegraph can outpace all but the most expert human senders and receivers. And where the person will ultimately need a break the machine just keeps going.

Morse code became unusual on wires in the 20th Century, however it became more widespread than ever with the rise of wireless telegraphy.

The ITU

By 1865, just 21 years after Morse had first sent a message to Washington the telegraph system had grown huge. International telegrams had become vitally important both to commerce and government. Submarine cables were working between Britain, France, and the Netherlands and it was becomming clear that if a Transatlantic crossing wasn't built the overland route via Moscow, Russian America(Alaska) and San Francisco would be. (A Transatlantic cable worked the next year)

International working was made difficult by lack of standards. For instance the US and much of continental Europe used Hughes printing telegraphs, but the British Post Office only used them on Continental lines, not internally. Even though the telegraph can be a relatively simple instrument there are all sorts of ways equipment can differ - voltage, resistance, timing and the codes themselves. Representaion of the alphabet on a line worked by hand could differ around a theme set by by Morse, but printing telegraphs needed a different serial code. Then there were different standards for charging. Issues like cryptograms and steganography were treated differently in each country.

Countries began to set up bilateral arrangements. Prussia had fifteen agreements with other German states. The problem with bilateral agreements, then as now, is that their numbers tend to grow as the square of the number of players - and because of slightly differing interests on each occasion each may be different in some important way.

The French government called for a meeting, and after several months of negotiation the 20 founding members signed the International Telegraph Convention and agreed to establish the “ International Telegraph Union ” to deal with future ammendments. In 1885 the ITU drew up rules governing telephony, and in 1903 they began to draw up rules for wireless telegraphy. Since wireless signals have no boundaries states don't have any choice but to cooperate.

Phones- Technological Revolution

Before looking at how later telegraphs worked the invention of the telephone is highly significant. Telephones change the context for telegraphs - which become tools for considered, printed business messages whilst personal messages became more likely to go by phone.

Since the telephone was much easier to use than the telegraph it became a much bigger system. The wires were similar - sometimes the same. The materials cost of a telphone is possibly higher because it is more complicated. However anyone can use a telephone, so it can be mass manufactured.


Telephones

Alexander Graham Bell patented the telephone in 1876, the same year Edison started his Menlo Park Lab. Bell was working on a harmonic multiplexer at the time. Edison's invention of the quadruplex had earned him a handsome 10,000 dollars, a lot of money at the time. Several inventors had noticed that ‘ sympathetic vibrations ’ might be used to communicate multiple telegraph channels on one wire. Bell's device had several resonant steel reeds, so essentially he was sending sound over the wire. By accident his voice was heard in a nearby room.

The idea of harmonic-multiplexing became frequency-dvision multiplexing and was widely used in the 20th Century. It is still the way the radio spectrum is shared today. To make it work well it needs tuned circuits and amplifiers and those were not available until the early 20th century.

//////////

Teleprinters, Typewriters and Telex

The telegraph system was to some extent overshadowed by the growth of the phone system. It didn't come to an end for another century and actually grew right through to the Internet age. Some models sold in hundreds of thousands and these were robust machines, intended to last 20 or 30 years.

With the invention of "Telex" the telegraph grew, extending to every large business. Telex was primarily a tool of big business, not so much because of the expense of the equipment but because of the charges by telecoms companies, often nationalised or monopoly powers and somewhat contemptuous of anyone questioning their operations.

Multiplexing

Alexander Graham Bell and Elisha Gray had been working on "harmonic multiplexing" when they separately discovered a way to transmit sound on wires and invented the telephone. Gray's lawyers famously advised him to drop his caveat to Bell's patent because the telephone was unimportant compared with the prize of multiplexing.

Multiplexing did prove important. Using different sound frequencies to divide up the space on a medium is the way radio works and it was used on telegraph wires as well. In Britain in 1927 the GPO decided to use voice frequency signalling giving 18 telegraph channels on one wire.

Printing

Samuel Morse originally intended the telegraph to be a printing device; a pencil would mark the dots and dashes on the paper. However it often wasn't neccessary for the operator to look at what was written since they learned the skill of interpreting them from the sound alone.

