Total Pageviews

Friday, 9 March 2012

New Zealand’s Donald Murray: The Father of the Remote Typewriter

How the ‘Australian Sausage Machine’
churned out pure gold
Sydney journalist was Wozniak’s
forerunner, creating the first 'emails'
People often ask me, “How come we were still stuck with QWERTY when we left typewriters behind and took to computers?” The reason is that QWERTYthe typewriter keyboard – had long since become an integral part of the codes used for electrically transmitting text.
And that is down to the work of a New Zealand-born Sydney-based journalist called Donald Murray.
Our predilection for communicating with one another by writing emails on a QWERTY keyboard owes much of its origins to Murray adapting an “ordinary typewriter” for sending and receiving telegraphic messages at the end of the 19th century.
Just as Steve Wozniak, in 1974, had got his idea for the Apple 1 computer from a TV set and a typewriter (“That made me think a computer should be laid out like a typewriter with a video screen”), so had Donald Murray, 82 years earlier, conceived his vision from looking at typewriters and cablecopy machines in the editorial office of The Sydney Morning Herald.
As Encyclopaedia Britannica puts it, “Teleprinters opened telegraphy to broader use, essentially by adapting it to the typewriter, which was then becoming a standard business machine and which could be operated by less-skilled personnel. Early teletypewriters were developed around the turn of the 20th century by Donald Murray …”
And as Murray’s system developed, he became responsible for the latter-day ubiquity of the QWERTY keyboard and the eclipse of the five-unit telegraphing chord keyset. Murray knew that training operators to use a new and extended Morse code alphabet was not feasible, but said anyone could type on a QWERTY keyboard.
Thus QWERTY became not just entrenched on typewriters, but on electric and electronic transmission devices as well – right through to the present day.
Those who today use Internet cafés will especially appreciate the prediction made in The New York Times on January 25, 1914. After an interview with Donald Murray, the Times wrote: “Can you use a typewriter? If you can, in a few years you will telegraph to London by paying a small fee in the nearest telegraph office and clicking out your message on an ordinary typewriter keyboard. The message will be printed almost immediately in London – printed so it will be intelligible to any one who can read.” Sound familiar?
By May 1899, Murray had worked out the actuating mechanism for his tele-typewriter, and in his United State patent application illustrated the perforated tape coding with an imaginary news report starting: “Bourke, Monday …”
When Murray started his project, in 1892, the young Henry Lawson had been sent packing to Bourke, with £5 and a railway ticket in his pocket, by The Bulletin’s editor J.F.Archibald, and found that “if you know Bourke, you know Australia”.

