⟦e190ffa18⟧

TextFile

INTRODUCTION TO
DATA COMMUNICATIONS SYSTEMS

First Edition
A/S REGNECENTRALEN August 1978
Information Department RCSL 42-i0847\f

F_ Author: David McLeod

KEY WORDS: Terminal, communications network, data transmission,
modem, on-line, batch processing, real-time applicati-
ons, input/output, telecommunications, files.

ABSTRACT: The purpose of this publication is to acquaint the
non-technical reader with the basic concepts of data
communications systems. It also presents a conceptional
introduction to the combination of data communications
and data processing techniques with little reference to
programming and other items of detail that are involved
in the main subject. The reader is introduced to the
use of the various types of data communication
equipment and applications in data communications.

TABLE OF CONTENTS

List of illustrations

1 INTRODUCTIONpage 7

2 COMMUNICATIONS13
2.1 A history of communication 13
2.2 A history and development of computers18
2.3 Development of data communications systems23
2.4 Satellite communications26
2.5 Communications and the environment28

3 EQUIPMENT AND SYSTEM SOFTWARE USED IN DATA
COMMUNICATIONS SYSTEMS31
3.1 The parts of a data communications system 32

4 COMMUNICATION FACILITIES35
4.1 Switched and non-switched lines35
4.2 Simplex, half-duplex and duplex channels38
4.3 Categories and transmission speeds of channels40

5 CHARACTERS AND CODES41
5.1 Information codes42

6 ERROR CHECKING, CORRECTION AND DETECTING CODES 44
6.1 Parity, blocking, buffering 45

7 SYNCHRONOUS AND ASYNCHRONOUS TRANSMISSION 51
7.1 Line control or data link control53

8 MODEMS 55

9 TERMINALS 56
9.1 Serial and parallel transmission61
9.2 Transparent, transparency62
9.3 Terminal simulation 63

10 NETWORK CONCEPTS 66
10.1 Distributed processing68
10.2 Packet switching 70
10.3 RC NET and network control program 70\f

11 APPLICATION SOFTWAREpage 72

12 GLOSSARY 73

List of illustrations.

Figure No.
1. Batch processingpage 9
2. Real-time processing10
3. Ages old traveller13
4. Early Roman runner13
5. Transportation of messengers for centuries14
6. Indian smoke signals14
7. Carrier pigeon14
8. First postal service15
9. First telegraph service15
10. The telephone 16
11. Mail train16
12. Air mail17
13. Radio17
14. Television17
15. Microwave transmission17
16. Satellite communication26
17. The parts of a data communications system32
18. A switched line36
19. A non-switched or private or leased line37
20. A trunk line37
21. Transmission types39
22. Direction of transmission39
23. Transmission speeds40
24. Combinations of bits form characters43
25. Types of errors44
26. Two-way parity checking48
27. Asynchronous transmission51
28. Synchronous transmission52
29. Airline reservation terminal56
30. Alphanumeric Display/Keyboard57
31. Terminal matrix printer57
32. Single line terminal59
33. Two line terminal59
34. Multiple keyboard terminal60
35. Multiple keyboard terminal with concentrator60
36. RC 3600 Minicomputer simulation64
37. Terminal simulation 65
38 RC 3600 Data Entry System66
39. RC 3600 terminal67
40. RC 3600 front-end computer68
41. RC Net data network71\f

F_1INTRODUCTION

Each science has its own terminology.
When a scientific or technical subject is introduced it is both
important and necessary to define its main terminology so that it
may be communicated easily to the reader.

While some of the more commonly used terms and definitions of the
subject are explained with the text this publication includes a
glossary.

d_a_t_a_ A general expression used to described any group of num-
bers, alphabetic characters or symbols which denote any condi-
tions, value or state, e.g. all values and descriptive data oper-
ated on by a computer program but not the program itself. The
word data is used as a collective noun and is usually accompanied
by a singular verb: 'data are' may be pedantically correct but is
awkward to say and therefore awkward to understand. Data is some-
times contrasted with information, which is said to result from
the processing of data, so that information derives from the
assembly, analysis or summarizing of data into a meaningful form.

d_a_t_a_ _p_r_o_c_e_s_s_i_n_g_ The operations performed on data usually by au-
tomatic equipment, in order to derive information or to achieve
order among files. A data processing system may incorporate
clerical functions and ancillary machine operations as well as
all arithmetic and logical operations performed by a computer.

d_a_t_a_ _t_r_a_n_s_m_i_s_s_i_o_n_ Pertaining to the automatic transfer of data
from one computer system to another, or to and from a central
computer and distant data collection points. The data may be
transferred by special equipment using either telegraph or tele-
phone circuits, or by radio link. The speed of transmission is
largely governed by the characteristics of the data transmission
line or channels.

Therefore, a data communications system may be defined as "a data
processing system with data transmission capabilities". From this
definition a user can assume that he can process data which has
been transmitted to his data processing system, also known as a
computer system, from "input/output" devices, called "terminals",
that are remote from his central data processing system. Similar-
ly, he can assume that information from his central data proces-
sing system may be transmitted to the terminals at remote loca-
tions. Data communications systems with a number of interconnec-\f

ting terminal locations are called "networks".

This publication, while not being technically oriented, will at-
tempt to present a clear picture of the communications resources
which can be used directly, adapted to, or modified for data trans-
mission.

By its own nature, data processing generally implies that source
data will be collected at remote points and brought in to a cen-
tral processor. In early systems this transportation problem was
solved by mail or messenger. Now, since electronics have advanced
many steps forward, there are devices to transfer data directly
between remote terminals and central computers. In most cases the
familiar telephone network, often called a "telecommunications
network", is the medium of exchange. So prevalent has this be-
come that in spite of paying large sums of money annually to the
telephone companies for data transmission, many users still look
at the telephone net work as pairs of wires, conveniently avail-
able. In many cases a better understanding of the telecommunica-
tions network and the constraints under which it must operate as
a monoplistic venture can lead to more efficient, less costly
data communication systems. In other cases it can contribute to a
more harmornious relationship between the customer and the
telephone company. In still other cases it can avoid
misapplication.