A printing telegraph uses codes to select positions on a typewheel and strike those onto paper. The mechanism can be quite similar to the Wheatstone ABC using magneto pulses, or a ratchet and cog mechanism (stock ticker), or something more evolved as in the Baudot machine. However the output is a thin paper tape containing the message in upper case in one long thin line of print. To turn the message into something more readable telegraph companies would often divide it into lengths and paste them onto a form. Cutting and pasting tickertape is laborious and messy.

Teleprinters produce a whole page, laid out more or less as though it had been typed.

Printing telegraphs evolved into teleprinters - by mating with typewriters.

Typewriters

Typewriters were invented somewhat later than the first telegraph printers. Typewriters and telegraphs do a rather similar job in some ways, so their development overlaps, as might be expected. Teleprinter inventors would later try to base their ideas on typewriters.

The Hansen Writing Ball was the first typewriting device, going into commercial production in 1870 (Patented 1865). As so often there is dispute about who should be credited as first, typewriters before that time were prototypes. The Writing Ball is an extraordinary looking machine. At the top a pin cushion with typewriter keys on plungers pointing down to the paper where a matrix of shaped letters converged on the cursor. Underneath the head the paper moved past the typehead on a curved carriage. Experiments were made typing against a stopwatch to optimise the typing speed.

The Hansen actually had an electromagnetic escapement so its inventor presumably knew about telegraphy, but doesn't seem to have tried to interest the companies in it. The Reverend Malling-Hansen was principal of the Royal Institute for the deaf-mutes in Copenhagen. One of his original motivations seems to have been a way to allow the children to communicate.

The Sholes and Glidden typewriter, also known as the Remigton No 1, was marketed in 1874. It has several claims to fame; its mechanism developed into that widely used in typewriters until the development of the golf-ball and daiswheel. The peculiar "QWERTY" keyboard layout was used on typewriters, then on teletypes and is now used on almost all modern keyboards.

Christopher Latham Sholes was a printer. He and Samuel W Soule had devised a machine to print page numbers on books and serial numbers on tickets. Carlos S Glidden suggested it might be adapted to print alphabetic characters as well. Together with Mathias Schwalbach, a German clockmaker, they developed a working prototype. The prototype looked like a piano (as did telegraph machines) and printed upwards to where paper passed across on a frame. The flying-key print mechanism is similar to a piano as well - Scholes modified a morse key to show people how it would work. In the original design the keys struck upwards through the paper to hit a ribbon or carbon paper on the other side. The limitation of this was that it could only use thin paper that would mould to the keys impact. People weren't going to want neat print on tissue paper so the idea was patented but not manufactured. A demonstration letter did attract James Densmore as a backer.

In the original machine the piano keys had numbers on the left keys and letters arranged in two rows of 13 on the right. The typebars meanwile were arranged around a metal ring so that the keys flew up into the center, then gravity would bring it back down. However if two keys were pressed in quick succession they would fly to the middle together and jam. It is thought that to get around this a combination of letter frequency analysis and trial and error led to a design with four rows of keys with the top row of letters marked "QWE.TYIUOP" whilst the "home row" retained the "FGH JKL" alphabet pattern.

A redesigned machine used a cylindrical platen, Soule and Glidden dropped out and sold their interest to Densmore. In 1870 Densmore demonstrated the machine to Western Union who bought some machines but thought they could develop a better machine for less than Densmore's asking price of $50,000. Sholes himself was eventually bought out for $12,000. Densmore then took the idea to E Remington & Sons, an arms manufacturer diversifying into sewing machines after the American Civil War. The Remington No 1 sold a few thousand machines - it was big and expensive and people didn't like the upper case only text. It was also overshadowed in exhibitions by Alexander Graham Bell's new Telephone. An improved version with lower case was introduced as the Remington No 2 in 1878.