1890s Bourke, NSW
Did the imaginative Murray picture himself in the role of such an illustrious writer-journalist, being sent to the furthest inhabited point in north-west New South Wales, to the edge of the Outback, and needing a quick and easy means of filing copy back to Sydney? It seems likely.
The dateline used in Murray’s example would no doubt have bemused his New York patent attorneys, not to mention US Patent Office examiners. Bourke was then an obscure blot on the Australian landscape, known to Australians for the expression “back o’ Bourke”, a town to which Americans were blissfully oblivious.
Appropriately, then, Murray has come to be labelled by Japanese QWERTY authority Koichi Yasuoka as “The Father of the Remote Typewriter”. Murray’s invention had, for newspapers, the potential and pronounced added advantage of being applicable to Linotype machines, with the coded tape able to be fed straight into the typesetters.
Murray devised the “Murray Automatic Printing Telegraphy System” based on a duplex telegraph printer and what became known as the five-level character code International Telegraph Alphabet No 2 (ITA2, also known as the CCITT No 2 code in the US). ITA2 was in use until the seven-level ASCII emerged in 1963. ASCII (American Standard Code for Information Interchange) is still used for transmitting text.
Incredibly, given all this, Murray has no entry in either the New Zealand or Australian dictionaries of biography, and is merely referred to, but has no individual entry on Wikipedia or in Encyclopaedia Britannica.
And yet he was a leader among a remarkable group of electrical engineers and inventors who revolutionised telegraphy. Indeed, Murray is considered the outright leader on the theoretical side.
Experts in the field rank Murray in an elite group of just five pioneers, alongside Sir Charles Wheatstone (1802-1875), William Thomson, 1st Baron Kelvin (1824-1907), Frenchman Jean-Maurice-Émile Baudot (1845-1903) and the Dane Kay Oscar Arthur Gulstad (1853-1927). [David Edward Hughes (1831–1900) should probably be in there, too, I suppose.]
A number of electrical engineers and inventors worked on using typewriters or typewriter-style QWERTY keyboards to transmit copy in much the same period as Murray. They included:
.Samuel Van Buren Essick (1841-??), of Alliance, Ohio, and later Yonkers, a Union army veteran of the Civil War who invented a printing telegraph in 1882. Mares described it as the “Electric Typewritten News Bulletin” (above).
.Michael Hoffman Wilson (1856-19??), of Brooklyn (referred to by Mares, who was frequently misinformed, as “M.Hoffman”, but known as M. Hoffman Wilson) patented an autographic telegraph in 1884. Mares dates it to 1896 and calls it a telescriptor. Mares says it was a typewriter with electric contacts under each key.
.John Burry (1861-1907), a Swiss-born electrician of Boston and later New Jersey who patented a printing-telegraph system in 1889 and worked on it for the next 15 years. Mares devotes much space to the page printer, which Western Union continued to develop into the mid-1930s.
The Zerograph demonstrated with a Hammond
for Scientific American, 1903
.Leonard Ulrich Kamm (above, 1861-19??), of London, invented a printing telegraph based on a Hammond typewriter in 1895. Perhaps as biased toward English inventors as I am toward New Zealanders and Australians, Mares called Kamm’s Zerograph “probably the most wonderful of all” these types of machines. Kamm, however, was Bavarian-born. He later became far more involved in motion pictures.
.Charles Elmer Yetman (1863-1949), of Oak Park, Illinois, whose Morse code transmitting typewriter of 1893 has been well covered elsewhere. Typewriter above from Richard Polt's collection.
.Hans Knudsen (1872-1938?), a rather mysterious Danish engineer and inveterate inventor who, like Murray, spent time in London and New York in the early part of the 20th century demonstrating and patenting his various devices; his main interest became the transmission of pictures, and he was later involved in the development of television.
.Henry Augustus Rowland (above, 1848-1901), the famous Baltimore physicist, applied for a patent for a QWERTY telegraphy keyboard just before he died. Murray, saying Rowland’s was the only fully-developed, high capacity multiplex system to compare with his own, told The New York Times in 1914 that the Rowland system was by then no longer used by the Postal Telegraph-Cable Company. As it transmitted straight from a typewriter keyboard, instead of a tape, it had attained no greater output and no labor saving over Morse.
.Charles Luman Buckingham (1852-?) and John Charles Barclay (above, 1856-1934), both of New York. Barclay was chief engineer for Western Union and developed the Buckingham-Barclay system, which Buckingham had started working on for Western Union in 1881. Barclay added to it a typewriter keyboard in 1905. An operators’ strike in 1907 hastened Western Union’s conversion to the Murray mechanical system. Western Union bought the rights to Murray’s system in 1912, to be adapted by Western Electric for Western Union conditions. Murray had fully supplanted the Buckingham-Barclay by 1916.
.John Edward Wright, of New York. Between 1891-1914, he worked for the American Type Telegraph Company and later established the Wright Telegraphic Typewriter Company, developing a printer telegraph which ultimately had a typewriter keyboard. In 1909 this was adopted by the Postal Telegraph-Cable Company in preference to the Murray system, for which Postal had underwritten the development. But Postal abandoned Wright in 1911 and reverted to Morse. Postal much later used the Morkrum system (see below).
.Elmer Allah Burlingame (above, 1879-1935), of La Porte, Illinois, who, like Murray before him, placed typewriters on top of telegraphic devices. In Burlingame’s case, they were L.C.Smith and Stearns typewriters. He has been labelled the “Steve Jobs of La Porte County”. He established the Burlingame Telegraphing Typewriter Company with a factory in Boston.