When data is transmitted via transmission links e.g. telephone
lines, the data is said to be transmitted "on-line". In a data
communications system data is originated at remote terminals and
entered on-line to the system. There is no need for human inter-
vention in the transmission of data from the terminals to the
central data processing system, also referred to as the "central
computer", and vice - versa, and no paper work is created. With
on-line processing the records at the central computer are updated
immediately upon the entry of each transaction and the computer
replies to the terminal/s immediately if required to do so. The
user builds a set of special conditions into his system according
to his needs and therefore only under certain conditions will the
system reply to terminals rather than on the receipt of every
transaction. System replies are usually exception reports and
messages to alert or request special action of the personnel at
the remote terminal site.
\f

Figure 1.
Batch processing \f

T_ It is unlikely that all information would normally be entered in
to and updated immediately in a data communications system. Cer-
tain applications such as inventory re-order reports may be pro-
cessed daily, factory payrolls weekly and customer account state-
ments monthly. This is called "batch" processing and its fre-
quency is decided by the user. A data communications system will
thus have both batch processing operations and those applications
where up-to-the minute information is required at all times and
must be updated immediately with each transaction. The latter are
called "real-time" applications and enquiries may be made as to
the status of items in the computer records or "files" at any
time.

Figure 2.
Real-time processing \f

T_ Other data has to be stored and accumulated over longer periods
e.g. monthly, quarterly and annual data in the current year as
well as data from each of the months in the previous year. This
data is used to provide the user with comparative information.
Thus performances can be measured between the current period and
the previous period or between the months, quarters and annual
figures of the current year and similar periods of the previous
year.

Data communications systems give the user certain real benefits.
Among these are:

T_1Up-to-the minute information on vital operations in a company's
individual and overall activities is available regardless of
geographical location, e.g. daily operating reports on costs,
&_sales and profits.

T_2 Direct control on costs and activities can be exercized as soon
as actions occur or are initiated. Management does not have to
&_ wait until, perhaps, the month-end to receive operating
results for decision making.

T_3 Managements in different locations have the same up-to-date and
&_consolidated data with which to review activities.

T_4 In manufacturing, production cycle times can be shortened by
direct control and economic benefits accrue from lower inven-
&_tory levels, less investment in work-in-process. Efficient in-
ventory control provides less work-stoppages due to lack of raw
materials and better customer service by having the right items
in stock when they are ordered.

T_5On-line processing of customer accounts with immediate transac-
tion updating provides real-time information (with immediate
&_updating) for excellent customer service.

T_ 6 Drops in performance in various activities can be noticed early
and appropriate corrective action can be initiated without de-
&_ lay.

T_ 7 Although it is known that computers can process large volumes
of data at very high speeds, getting the data to and from the
&_ computer itself by mail or courier system can be very slow.
Therefore, this communication lag is eliminated in a data com-
munications system and so is a great amount of paper work.

The facilities and benefits of data communications systems can
be taken advantage of by all types of organzations such as ma-
nufacturing industries, financial and banking organizations,
government and educational institutions, public utilities,
transportation and commercial enterprises. There are data com-
munications systems to suit the requirements of small, medium
and large organizations and wherever they are geographically
located. Whether an organization may have a few branches of
operation in the same city or in different cities and towns in
the same country or in a number of countries around the world,
there are no limitations to the distances involved. The bene-
fits of data communications systems are equally important to
all of them.\f

F_2. COMMUNICATIONS

2.1 A HISTORY OF COMMUNICATION

Communication is the most important action in this universe. With-
out it almost nothing would happen. Man has been communicating in
one way or another since time began for him. History is filled
with accounts of communication between many peoples. As people
moved to other parts of the world the need for communication to
and from them was carried in written form or by word of mouth by
travellers. Man recognized the need for communication for the
knowledge and trade that it brought him.

Figure 3.
Ages old traveller

Later civilizations used messengers for peaceful and aggressive
reasons since it was safer to know about one's enemies and a good
idea to keep in touch with one's allies. The early Greeks, Romans
and other nations had specially trained runners who were capable
of covering great distances. They would carry messages between
the rulers of countries.

Figure 4.
Early Roman runner

Through the ages man developed ways to speed his communications.
The fleetness and endurance of the horse was the transportation
of messengers for many centuries.

Figure 5.
Transportation of messengers for centuries

Code signals were introduced with flags during the day and flares
at night, smoke signals were employed by Indians in North America

Figure 6.
Indian smoke signals

In later years carrier pigeons were used in many countries and

Figure 7.
Carrier pigeon

France introduced the first public postal service in the 15th cen-
tury.
\f

Figure 8.
First postal service

Oersted discovered the relation between magnetism and electricity
in the year 1819. In 1834 a telegraph wire was strung over the
roof tops of the city of Gottingen. Samuel Morse set up a
telegraph system in 1844 between Baltimore and Washington. The
information was a code of dashes and dots.

Figure 9.
First telegraph service

Then came Bell's invention in 1876 of the first practical tele-
phone. Each telephone was connected only to the telephone at the
other end and not to any other. In order to connect a number of
telephones in one locale, an exchange or switching office was
employed. \f

Figure 10.
The telephone

As the demand for more telephones increased, a method of increas-
ing the number of circuits without adding more wires became impe-
rative. Research led to the combining of several voice circuits
using the same transmission line. This method is called "multi-
plexing".

Another big advance in the field of communication was the inven-
tion of radio by Marconi.

In 1918 the Bell system placed a system designated as Type "A"
into service. Type "A" service provided four two-way channels for
use over a single open-wire line. Type "B" provided three two-
way channels using different frequencies for each direction of
transmission. Both Type "A" and "B" systems used "amplitude modu-
lation" to super-impose the voice signals onto the carrier fre-
quency. The Type "C" system was introduced in 1925, which used
the best features of the two systems. Technological advances soon
proceeded far enough to permit the development of a 12 channel
carrier system, which operated over cabels and open-wire circuits.
This was done by using a different wire pair for each direction.
establishing a physical four-wire system.

Figure 11.
Mail train

The Telecommunications industry concentrated its peacetime endea-
vors on the war effort as WW II was entered by the United States.
During the war great advances were made in the field of military
communications. These advances were later to have influence upon
commercial communications. \f

Figure 12.
Air mail

With the end of the war in view, men with foresight realized that
there would soon be a demand for expanded communications facili-
ties. With their experience in military communications, they be-
gan to design carrier equipment that would be suitable for commer-
cial purposes.