Takeup of typewriters was slow at first. The machines were not easy to make, containing about 2,500 parts. A competitor called the American Writing Machine Company started business in 1881, so Remington lowered the prices and negotiated an agreement with a marketing firm to take all the machines produced. Sales increased dramatically. Material could be prepared twice as quickly and far more legibly on a typewriter than it could by handwriting so as businesses became interested in efficiency the take-up increased.

Typewriting had a dramatic effect on the employment of women. Typing and stenography paid much more than factory work, and seemed more dignified. In 1874 less than 4% of the clerical workers in the United States were women; by 1900 75% were.

Typewriters were once so ubiquitous that they were almost definitive of an office environments. In the 1980s and 1990s they were almost entirely replaced by computers.

James Densmore himself ultimately earned one and a half million dollars from royalty payments on typewriter manufacture.

Remington is notable in computing. In 1927 they merged with Powers Accounting Machines and Rand Kardex Company to form Remington Rand. In 1950 Remington Rand bought the Eckert-Mauchly Computer Corporation, founded by the builders of ENIAC. In 1952 they bought Engineering Research Associates (ERA), another computing pioneer. In 1955 they in turn were bought by Sperry which later merged with Burroughs to form Unisys.

Teleprinters

Teleprinters tended to be quicker and easier to use than a Morse telegraph. Morse still had a use because a simple key and sounder are lightweight and will work more or less anywhere. Teletypes look like a cross between a typewriter and a communications machine - which is largely what they are. A teleprinter tends to weigh about what a man can lift so they aren't exactly portable. The relatively simple electronics needs a nice, clean, binary signal.

Teleprinters use binary codes usually 5, 6 or 7 bits wide - or long if your looking at the serial signal. The original Baudot machines use 5 channel paper tape and 5 bits for a character. There were a few 6 bit machines; most late models of teleprinter - typically a "Teletype" had 7 bit codes.

Teleprinters generally don't use Morse code because:
  • Fixed length codes are much easier for a machine to send and recognise
  • A machine can't recognise the different cadences a human sender might use

Machines that could send and receive Morse became feasible in the 1980s but they used microprocessors to recognise the dot-dash patterns. If there is a microprocessor involved it probably makes more sense to send a telprinter code like ASCII.

Teleprinter designs tended to vary on a national basis. The telegraph operators had established national monopolies - Western Union in the US, the Post Office in the UK.

Teletype

"Teletype" is probably the most famous name because their machines were very widely used as computer input terminals in the 1960s and 70s. Teletype was the merger of two rival firms, Morkrum and Kleinshcmidt. However Kleinshcmidt then started his own company specialising in lighter weight machines, so they were rivals.

Joy Morton provided the finance that started the company. Morton made his money from the Morton Salt company which In 1930 it was sold to AT&T for $30 million.

The UK GPO tended to favour Creed & Cos telegraph equipment.

Wireless Communication

By the early years of the 20th century paper tapes and printing telegraphs were becomming commonplace and Morse code was less used on wires. But in the last years of the 19th century wireless communication was discovered and in 1901 Guglielmo Marconi sent a message across the Atlantic. Early wireless appartus used Morse code.

As with the telegraph Marconi was not the sole inventor. Oliver Lodge, Hertz, Popov Branley and Tesla had all done similar things. Tesla commented "Marconi is a good fellow. Let him continue. He is using 17 of my patents."

The sinking of the Titanic in 1912 contributed greatly to growth of wireless communication. The Titanic had two of Marconi's operators on board and the SS.Carpathia had another. The Carpathia heard the distress calls and rescued 711 people - but 1513 died. In Britain the postmaster general said "Those who have been saved, have been saved through one man, Mr. Marconi...and his marvelous invention." It became commonplace for ships to have radio apparatus and although voice communication quickly became possible Morse was often more reliable.

Economic Impact

Telegraphy undoubtedly had a huge economic and social impact. It was the first technology stock, fortunes were made and lost. Contrast the carreers of Ezra Cornell or Jay Gould who made fortunes with Alexander Bain, who lost his.

The first phase of industrialisation pre-date both the railways and the telegraph.