.Edward Ernst Kleinschmidt (above, 1876-1977), a Bremen-born émigré who in 1924 joined forces with Morkrum (see below) and whose subsequent Kleinschmidt Laboratories became part of Smith-Corona in 1956; he was probably the “Bill Gates of Brooklyn”.
.Charles Lyon Krum (1852-1937) and his son Howard Lewis Krum (1883-1961), of Chicago, who took up the work of Frank Dillaye Pearne (1876-1927) and developed a teleprinter with Joy Morton (above, 1855-1934) and his son Sterling Morton (1885-1961). Krum machines used Oliver and Blickensderfer typewriters, but were also influenced by the Hammond keyboard.
The Morkrum company became Morkrum-Kleinschmidt (see above), which was sold to the American Telephone and Telegraph Company (AT&T) in 1930 for $30 million. Out of this came the Teletype Corporation, which, as with Morkrum machines, continued to use Murray’s principles.
.Tullio Venier Giara (1874-??), of Milan, travelled to the United States from Genoa at least three times in the early part of the 20th century, each time with a new, grandiose scheme, including the pneumatic tyre and a flying machine. Among them was also a multiplex telegraph system in 1904-05, which became a "diplex typewriting telegraph" in 1911. It was called the Telegraphic Mail and at various times looked to be based on a Blick Electric and a Hammond (see above). Giara based himself in first Boston and later New York while in the US, but remained an Italian citizen.