Figure 13.
Radio

By 1952, the telecommunication field had achieved a high degree
of sophistication. Transistors, silicone diodes and printed cir-
cuits began to have practical applications. One of the develop-
ments has been in the field of miniaturization.

Figure 14.
Television

In the telecommnunications industry a backward glance at the past
is permitted only occasionally. The usual point of attention is
to the unlimited possiblities and applications that lie in the
future. Challenges are being met daily to keep products ahead of
technological changes, and at the same time new demands are being
met with sophisticated communications systems. Indeed, more modern
development in communications technology are microwave and laser
beam transmission as well as communication satellites.

Figure 15.
Microwave transmission\f

T_2.2 A history and the development of computers.

Humans have been calculating since the dawn of history and be-
fore. Stone age man, making scratches on animal bones, tried to
&_ keep track of the phases of the moon. Other prehistoric people
reckoned with pebbles. Indeed, the Latin word Calculus means a
stone used for counting. Perhaps the most enduring calculating
device is the abacus, which was used in China as early as the 6th
century B.C. But the first of really serious efforts to make me-
chanical calculators in which some of the tallying was done auto-
matically, did not come until the 17th century.

By then numbers had become especially important because of great
advances in astronomy, navigation and other scientific disciplines.
More than before, it was necessary to rely on long tables of such
elementary mathematical functions as logarthims, sines and cosines.
Yet compiling these essential tools often required years of sla-
vish toil.

The word "computer" comes from the Latin verb "computare", which
means to count or reckon. A digital computer is a machine that
reckons or calculates with digits.

In 1642, a Frenchman named Pascal invented what was perhaps the
first actual accounting machine. This machine was used to figure
currency in a custom's house. It was basically a hand-operated,
gear-driven counter, with addition performed by turning a wheel
distance equal to the currency to be added. It was the first in-
strument with provision for an automatic carry into the next order
column when the sum of a column exceeded nine.

In 1820, an American named Thomas invented a desk calculator
which was used for the next 100 years.

Still, mathematical illiteracy continued to plague Europe. In Eng-
land in the early 19th century, Charles Babbage, an idiosyncratic
mathematician and inventor of the railroad cowcatcher and the
first tachometer, was becoming increasingly incensed by the
errors he found in insurance records, logarithm tables and other
data. His fetish for accuracy was so great, in fact, that after
reading Lord Tennyson's noted line "Every moment dies a man/Every
moment one is born", he wrote the poet: "It must be manifest that
if this were true, the population of the world would be at a
standstill." Babbage's recommended change: "Every moment dies a
man/Every moment 1 1/6 is born."\f

In 1822, Babbage began work on a machine, called the difference
engine, that could help solve polynomial equations to six places.
The Chancellor of the Exchequer was so impressed by the machine's
potential for compiling accurate navigational and artillery tables
that he subsidized construction of a still larger difference engi-
ne that could compute to 20 places. Unfortunately, the metalwor-
kers of Babbage's day were not up to making the precision parts
required, and the machine was never completed. But Babbage had a
bolder dream: he wanted to build a machine, which he dubbed the
analytical engine, that could perform any arithmetical and logi-
cal operations asked of it. In effect, it would have been program-
mable - that is, a true computer instead of a mere calculator.

To "instruct" the machine, Babbage borrowed an idea that had just
revolutionized the weaving industry. Using a string of cards with
strategically placed holes in them, like those in a piano roll,
the Frenchman Joseph Marie Jacquard automatically controlled
which threads of the warp would be passed over or under with each
pass of the shuttle. Babbage planned to use the same technique to
program his machine, instead of the positions of threads, the ho-
les in his cards would represent the mathematical commands of the
machine. Wrote Babbage's mathematically knowledgeable friend,
Lady Lovelace, daughter of the poet Lord Byron: "We may say aptly
that the analytical Engine weaves algebraical patterns just as
the Jacquard loom weaves flowers and leaves."

Babbage's loom, alas, never wove anything. By the time the
eccentric genius died in 1871, he had managed to put together a
few small parts; only his elaborate drawings provide a clue to
his visionary machine. Indeed, when Harvard and IBM sciensists
rediscovered Babbage's work in the 1940s while they were building
a pioneering electromechanical digital computer called Mark 1,
they were astonished by his foresight. Said the team leader,
Howard Aiken: "If Babbage had lived 75 years later, I would have
been out of a job."

In 1886, Dr. Herman Hollerith invented a machine which used
punched cards. Hollerith was head of the U.S. Bureau of Census.
He discovered that the 1880 census was not yet completed. All
calculations were being made by hand. He set to work to find a
way in which all recording, tabulating and analyzing of facts
could be done by machine. His solution was to record the facts of
any situation by punching holes in a definite code in a piece of\f

paper. He cut the strips of paper into a standard size and shape
and thus had a "card" for each situation. Once the card was
developed, Dr. Hollerith developed a sorting device and opened up
a new field of computing aids.

The punched card system was the most widely used method of data
processing before the advent of electronic computers. The punched
card industry developed into a large business in the 1940's and
1950's.

IBM, Univac and ICL developed into multinational companies
through marketing their punched card systems. The machines were
mainly electro-mechanical and an installation would usually con-
sist of keypunch machines to punch the source data into cards,
sorters, collators and tabulators. This business was taken over
by computers at an increasing rate from the latter 1950's.

Dr. H.H. Aiken of Harvard University directed a project which
started in 1939 and was climaxed by the construction of the
automatic sequence controlled computer in 1944. This computer The
Mark 1, was built by International Business Machines for Harvard
using IBM's electromechanical business machine components.

Although the Mark 1 was electromechanical in nature, it marked
the appearance of the first of a long line of large digital
computers.

The Harvard machine occupied a large room and sounded in the words
of Physicist-Author Jeremy Bernstein, "like a room full of ladies
knitting." The noise came from the rapid opening and closing of
thousands of little switches, and it reprensented an enormous in-
formation flow and extremely long calculations for the time. In
less than five seconds, Mark 1 could multiply two 23-digit num-
bers, a record that lasted until ENIAC's debut two years later.
But how? In part, the answer lies in a beguilingly simple form of
arithmetic: the binary system. Instead of the ten digits (0
through 9) of the familiar decimal system, the computer uses just
the binary's two symbols (1 and 0). And with enough 1s and 0s any
quantity can be represented.