Matthew Boulton's manufactory was was built in 1761 on the site of a water driven rolling mill, Boulton built his residence "Soho House" next door. Boulton enclosed common land to build the factory but said the people who had used it were "idle and beggarly". The factory made "toys": buttons, buckles, knife handles and decorative boxes called japanned ware. Later it also made silverware, ormolu and had a steam powered mint.

Boulton & Watt began the steam age, selling stationary steam engines to mines and mills around Britain. Watt patented a steam engine with a separate condenser in 1769 but did little with it. In 1772 Watt's partner Dr John Roebuck had financial difficulties and owed Boulton money, which was traded for his share in the patent.

Boulton succesfully lobbied parliament to extend Watt's patents for an extra 17 years to 1800. They jointly began work improving the engine to a commercially viable state.

Boulton and Watt had customers get their parts from local forges and charged a percentage on the economies from using the designs; its an interesting business model but often led to conflicts.

Industrialisation in this generation relied on canals --

It is common to credit the Industrial Revolution on ready availability of books and learning - and perhaps on what is sometimes called the "Protestant Work Ethic". Mathew Bolton's factory in Birmingham and Stephenson's "Rocket" are earlier than the telegraph

There is no way to gauge the issue, but the impact of telegraphy might be compared to that of railways. Before the railway goods moved by horse in rutted lanes, or perhaps on tramlines between a colliery and staithes and then by sea. A railway creates one long smooth iron way between towns.

Fast railways were only really possible using telegraphs, to allow signalmen to know what trains are comming. The railways also used telegraphs and special telegraph codes heavily for operational reasons - to request and engine for shunting or to list the rolling-stock to be moved. Railway companies were often telegraph companies as well because they had the wayleave for a wire alongside the track.

The impact of the telegraph was widespread. Telegraphs allowed newspapers to carry reports from distant places, (hence continued use of the name in the title of several newspapers). The telegraph also founded businesses like Reuters (he supplemented the Aachen-Berlin Line with homing pigeons from Brussels). In the UK a special rate applied to telegrams carrying news so that the Post Office claimed it lost money cross-subsidising newspapers.

The telegraph made weather forecasts, money transfers and stock markets possible. Big projects like transcontinental railways need stock markets and national stock-markets became possible with telegraphs. There were once regional stock exchanges, for instance The Newcastle Stock Exchange building still stands (it was in Neville Hall) but the regional exchages faded away in the 20th century.

The really big change brought about by the telegraph was the rise of big business. The telegraph allowed a head office to control distant branches. Big banks, department stores and factories become possible allowing economies of scale.

The telegraph had an impact on natural language. Most notably it gave rise to brusque messages.

Telegrams were charged per word so short, clear messages were wanted. Sometimes parsimony gave rise to farce as in the incomprehensible telegrams in Evely Waugh's novel "Scoop" with messages such as "CO-OPERATING BEAST AVOID DUPLICATION BOOT UNNATURAL".

The data flow through the Internet is much greater than that of the telegraph. It is not so clear that the flow of knowledge is much greater.

Telegrams have been so useful that the service still continues in several countries. Even where email and texting are common the telegram may have a legal or social status that they haven't usurped.

Other References

Different countries had very different conditions for telegraphs and phones to operate in. A system that needs cables almost innevitably needs some political support - if not direct finance - because it needs wayleaves alongside or under the public highways. Government already used private telegraphs and within a few years they were significant users of the telephone as well.

The UK

There had been some private telegraphs. In the UK the Wheatstone ABC apparatus provided a compact and easy to use little unit. The UK Post Office had a particular problem with telephones because telegraphy had been nationalised in 1869. Telephones needed the same sort of wires and the court ruled that they were "telegraphs" within the meaning of the act. For about 30 years private phone companies operated under licence from the Post Office. Phone services in the UK were often compared unfavourably with those elsewhere. They were finally nationalised in 1912.

Conflict between legislation and technology continued; by 1912 wireless broadcasting was a rising possibility. In 1922 electrical manufacturers were forced by the GPO to create a single licensed British Broadcasting Company (BBC).

ould take a message from a "QWERTY" keyboard