.And the Canadian-born Frederick George Creed (above, 1871–1957), who bought the North American rights to Murray’s patents in 1925. In 1897 Creed spent 15 shillings on a second-hand typewriter at the Sauchiehall Street market in Glasgow to start his experiments, despite being told by Lord Kelvin "there is no future in that idea". In 1928 Creed’s company was sold to International Telephone & Telegraph (IT&T, now ITT Corp).
All of these men deserve posts in their own right. But, in the main, they followed Murray with their developments, and, in a strictly technical sense, none was to have the enduring success of Murray.
Having said that, Murray’s name is probably the most obscure among them. This is because, largely for the sake of convenience, Baudot’s 1874 multiplex system continued to be known by the Frenchman’s name (immortalised by the “baud rate”) even after Murray had modified it for use on a full typewriter keyboard. ITA2 is basically the Murray code.
When he is mentioned in telegraphy histories, Murray is often referred to as English, or at best British (as a New Zealander by birth, he did remain a British subject throughout his life). But Murray began his work in November 1892, just after he graduated as a Master of Arts in logic at Sydney University and while working as a journalist for The Sydney Morning Herald.
And Murray got the idea to convert typewriters for use as teleprinters while working as a journalist with The New Zealand Herald in Auckland in 1887. Always “a born mechanic”, the mechanical features of the newspaper office fuelled Murray’s fertile imagination.
Downtown Auckland when Murray worked there
It is often stated that Murray developed his five-unit code in 1900-1901, after he had moved from Sydney to live in Manhattan. But he had first patented his idea for a printing telegraph in the US in 1893, while still living in Australia.
Murray’s allocations of signal combinations differed very considerably from that used in the Baudot code. The main reason for this was that he chose to use a typewriter keyboard, which relieved the operator of the burden of setting up the individual code elements. This allowed Murray to allocate the codes so that those characters having the greatest frequency of occurrence were given a combination which involved the least number of mechanical operations, thereby reducing the wear in the equipment.
The Murray code also introduced what became known as "format effectors" or "control characters" - the CR (carriage return) and LF (line feed) codes. Another advantage was the ability to correct copy.
At the transmitting end, the Murray system comprised a keyboard perforator and close by a transmitter. A novel feature on the transmitter was a start-stop device. At the receiving end was a reperforator and a printing receiver. Murray’s phonic wheel motor, in combination with Baudot's epicyclic correcting train, gave excellent synchronism. Synchronous printing telegraph systems employing constant length codes, such as the Baudot and Murray, were a great advance over previous telegraph systems.
While Murray’s early patents allowed for the adaption of any “ordinary typewriter”, they were mostly based on the Columbia Bar-Lock, which he used at Astor House in New York in 1900 to demonstrate the practicality of his invention.
When his idea first came to him at The New Zealand Herald in Auckland in 1887, however, it was based on the Remington.
For his first US patent, Murray wrote, “My invention relates to improvements in printing telegraph apparatus, and is designed, by means of a transmitting device provided with an ordinary keyboard, such as the Remington, to operate at a distant station any ordinary typewriter …”
His later patents, however, clearly show a Bar-Lock.
At first Murray, like Burlingame after him, had the typewriter sitting on top of his transmitting apparatus. But when Western Electric adapted the system for Western Union, it built a keyboard which looks very similar to the Oliver’s, maintaining, of course, the QWERTY configuration.
Murray signalled the future direction of his work when, in 1902, he wrote an article for The World’s Work titled “How cables unite the world”. Referring to “swift automatic transmission”, he said, “Recently several still more wonderful inventions have been perfected [referring, no doubt, to his own]. There is good reason to believe that it is now possible to work a typewriter in New York by playing on a typewriter keyboard in London, and vice versa.”
Murray described his system in a paper presented to the [British] Institution of Electrical Engineers on February 23, 1905, and later published by the institution in its Journal as "Setting Type by Telegraph". In 1910 he was made a member of the institution. He won its Fahie Premiums in 1905 and 1911, when he made another presentation. His papers were read to the American institution by his New York attorney.
Invercargill, 1860s
Donald Murray was born in Invercargill, New Zealand, one of the southernmost cities in the world, on September 20, 1865. He was the son of a Bank of Otago branch manager, John Murray, who had emigrated from Glasgow in Scotland in 1863. In 1866, John Murray (below) rejoined the Bank of New Zealand in Auckland as an inspector. Donald attended Auckland Grammar School from 1875 to June 1881, winning many academic prizes.
Given Murray’s Glaswegian connection, it is interesting that the first newspaper in Europe to use his tele-typesetter was The Scotsman, which from 1903 transmitted 40 columns a day from London to be automatically set in type in Edinburgh. The Glasgow Herald also adopted this groundbreaking system.
Towards the end of his life, Murray wrote that he had “Suffered the usual idiocies of better-class English education in Canterbury”. From 1882 he attended Lincoln University’s School of Agriculture outside Christchurch for two years. He began his working career as a farmer in Canterbury, but found it not to his liking. He wanted to be around machines.
In 1886, aged just 21, Murray sailed for Europe, around Cape Horn, and spent a winter in Dresden. He returned to New Zealand and settled in Auckland, where he joined the staff of The New Zealand Herald. He completed his studies at Auckland’s seat of the New Zealand University, gaining a Bachelor of Arts degree in December 1890.
Sydney Morning Herald staff picnic, 1890s
Murray moved to Australia in 1891 and joined the staff of The Sydney Morning Herald, as well as taking his MA degree at Sydney University. He later said that for his degrees at both the Auckland and Sydney universities, “I took logic among other subjects. That set the ball rolling and it has been rolling ever since.”
He applied to sit for the MA examination in logic in 1892 and, despite an illness, completed the examination requirements, being admitted “to the degree of Master of Arts in the School of Logic, Mental and Moral Philosophy, and Political Philosophy” on April 23, 1892.
However, Murray said later his “first-class scientific and, particularly, electrical and mechanical education” had come from “72 years of extremely varied experience all over the world, from Fiji to Moscow and Rio de Janeiro, including 12 years of newspaper work and 25 years of telegraph engineering”.
In late 1899 Murray went to New York seeking a financial backer for his enterprise. To his considerable horror, his system arrived at the Scientific American’s offices at 11 Spruce Street, New York, smashed into little pieces in transit from Australia. But Murray was able to patch it up and successfully demonstrate it at Astor House on Broadway, conveniently positioned between Vesey and Barclay streets, diagonally across from the offices of the New York Herald.