In the decimal system each digit of a number read from right to
left is understood to be multiplied by a progressively higher
power of 10. Thus the number 4,932 consists of 2 multiplied by 1,
plus 3 multiplied by 10, plus 9 multiplied by 10 x 10, plus 4
multiplied by 10 x 10 x 10. In the binary system, each digit of a\f

number, again read from right to left, is multiplied by a progres-
sively higher power of 2. Thus binary number 11010 equals 0 times
1, plus 1 times 2, plus 0 times 2 x 2, plus 1 times 2 x 2 x 2,
plus 1 times 2 x 2 x 2 x 2 - for a total of 26.

T_ BINARY NUMBERS
and their decimal equivalents
1=1 101=5 1001=9 1101=13
10=2 110=6 1010=10 1110=14
11=3 111=7 1011=11 1111=15
&_ 100=4 1000=81100=12 10000=16

Working with long strings of 1s and 0s would be cumbersome for
humans - but it is a snap for a digital computer. Composed mostly
of parts that are essentially on-off switches, the machines are
perfectly suited for binary computation. When a switch is open,
it corresponds to the binary digit 0: when it is closed, it
stands for the digit 1. Indeed, the first modern digital computer
completed by Bell Labs scientists in 1939 employed electrome-
chanical switches called relays, which opened and closed like an
old-fashioned Morse telegraph key. Vacuum tubes and transistors
can also be used as switching devices and can be turned off and
on at a much faster pace.

But how does the computer make sence out of the binary numbers
represented by its open and closed switches? At the heart of the
answer is the work of two other gifted Englishmen. One of them
was the 19th century mathematician George Boole, who devised a
system of algebra, or mathematical logic, that can reliably deter-
mine if a statement is true or false. The other was Alan Turing,
who pointed out in the 1930s that, with Boolean algebra, only three
logical functions are needed to process these "trues" and "falses"
- or, in computer terms, 1s and 0s: The functions are called AND,
OR and NOT, and their operation can readily be duplicated by sim-
ple electronic circuits containing only a few transistors, resi-
stors and capacitors. In computer parlance, they are called logic
gates (because they pass on information only according to the ru-
les built into them). Incredible as it may seem, such gates can,
in the proper combinations, perform all the computer's high-speed
prestidigitations (quick actions).

The first electronic computer was the Electronic Numerical
Integrator and Calculator, or ENIAC, built in 1946. The ENIAC
used 18000 vacuum tubes as storage elements, instead of relays\f

and switches as used in the Mark 1. The fact that vacuum tubes
were used at all represented a considerable venture in computing
techniques, since the performance of tubes at the time was not
very reliable.

As an example of the improvement in arithmetic speed between Mark
1 and ENIAC, consider the addition of two numbers. The Mark 1 re-
quired 300 milliseconds (millesecond = 1/1000th of a second) to
perform this task, whereas ENIAC could do the same thing in
two-tenths of a millisecond.

Also in the year 1946 the designers of ENIAC resigned from the
University of Pennsylvania and set up their own firm. They began
to develop an electronic computer which could handle alphabetic
as well as, numeric data. This machine was called the Universal
Automatic Computer, or UNIVAC, first produced in 1951.

England's University of Manchester began design of their Mark 1
computer in 1946. The first goal of the Mark 1 was to construct a
realistic test environment for a novel digital store - the elec-
trostatic Williams Tube. The prototype Mark 1 simply consisted of
a 32 bit x 32 bit Williams Tube store plus elementary computational
facilities. Nevertheless, when it successfully ran a 52-minute
factoring program on June 21, 1948, it became the first general
purpose stored-program computer to work.

The University of Manchester then went on to develop the famous
Atlas computer. That machine, while not a commercial success sin-
ce just three Atlas machines were sold, was nonetheless the ar-
chitectural father of many later powerful computers.

Additional developments enabled manufacturers to incorporate more
speed and reliability into their digital computers. Among these
developments were a memory composed of magnetic cores and solid-
state transistorized circuitry.

The electronics industry has undergone a progressive miniaturi-
zation of equipment parts as well as entire electronic circuits.
From the early vacuum tube circuits, the progression has been to
transistorized printed circuits to thick and thin film integrated
circuits.
\f

Today, in the 1970's, with the arrival of the new technology of
integrated circuits (IC) and microprocessor "chips", a whole new
generation of computers is now in development. The microprocessor-
chips, first released to the electronics market in 1971, contained
2250 transistors in an area one sixth of an inch long and one
eighth of an inch wide (4.2 millemeters by 3.2 millemeters). This
meant that nearly all the elements of a computer's central pro-
cessing unit were concentrated on a single silicon chip.

The benefits of miniaturization have been to drastically reduce
the size, cost and electrical drain of any equipment in which
they were used. However, they have had an even greater effect on
the speed of computations in computers. Electric current passes
through circuits close to the speed of light which is about 1
foot (30 centimeters) per billionth of a second. Even so, an
electrical impulse required a significant fraction of a second to
move through the miles of wiring in the early, large computers.

Now even circuitous routes through chips could be measured in
inches and traversed by signals in an electronic blink.

T_ 2.3 Development of data communications systems.

During the 1960's many people in the data processing industry and
&_ other organizations recognized the potential advantages and bene-
fits inherent in a combination of the two technologies of compu-
ters and communications. Both technologies had a complimentary
effect in providing a vast scope of application possibilities for
the user. Thus the development of data communications systems was
started on its way.

Although communications facilities were developed over a long pe-
riod of time we can see that data processing systems were a com-
paratively recent development and still more recent was the mer-
ger of the two technologies.

