Rhodri Johns Speech
Donald Watts Davies - Blue Plaque Unveiling-Treorchy 25 July
2013
Mayor of Rhondda Cynon Taf Cllr Ann Crimmings,
honoured guests, including Donald Watts Davies’ son Michael Davies and his
family, Donald’s twin Sister Marion, Ladies and Gentlemen;
My name is Donald
Rhodri Johns. My mother, Margaret Johns (nee Davies), born in Treorchy, is a
first cousin of Donald Davies. I will provide some family history, a summary of
Donald’s achievements, and my perspective as a scientist.
Family Background
Donald Watts
Davies was born at home in 102 Dumfries
Street, Treorchy on the 7th June 1924.
He was a second twin of an undiagnosed twin pregnancy. His sister Marion had
been born safely when the doctor pronounced that another baby was on its way. Donald's
birth was a difficult one and he was pronounced stillborn by the doctor. The doctor declared ‘one is enough’. However his grandmother or Mamgu, Margaret Anne Davies,
revived him thereby saving his life. The story stands as part of our family’s
folklore.
Griffith Davies, Donald’s
grandfather, had arrived in Treorchy from Pembrokeshire in the early 1890's
to work in the coal mines. Griffith’s mother
Jane Watts was descended from Scottish migrants that had settled in the Teifi Valley
in the early 1700s. Margaret Anne was born in the coastal town of Dinas Cross, North Pembrokeshire
and was the eldest sibling of a large family. Griffith and Margaret Anne married
in Porth in 1895.
They had four children. Benjamin (my grandfather), John, Annie and Joshua.
John Watts Davies, Donald and Marion's
father, married Hilda Stebbens, a native of Portsmouth, but a year after the twins’
birth, whilst Hilda was away visiting family, tragedy struck and her husband John
Watts Davies died from gastro-enteritis.
Hilda took Donald and his twin sister back to her home town of Portsmouth,
where he went to school, eventually attending Portsmouth Southern Secondary
School for Boys.
Donald’s early recollections were documented in a Biography for
the Royal Society
My
first memories are of a small street house with trams passing by. The trams
waited at a crossing place in the middle of the street before going on to Bradford Junction, where several tram lines met. Their
clanking and squealing was a familiar and homely sound. Other early memories
are of train journeys to Newport, a small
village in Pembrokeshire, south west Wales. Mamgu and Dadcu had retired
there, living near the smallholding close to Newport Castle
which was farmed by their daughter and son-in-law. There was a working water mill,
grinding flour, which had large wooden gears. Staying on the farm was a great
thrill for us and the foreign language (Welsh) that was spoken around there
made it an exotic place.
Donald’s mother worked for the Post Office,
this gave him an early familiarity with the way telephone networks worked.
Education and Career
He received a 1st class degree in Physics at Imperial
College London, 1943 and then joined the war effort working on the nuclear
weapons Tube Alloys project. He then returned to Imperial College
achieving a first class degree in mathematics (1947).
In 1947, after hearing a lecture about the
early development of Computers by John Womersley, and excited by the potential of the new technology, he
applied successfully to join the National Physical Laboratory (NPL).
There he worked briefly with the famous Alan Turing,
whose remarkable pioneering plans for a computer to be called the Automatic
Computing Engine or ACE had not yet been
realized.
Donald found mistakes in Turing’s famous paper ‘On
Computable numbers’ but Turing’s response was apparently not very favourable
when these were pointed out to him.
In
1948, after Turing's departure, his group was reconstructed as a balanced team
of mathematicians and engineers including Donald, and led by F. M. Colebrook. The result was the successful Pilot
ACE computer which ran its first program on 10th May 1950. The Pilot Ace, now
in the Science Museum, was at the time one of the
fastest computers in the world. A commercial spin-off, Deuce, became one of the
best-selling British machines of the 1950s.
Recommending DWD for a Commonwealth Fund fellowship in
1954, Colebrook wrote:
“D. W. Davies is
one of the most brilliant young men I have ever met; outstanding not only in
intellectual power but also in the range of his scientific, technical and
general knowledge. He is equally unusual in his ability to apply this knowledge
to mechanical and electrical design and even to the actual construction of
complex equipment. He is, for example, one of the very small number of persons
who could draw up a complete logical design of an electronic computer, realise
this design in
actual
circuitry , assemble it himself (with a high probability that it would work as
designed) and then programme it and use it for the solution of computational
problems.”2
He recalled one by product of his work in a humorous and modest
way
A by-product of my electromechanical work
was to finalise my quest for a machine to play noughts and crosses.
In the evenings I built a machine using
Post Office relays. NPL put it forward as an exhibit at the Royal Society
Soirée in 1949.
Its clicking noise and nice display
attracted people, so it received more attention than it deserved. I felt rather
bad for the other exhibitors with real scientific achievements to show.