Scientific American at that time described the printer “as a sort cross between a sewing machine and a barrel organ”. It was worked by the operator turning a handle, and was dubbed “Murray’s coffee mill” and the “Australian sausage machine”.
A "sausage machine"?
Some latter day telegraphic historians have behaved like sensationalist journalists, and have highlighted these dismissive if jocular words about the resurrected wreck of 1899, taking them out of context and ignoring the very positive things Scientific American wrote about the completed system of 1906. Scientific American wrote, in fact, that the Murray system was “without rival” for long telegraph lines, underground cables and press messages. It really doesn’t take that much effort to get the facts right.
Regardless of what this very poor research work may indicate, Murray’s demonstration at Astor House went off well enough for Murray to get financial assurances from the Postal Telegraph-Cable Company. Murray took out three patents for a page-printing telegraph, which he assigned to this company, as well two further patents, for improvements to the actuating mechanism, in his own name. Postal workers dubbed the reconstructed system “The Baby”, and after two years work, it grew “almost out of recognition”. Yet Postal decided to throw The Baby out with the bath water, and not adopt it.
In the meantime, in October 1900, Australian newspapers, including the Sydney Mail, had leapt with justifiable pride on to a story which had appeared in the New York Post. “Typewriting by wire: A young New Zealander’s invention” the headlines ran. The Post said: “By the transfer of three patents from the inventor, the Postal Telegraph Company has just obtained control of a new device for the mechanical transmission of telegraphic messengers which it is expected will accomplish a revolution similar to that brought about in typesetting by the development of the linotype machine.”
The Post went to lengths to point out that with the Morse and Phillips codes, “the trouble has been in the physical limitations of the human operator.” It said Murray’s invention “is considered by far the most promising” solution.
The Post quoted a “noted expert” as saying, “It's a comparatively simple matter to make one typewriter control another by electricity, but for telegraphy the device so far has been a failure. Patent offices are full of these devices, interesting enough as curiosities, but they do not give any greater speed than the Morse key, and they are, therefore, failures. The typewriter, in theory, can run off a minimum of 100 words, but in writing messages with the needed care, experience reduces this figure to about 40 words at the utmost. We have had enough of those affairs; we want something different.”
Creed machine using Murray system
The Post said Murray had reached 114 words a minute “on the postal wires over a circuit of 388 miles, to Albany and back [to New York]; while over the line to Chicago, a distance of nearly 1000 miles, a speed of 103 words has been obtained.”
In May 1905, the Boston Evening Transcript, also pointing to previous attempts, said of Murray’s invention: “There is every indication that the era of fully-developed machine telegraphy has now arrived.”
Nonetheless Murray, disillusioned by Postal and with New York, had taken his Automatic Printing Telegraphy System to London in 1901, where he lived at 3 Lombard Court. From there he patented a tape-controlled telegraphic transmitting apparatus.
He worked for the (British) General Post Office until 1909, by which time he had established a telegraph engineering factory at 55 Goswell Road, London, with Sidney George Brown FRS (above, 1873-1948). Murray sold installations to Germany, Austria, Russia, Sweden, Brazil, Australia and New Zealand. At various times he was based in Berlin and Petrograd (St Petersburg).
In September 1906, Scientific American reported on successful trials in Russia and Germany. It said the system was working between London and Edinburgh, Berlin and Hamburg, St Petersburg and Moscow and sets being installed for London-Dublin, Bombay-Calcutta and Vienna-Prague. Scientific American said that unlike previous systems, the Murray had already paid its own way.
It did much more than that on April 12, 1912, when Murray announced he had sold the North American rights to his system, as it had been developed to that point, to Western Union in New York [the British GPO retained the European rights].
Western Union gradually replaced all of its Morse telegraphs with the new 'teletypewriters'. Murray continued to improve the system for Western Union until 1925, when there was a falling out. In all, Murray was to be issued with 16 patents, of which nine were assigned to Western Union. Western Union still used Murray's code until the 1950s.
While Murray continued to work with Western Union for 12 years, there was mounting tension. Western Union had a close relationship with the Western Electric Company, which supplied it with relays and other equipment. Western Electric was the manufacturing arm of AT&T, and had been employed by Western Union to adapt the Murray system to Western Union’s conditions between 1912-1916. In 1914, AT&T made Murray a counter-offer, to transfer his rights, but Murray considered it inadequate. AT&T responded, according to Murray, with a veiled threat that "they could get round any patents”. Murray closed one of his “Press the Button Telegraphy” articles in the Telegraph and Telephone Journal in 1914 thus: “I ask for the sympathy of all fair-minded men against the unscrupulous tactics of a big American company, which is trying to deprive me of the legitimate fruits of my labour.”
In January of that same year, The New York Times had called Murray the man “who is taking the leading part in revolutionising the telegraphic service to the world”. I have always believed in accepting the word of authoritative journals such as The New York Times, as published at the time these developments were happening, over the half-baked observations of amateur historians, looking back selectively more than 80 or 90 years later.
In 1925, Murray severed his ties with Western Union and instead switched the North American rights to his patents to Creed, who in turn sold out to IT&T three years later. Thus, by 1930, Murray’s principles had spread from Western Union to Creed and Morkrum-Kleinschmidt and on to IT&T and AT&T. They were there to stay.
Murray was to write that he had “spent 12 years and some thousands of pounds in developing the Murray Automatic, my high-speed typewriter, which worked at nearly 200 words [1200 letters] a minute." It had proved to be a worthwhile investment of time and money. His New Zealand relations told Murray’s Californian biographer Bob Mackay that Murray did “become rich”. Even if, as an 1923 advertisement in the Telephone Journal, put it, “The Murray Multiplex is the Rolls-Royce of Printing Telegraphs in performance, but not in price."
In 1940, Murray was diagnosed with a brain tumour. He underwent three operations in Monte Carlo, where he had settled in his new career as a philosopher. Later he and his wife Patricia (nee Cosgrove) moved to Switzerland (The couple had married when Murray was 47; they had no children). Murray died, aged 79, at Territet, a suburb of the town of Montreux, on July 14, 1945. Patricia remained in nearby Veytaux.
Late in life, Murray published two books, The Philosophy of Power and The Theory of Control. Murray’s abiding “New Zealandness” is reflected in the dedication to The Philosophy of Power, in which he referred to the “Winds of Ururangi”. Those winds eventually, thanks to Murray, blew our way the QWERTY keyboard on computers, and there it has remained.
Donald Murray in later life