In 1944 the input data for scientific calculations was sent by
telegraph to the engineers operating the IBM Mark 1 computer at
Harvard. The results were then sent back to the user by telegraph.
For some years before this the telegraph network had facilities
for reading data that was punched into paper tape. The data was\f

transmitted along telegraph lines. At the receivng end a decod-
ing unit translated the telegraph message into punched paper tape
and this was fed into a printer which produced a printed output
of the telegraphed data on paper. In 1941 IBM developed the punch
to-tape and tape-to-punch machines. This was in the early days of
data processing when the punched card system was the most widely
used mechanical method for data analysis. The data that had been
recorded on punched cards was converted by the punch-to-tape ma-
chine into paper tape data.The paper tape was then fed as input
into a telegraph network for transmission to another location.
The receiving location punched the data into paper tape and then
the tape-to-punch machine was used to convert the paper tape into
punched card data for processing at that site.

The above were among the pioneering efforts towards data communi-
cations systems. During the 1940's and early 1950's the develop-
ment in computers and data communications was comparatively slow.
However, the earlier efforts pointed out the value and convenien-
ce of ideas that would be used in future communications systems.
Remote terminals could be linked to computers and this meant that
communications would bring the power of the computer to the user
wherever he is. Geographical considerations were no longer a
theoretical problem. The basic concepts were in the minds of many
people in the communications and data processing industries.

In the mid and latter 1950's there was a quantum leap forward in
the computer industry. This was brought about by the invention of
solid-state technology. This was a circuitry consisting mainly of
trasistors, diodes, capacitors and core storage. Computers with
electronic speeds and large storage could be manufactured at re-
markably lower costs. Large-scale production commenced. Modern
computer technology had arrived and the widespread use of compu-
ters had begun its explosion.

Much of the leadership in advanced communications came from tele-
metry systems developed for missile and space technology.

The new solid-state technology was just as beneficial to the
communications industry as it was in the manufacture of computers.

The new breed of computers carried out arithmetic and other pro-
cessing functions at internal speeds which were measured in mille-
seconds and microseconds, i.e. in thousandths and millionths of a\f

second. The internal speed of the computer could handle a number
of input/output peripheral units simultaneously. These would be
card and paper tape readers, magnetic tape and disc storage units
as well as printers and keyboard - typewriter consoles. Compared
with the speed of the computer, the communication lines of
telephone networks and communication devices were slow. In
addition to handling the usual above-mentioned peripheral units
of a data processing system, the speed of the computer enabled it
to service many remote terminals concurrently.

This meant that a computer could be shared by many users. The
concepts of time-sharing and multiprogramming were developed and
implemented. Significant advances were made in programming and
software i.e. programs which assist and support programmers to
get the best use from their computers. Large networks with hund-
reds of terminals and lines took advantage of geographically
distant computer centers.

This great development did not take place without its problems.
Countless meetings were held between the executives of electro-
nics, computer and telephone companies in many countries to ne-
gotiate data transmission rates, interfaces for a variety of
equipment, transmission speeds and many other facets of the data
communications industry. A new breed of professional was necessa-
ry to understand the status and structure of a growing new indu-
stry. A whole new range of equipment specially designed for data
communications was necessary to provide customers with the hard-
ware to handle their applications.

This development was aided and stimulated by the U.S. military in
their planning of air defense systems for the North American con-
tinent. These systems required large networks extending over vast
areas of land and ocean. The networks were later extended to cover
Europe. Large volumes of operational data from radar and weather
stations were gathered and analyzed in a continuously updated in-
formation system. The air defense personnel could then plan im-
mediate and effective actions. Once again many new technological
developments took place in the development of equipment and pro-
gramming necessary to supply the military with systems of 100%
reliability.

2.4 Satellite communications

The Communication Satellite Act of 1962 called for the establish-
ment of a new and unprecedented private corporation to act as an
instrument of the United States in establishing a word-wide com-
munication satellite system.

A new corporation was formed, the Communication Satellite Corpo-
ration (Comsat).

After holding meetings with a number of countries around the glo-
be, COMSAT drafted the Interim Agreements and opened them for
signature in August of 1964.

Under the Interim Agreements, COMSAT represents the United States
in INTELSAT, has 35% ownership in the system and acts as
manager.

The successful launching and orbiting of Telstar and Relay in
1962, with their low, non-synchronous orbits and brief 30 minute
transmission periods assured the future of satellite activity.

Figure 16.
Satellite communication

It is an interesting fact that the concept of a geo-stationary
satellite was first published in 1945 in an article by A.C.
Clarke. \f

The world's first commercial communication satellite was orbited
on April 6, 1965. This synchronous satellite, named Early Bird,
was placed in a geo-stationary equatorial orbit over the Atlantic
Ocean. From its vantage point above the ocean, Early Bird linked
the U.S. and Europe with 240 voice circuits and made live televi-
son commercially available across the Atlantic for the first
time. Early Bird has far exceeded its expected lifetime of 18
months.

Early Bird, designated INTELSAT I was the first of the INTELSAT
series. By September 1967 four INTELSAT II satellites had been
launched. INTELSAT I and II satellites were designed with 240
channel capacities.

By May of 1969 three INTELSAT III satellites had been placed in
orbit, one each over the Atlantic, Pacific and Indian Oceans. The-
se three satellites are capable of handling 1200 separate two-way
telephone conversations, or four TV channels or any combination
of the two. A few years later Western Union established its own
satellite "Westar" in synchronous orbit 22,300 miles (35,680 km.)
above the central United States. Intelsat IV and IVA satellites
were also poised over the Atlantic in 1977. Intelsat V is sche-
duled to begin service in 1979.

The possibilities confirmed by the success of the synchronous sa-
tellite were revolutionary. With a minimum of three satellites,
global coverage could be achieved. This would provide telephone,
radio and TV services, a vast communication and navigation ser-
vice for aircraft and shipping.

When Early Bird went into service in 1965, there were only a few
experimental earth stations in the United States, Japan and Euro-
pe. By mid 1969, 25 earth stations were operating in 15 different
countries. 21 additional earth stations in 14 countries were ei-
ther under construction or contract, with many more in the
planning stage.

The Soviet Union, on October 4, 1957, launched Sputnik I. This
event was man's first step into the field of earth satellites.

Only seventeen days after the launching of Early Bird, the Soviet
Union successfully orbited their Molniya I on April 23, 1965.
\f

To date, Russia has launched several Molniya-class satellites in
12 hour elliptical orbits. Successful transmission of many types
of signals, including color television, have been carried out in
joint experiments with France. While the Russian satellites have
an apparent lack of channel capacity they do have a high-powered
transmitter, 40 watts.