Next day I was on the Daily Express front
page and much more publicity followed. I appeared on TV with Richard Dimbleby
at NPL and then at Alexandra
Palace, I was also
interviewed on children’s TV.1
Donald then worked for a while on applications such as traffic
simulation and machine translation. In the early 1960s, he worked on Government
technology initiatives designed to stimulate the British computer industry.
In
the 50s and early 60s Computers had began to be increasingly used by industry and
universities. Computer communications could be slow and inefficient, and costs
for long distance communications could be overwhelming. These were real barriers
to the creation and growth of computer networks.
In 1965 DWD had an idea: that to achieve
communication between computers, a fast message-switching communication service
was needed, in his idea long messages were split into chunks sent separately so
as to minimise the risk of congestion. The chunks he called packets, and the
technique became known as packet-switching.
A packet was rather like a short telegram, with each
packet having the address of the destination.
A long message would be broken up into a stream of
packets, which would be sent as individual items into the network. This would
enable a single communications line to carry many simultaneous human-computer
interactions by transmitting packets in quick succession. The computers that
acted as the switching centres-would simply receive packets and pass them along
to the next node on the route toward the
destination. The computer at the destination would be responsible for
reconstituting the original message from the packets.
He determined at a very early stage that the
system could work very efficiently
This was the simple but brilliant idea that
enabled computers to communicate data at speed and allowed for the economic simultaneous
use of communications lines by many users. This idea is the basis for modern
digital data communications and the Internet.
Interestingly his
vision was for a UK
national communications network which also would have everyday people as its
users. He was seeing far into the future. In a paper written in late 1965 he wrote
a list of services for a data communications network6.
These included;
People
to people communication (email or messaging)
Remote
access to records
Numerical
calculations
Editing
and typesetting of text
Availability
of goods for sale
Invoicing
Transport
booking
Banking
Betting
In 1966 he was appointed to NPL's autonomics
division and soon turned it into a Computer Science division.
A team was created under his tutelage and set
to work on a packet switching network to translate his idea into reality.
In 1967, at a symposium in the United States, his
packet switching network design was received enthusiastically by America's Advanced Research Project Agency
(ARPA) and the Arpanet and the NPL local network in the UK became the
first two computer networks in the world using the technique.
Spurred by his invention many independent
packet switching networks were set up world wide and the eventual integration of
these networks in the 80s led to today’s Internet.
It seems to me that, just as
Watt, Stephenson, Trevithik and others, through their scientific and
engineering genius and inventions, powered the Industrial Revolution in the 19th
century Donald Watts Davies through his invention of the packet switching networks
helped power the information and communication revolution of the 20th
and 21st centuries.
In 1979 Davies relinquished his managerial post at NPL
to concentrate on technical work. Realising that computer networks would be
used widely only if malicious interference
could be prevented, he started a group to
work on data security, concentrating on the new method of public key
cryptosystems including smart cards. The group built a strong consultancy role
round his expertise; all the major British clearing banks, for instance, used
its services. He retired in 1984 and
continued his work as a data security consultant.2
He was nevertheless very busy Donald himself lists no
less than 84 projects undertaken between 1984 and 1998.1
It is clear that he also made significant
contributions to data security.
Awards
He was the author or joint author of four influential
books, notably the classic Computer Networks and their Protocols published
in 1973. His contributions, in particular his work on packet-switching, were
recognised by the British Computer Society, which conferred on him the John
Player Award in 1974 and a Distinguished Fellowship in 1975; he became its
technical vice-president in 1983. He was appointed CBE in 1983 and a Fellow of
the Royal Society in 1987. He was also a visiting professor at Royal Holloway
and Bedford New College
in 1987.
In 1999 he was awarded the Internet Award
(with 3 others) by the IEEE for his work on packet switching networks ‘the foundation technology of the internet’
In 2007 DWD was inducted into the US
National Inventors Hall of Fame
In 2012, DWD was inducted into the Internet
Hall of Fame by the Internet Society.
As the NPL Biography stated;
"An Amazingly versatile man and an inspiring and much respected colleague"
NPL provided a statement for this event. I shall quote a portion of it here;
"NPL is proud of its legacy in computer science and the role that people. like Donald Davies, played in making the United Kingdom a technological world leader and it is a special honour for his birthplace in Treorchy to be recognised with this blue plaque."
Thanks
Three generations of the Davies family
lived here in Treorchy and on behalf of their descendants here today I would like
to thank the County Borough of Rhondda Cynon Taf for recognising Donald Watts
Davies in this way.
As "The Times" put it:
Donald Watts Davies – The Scientist who enabled computers to talk to each other, and so made the Internet possible 4
-----------------------------------------------------------------------------
Donald Rhodri Johns, BSc (Hons), PhD (Cantab),
FGS
July 2013
D. W. Davies, 'An Historical
Study of the Beginnings of Packet Switching,' Brit. Comp. Soc. J., vol. 44, no.
3,2001, pp. 151-62.