For this post, I am indebted to the following authors, historians and researchers (I won’t mention those who have got the story wrong):
Bob Mackay, Murray family historian, of Berkeley, California, for details of Donald Murray’s life and that of his father, John.
Koichi Yasuoka, for putting me on to this story. See
And historians Alan G. Hobbs, Sam Hallas, Cybrations of Williamsburg, VA, Tom Jennings, Gil Smith, Lloyd Butler,
David M. MacMillan and Rollande Krandall, and finally author Thierry Bardini.


notagain said...

That would make a good movie. Now I want to read his books.

Richard P said...

Fascinating. This is one of your best articles ever. I don't really understand this whole field of teletypewriting, and I can see that I have plenty to learn.

Naturally, like notagain, I want to read Murray's books. I see that the University of Cincinnati has a copy of The Philosophy of Power (in some remote storage location -- probably hasn't been touched for decades). I've ordered it!

Thanks for all your work.

Bill M said...

This is one grand article! Those old photographs are superb. Thanks very much for such a wonderfully informative post I also now have a few more books to add to my list of books to read.

Koichi Yasuoka said...

Thank you for linking my article at Sanseido. Well, I have an information about his marriage. Donald Murray were married with Patricia Cosgrave on January 5, 1912, at New York. He was at the age 46 then. Please see detail here.

ybennett said...

I am looking for someone in Auckland who can type on a 1940's typewriter with speed and accuracy. I am working on a large-scale feature film set in this period in the U.S and Japan and need a typing hand double for our American actor. Please if there are any capable male hands out Please send me an email: thanks!! YB

Unknown said...

I'm aware I'm a little late here, but I have to tell you that Donald Murray also was a war hero, although he was probably not aware of that. And none else seem to have realised it. Some years ago I became interested in WW2 cypher technics, the reason for this was that Swedish crypt analysers were very succesful in breaking German traffic sent over Swedish landlines beetwen occupied Norway and Germany. Arne Beurling, a Swedish matematician, managed to completly recover the inner working of the Siemens T52ab machine in just two weeks. I can only briefly describe how Beurling was helped by Murrays changes to the baudot code, but the essential change was that Murray used separate codes for space, letter shift and figure shift. Operators used to combine space with letter shift to avoid garbled text becoming longer than one word and this made it easy to find spaces beetwen words. It also gave Beurling a head start as he immediatly realised that the machine in addition to XOR-ing the code bits also transpositioned them afterwards. Of course sloppy, as usual, German operators were very helpful. Here's a paper on the machine.