Modern satellites are capable of transmitting all forms of commu-
nications simultaneously - telephone, telegraph, television, data
and facsimile. Satellites can operate competitvely with cable
networks and have the advantage of multipoint, multiple access.
This means that synchronous satellites can make direct communica-
tions possible between all countries having earth stations within
a satellite's line of sight.

In addition to people, computers in one country can talk to com-
puters in other countries at speeds many times faster than it is
possible by conventional means.

High-quality telephone service is available from the United
States to a growing number of countries that were previously dif-
ficult to reach by cable or short wave radio transmission.

Weather maps are being transmitted from one country to another to
assist airline pilots in charting plans for flights.

The most dramatic role of the communication satellites is the
part they play in the space program.

Today, in 1978, the data communications industry has such a wealth
of experienced personnel, expertise, know-how, manufacturing ex-
cellence and professionalism that it seems as though any user's
requirement in data communications can be provided for. And yet
the future promises to be just as exciting and beneficial to the
user.

T_2.5 Communications and the environment

Proper application of today's communications technology can bring
&_ people together and improve the atmosphere in which they live.
Present communication links can be expanded for environmental
channels. Environment includes not only geographic features, but
also the culture of an area. Remote data collection and centrali-
zed computer analysis of the data can provide an efficient means
of measuring, analyzing, and correcting environmental pollution.
\f

Although it is not physically or financially feasible to establish
manned laboratories in every geographic location where pollution
is most likely to occur, it is possible, by unmanned data collec-
tion stations, to sample the surroundings and transmit informati-
on on air, earth and water conditions to a central processing la-
boratory for analysis. In this way, computer technology and remo-
te data acquisition can contribute to pollution control.

Prototype pollution monitoring systems are presently in operation.
What look like ordinary navigation buoys are really ocean pol-
lutant detectors. These unattended buoys are able to measure and
transmit such data as water and air temperature, wind speed and
direction, and barometric pressure. A network of ocean monitoring
buoys, or stations can communicate with a central processor
either over a direct microwave or via satellite relay links.

Another pollution detection device employs a patrol aircraft that
measures the changes in microwave radiation from the surface of
the water: thereby, determining what the pollutant is and how
thick the spill is.

Information can be transmitted from the data collection points to
a central processor by microwave techniques. For getting informa-
tion from the remote data collection points, satellites seem to
offer a convenient means. A network of satellites and a surface-
probing sensor system may be used to study natural resources. In
addition to the oceans and air, this network can take inventory
of what, where, and how well forest and crops are growing and
the condition of the soil and its ability to be put to work.

Aerial data collection can also be used to map ocean currents,
ice and other navigational hazards. Fish and other marine biology
of interest, as well as pollutants, can be studied for the sea-
food industry, shipping and marine ecology.

The computer's role in this overall environmental management sy-
stem is that of a soothsayer - if, for example, a decision were
made to irrigate thousands of square miles of desert to create a
new agricultural area, the computer could predict such things as
the plan's affect on: climate, population, water resources and
international trade. With the mass acquisition of data and sophi-
sticated computer processing, it may be possible to stem the tide
of diminishing resources, and pollution of the existing resources.

Solving the problems of an area's pollution and diminishing natu-
ral resources will do little to improve the environment, if the\f

people in the area are unable to communicate and clear up their
differences.

Through proper education and exchange of ideas it is possible to
bring expectations in line with reality.

Educational television and remote conference participation are
similar to videophone with instantaneous voice and video commu-
nication. This service has the greatest potential for bringing
people together because it is possible to clear up any misunder-
standing that might arise before they have a chance to cause dis-
sention in the ranks.

Expanded means of communication have the potential to provide a
more efficient society with an informed public living in a heal-
thy, plentiful environment. Presently, the possibilities are prac-
tically unlimited. Technology has developed these services, eco-
nimics will dictate their future.\f

F_ 3 EQUIPMENT AND SYSTEM SOFTWARE USED IN DATA COMMUNICATIONS SYSTEMS

Taken as a whole, a data communications system is fairly complex.
It comprises the hardware which is data processing equipment,
communications equipment, telephone lines and terminal equipment.
There is also software which includes programs and programming
languages and transmission codes. The software enables us to in-
struct the system to carry out the functions we want. Most impor-
tant we have the people that make the system work. The personnel
involved in the system are those of the user, the equipment manu-
facturers, the telephone companies (also known as communication
carriers), software companies and sundry ancillary suppliers.
These people all play their part in keeping the system running,
by programming it, feeding it data, maintaining the software,
servicing the hardware and planning the development of their
applications towards more efficient use of the data communications
system.

However, the purpose of this chapter is to analyze the equipment
used in data communications systems and to describe the various
parts and explain their functions.

The main two divisions of a data communications system are the
data processing system and the communications system. The two
systems are interdependent. The following annotated drawing shows
a typical data communications system. Only one terminal is shown
for the simplicity of illustration whereas in practice it is
usual forsystems to have a number of terminals.\f

T_ 3.1 The Parts of a Data Communications System

1. Central
processing unit
2. Display
keyboard/console3. Disc
Arithmetic and storage
logic

Programs9.
4. Paper tapeData
proces-
sing
5. Punched cards10. Modem 6. Tapecenter
storage

11. Transmission
line7. Printer

10. Modem

Display keyboard Printer 8. Terminal

Figure 17.
The Parts of a Data Communications System.

1. The central processing unit controls all the peripheral units
connected with the computer. It performs all the arithmetic and
logic functions in the system and directs programs to operate on
data fed into the computer. A program is a set of instructions
that a computer uses to carry out a job such as calculating and
printing the payroll for a company. In doing so the computer uses
its arithmetic and logic capabilities as well as taking in data
from storage and other input devices and terminals. The computer
then sends the results to be printed out at the main printer 7
and terminals 8 and also stores the latest payroll information.
From this it can be seen that the central processing unit also
controls the flow of data into and out of the computer system as
well as reading data from its own storage and writing it to its
storage units for future use and record keeping. \f

2. The display keyboard console is used by the operator to tell
the computer what jobs to do and it gives the operator any neces-
sary control he may need on the computer's actions.Usually jobs
are carried out automatically on a computer but when an exception
occurs the computer sends a message to the display screen on the
console asking the operator to handle the situation. These
messages vary but examples would be to ask the operator to fit
more pa per forms on the printer or to load a particular magnetic
tape reel onto the tape storage unit.

Display keyboard units are commonlyused in data collection where
numbers of them are used by operators who key in data which is
transferred to magnetic tapes or disc units which are then put on
the computer for processing.

3. Disc storage units are like long-playing LP records with in-
formation stored on them instead of music. A number of these
discs are arranged to spin together in one unit and mechanisms
called 'read/write heads' can read or write data directly to and
from any part of the discs. They are also known as 'direct access
storage devices' and data can be read from them or written onto
them very quickly.

4. Paper tape is punched with holes to represent data and is
often read into a computer as input of data by a device or machi-
ne called a paper tape reader.

5. Punched cards are also used with holes punched in them to
represent data and are read into a computer as input data by a
device or machine called a card reader.

6. Reels of magnetic tape are used for storage of data and are
placed on magnetic tape units which read data from them and
transfer it to the computer and also write data to magnetic tape
from the computer.

7. Computer systems need printers to print out reports, schedu-
les, stock records, analyses and thousands of different types of
printed output. Computer printers are highly capable machines and
have to print at speeds of hundreds or thousands of lines per
minute to handle the output from a computer system. They are also
used for printing out at terminals.

8. A terminal is a device or a group of devices which is situated
at a location different from the data processing centre; to which
it is connected by a transmission line. \f

A terminal is used to input and send data to the central computer
and receives processed data from the computer.

A basic terminal would consists of a keyboard display unit and
perhaps a printer. A terminal can be quite a large collection of
machines too, depending on what jobs it has to do.

A terminal can also have its own capabilities in processing and
does jobs which need to be done at its own site and then sends
bigger jobs concerning the whole system to the central computer.
A terminal can also draw data from the storage at the central
computer.

9. The data processing centre is the primary and main computer
installation in a data communications system. It supplies the
main computing power and storage of information for the user and
his terminals.

10. A Modem is a unit that translates computer data into messages
that it can send along a transmission line and translates
messages from a transmission line into computer data.

11. Transmission lines are described in the next chapter. They
carry the data from computer to terminal and vice-versa or from
computer to computer.

The devices 2 to 7 in the illustration are known as 'peripheral
devices' i.e. they are around and near the central processing
unit. \f

F_ 4 COMMUNICATION FACILITIES

Data is information. For our purposes it is, moreover, information
in the form of numbers, letters or special symbols with a pre-
assigned meaning. If data goes from one place to another, then we
are in the area of data communications. The system of ship-to-
ship flag signals is a data communications system, and so is the
telgraph network, but our use of the term will be more specified.

When we talk about a data communications system we will be prima-
rily concerned with methods by which numbers, letters, and other
symbols are sent back and forth between two places by means of
radio waves or wires. Such a system is called a telecommunication
system.

Until the development of microwave communications, telecommunica-
tions concerned itself predominantly with communications systems
that utilized wires, in particular the lines of the telephone
network. Today there are many ways of designing telecommunications
networks; in the future, perhaps, new developments such as laser
beam technology will be a part of ordinary data communications
systems.

The communications industry refers to the information paths as
communication channels and more broadly as communication facili-
ties.

However, telephone systems are so designed that the data communi-
cations user is unaware of the actual structure of the network
and can proceed exactly as though the communications took place
over telephone lines.

Data communications systems can use the common communication chan-
nels that public carriers or telephone companies offer or specia-
lized facilities from carriers or their privately-owned channels
or a mixture of all 3 categories.

T_4.1 Switched and Non-Switched Lines.

&_ Transmission facilities can be divided into two types:

- switched lines (also called public dial lines)
- non-switched lines (also called Leased or Private or dedicated
lines).\f

A switched line is basically the normal dial telephone facility
and the communication between computer and terminal goes through
the public telephone service. The line is not permanently switc-
hed on between the computer and terminal but dialling the connec-
tion is made by a program in the communication equipment or it
can be dialled manually by an operator.

Figure 18.
&_ A switched line

Switched lines are flexible in that any terminal that can be
reached by dialling can be connected to the system. Switched
lines are economical where low volumes of data are to be transmit-
ted and a non-switched line is an established and permanent con-
nection between the computer and terminal. The non-switched line
requires no dialling every time a message is to be transmitted.
When the volume of data to be transmitted on a switched line is
sufficiently high to warrant a non-switched line, lower trans-
mission costs can result thereafter because a fixed monthly
charge will be applied regardless of the volume of data trans-
mitted and hours of usage. There is an advantage to using a
non-switched leased or private line instead of the public or
switched line. Non-switched lines can be conditioned or techni-
cally upgraded and make them more efficient for data transmission.\f

&_ Figure 19.
A non-switched line

A "leased" or "private" line is connected permanently between the
transmitting machines. Some private line systems are owned by
their users. A private line between two or more private branch
exchanges (PBX) is called a "tie line" or "tie trunk".

Voice lines and telegraph lines can be either switched through
"public" exchanges or "permanently"connected. When you make a
call from your home telephone, you must go through a nearby
public exchange or central office. These wires are permanently
connected from your telephone to the central office.

When you call a friend in another city, your call is routed over
an interoffice trunk. A line between the two central offices in
each city is called a "trunk". Trunks are designed to carry many
conversations at a time.

Figure 20.
A trunk line
\f

T_4.2 Simplex, Half-Duplex and Duplex Channels or lines.

Channels can be classified into different types dependent upon
the direction in which they transmit and speeds of transmission.
These are called simplex, half duplex and duplex channels.

Like water flowing through a pipe, it cannot flow in both direc-
tions at the same time. Simplex channels transmit data in only
one direction and can be used when information need flow only
from a terminal to a computer, or vice-versa, for example.
However, simplex channels are seldom used in data communications
because we usually want to transmit both ways to and from a
terminal or computer.

Half duplex (HDX) channels are like pipes which can have the
direction of water flow changed from one direction to the other.
They can transmit data in both directions but only in one direc-
tion at a time, i.e. the data cannot flow in both directions
simultaneously. Depending on the users requirements HDX channels
can be the most economical choice for data transmission where
volumes of data are not required to transmit at high speed. The
slow-down in transmission is caused by the time it takes to
change the flow from sending to receiving.

Duplex (or Full Duplex - FDX) channels are like having 2 pipes
that can flow water, one in each direction. They can transmit
data in both directions simultaneously. Checking the data can be
overlapped with sending the data. The flow rate of transmission
is significantly higher.
\f

T_
T_e_r_m_i_n_a_l_ C_o_m_p_u_t_e_r_

Figure 21.
Transmission types

Simplex- one way flow water pipe.
Transmitted data only

Figure 22.
Direction of transmission

Half Duplex -
a pipe which can change to flow water in either direction.
Transmitted data is received
Check character is compared
Data is accepted and acknowledged.
Transmission is in one
direction at a time.

Full Duplex -
two water pipes flowing water, one in each direction.

Data is transmitted; received,
checked, accepted and acknowledged
at both terminal and computer.
The data can flow in one direction
while checking and acknowledgement
&_ are overlapped in the other direction.
\f

T_ 4.3 Categories and Transmission Speeds of Channels or Lines.

The categorization of channels is related to the speed of trans-
mission. The speeds are listed in terms of the number of data
&_ bits per second that may be sent over the channel.

A bit, which is described in 2.2 of this manual, is a binary
digit; a character is 5, 6, 7 or 8 bits (depending on the trans-
mission code used); a word is normally 6 characters as used in
telegraph; and transmission speeds are mainly defined in terms of
bits per second (bps), characters per second (cps) and words per
minute (wpm)

Communication channels fall into one of three categories of speed:

1 "Sub-voice grade" are channels designed for telegraph and si-
milar machines transmitting at speeds ranging from 45 to 150 bits
per second.

2 "Voice grade" are telephone channels transmitting at speeds
from 600 to 9600 bits per second.

3 "Wideband" channels transmit at speeds ranging from about 18,000
to 500,000 bits per second.

&_ Figure 23.
Transmission speeds\f

F_ 5 CHARACTERS AND CODES

Because information is transferred, the telephone industry is in-
the business of designing, developing and providing the trans-
mission channels by which information in the form of data is mo-
ved. As in any mathematical analysis there should be a way to
measure the value of the commodity.

The bit has been introduced as a measure of information and bit
per second as the rate of information transmission. Another term,
baud, is sometimes confused with bit, or bit per second. The baud
named after Baudot, is a measure of signaling speed. The baud
rate is the number of signal elements per second transmitted over
the channel. Baud should be thought of as a measure of channel
capacity. When nothing is being sent, the information rate is
zero bits per second, whereas the baud rate of the channel is
unchanged.

As an introduction to a data transmission course, it is desirable
to have a general familiarity with the coding process. By coding
is meant how bits are arranged to represent alphabetic, numeric
and other characters.

The first significant code that is referred to is Baudot code.
Jean Maurice Emile Baudot of the France Postal Telegraph Service
set out in the 1880's to develop a scheme to multiplex a number
of telegraph signals onto one channel without crosstalk. Multi-
plex means to cause a line to carry more than one transmission
simultaneously.

The Baudot, the 5-level code, has 2 to the power of 5 or
2x2x2x2x2 possible combinations for a total of 32 characters.
This is sufficient for the 26 letters, certain operational
functions, and an escape character which allows the duplicate use
of the remaining codes until a return character (letter) is sent.
Thus, a total of 58 characters may be transmitted.

The 5-level code has served well for teletypewrite service and is
still much in use. However, the data processing industry has
introduced many other ways of representing information.

The primary reason for departure from the teletype code has been
to gain efficiency in the storage area of the computer.
\f

T_5.1 Information Codes

A) 5-level Teletype Code.

Even though it has limitations, the 5-level code is used as an
&_ input to some processing systems, primarily because of the
continued use of older teletypewriters. It is probable that
the code is retranslated at the central processor input.

T_ B) 6-Bit Alphameric Code (BCD) Binary Coded Decimal.

With 6 bits this code can have 2 to the power of 6 combina-
tions and is a code suited to punch cards. A seventh bit is
used to provide parity checking. (Parity is explained in the
&_ following chapter.)

T_ C) 8-Bit Alphameric Code (EBCDIC) Extended Binary Coded Decimal
Interchange Code.

With 8 bits, 256 characters and symbols are available. This
provides a powerful code for many applications. Parity
checking is done by the ninth bit. These 8 bit characters are
&_ commonly known as bytes.

T_ D) USASCII USA Standard Code for Information Interchange.
Establishedby the American National Standards Institute.

This is a 7-bit code developed by data processing industry
representatives as a standard for machine to machine and sys-
tem to system communication. It has a capacity of 128 symbols.
The code includes a parity check bit.

Summary

A number of other codes have been in use but the above are the
&_ most common ones.

For transmission purposes, all data can be reduced to a digital
format. This can then be encoded into binary language and sent
over the communication channel. Although there are many possible
ways of encoding the data, the final result is quite similar.

The communication channel must carry the message to the other end
of the line to the terminal or computer receiving it and deliver
the message in exactly the same form as it received it.
\f

T_ An example of how bits are coded to represent alpha and numeric
characters is shown below with samples taken from the EDCDIC code.

T_h_e_ _8_-_b_i_t_ _b_y_t_e_ _c_o_d_e_.

This code has 256 combinations and is widely used by computers.

A If no bits are on the value = 0 or zero
If 1 is on the value = 1
B If 2 is on the value = 2
If 1 and 2 are on the value = 3
C If 4 is on the value = 4
8-bit etc. to character 9.
charac- D
ter or If A,B and 1 are on the character= A
byte 8 If A,B and 2 are on the character= B
If A,B and 1 and 2 are on the
4 character = C

2 etc. to I

1If A,B,D and 1 are on the character= J
IF A,B,D and 2 are on the character= K
etc. to R

&_ Figure 24.
Combinations of bits form characters.

This code has 256 combinations and is widely used by computers.

Equally widely used by computers is an arrangement of bits into
words which are 2, 3 or 4 or more bytes long, that is, 16, 24 or
32 bits or longer words.\f

F_ 6 ERROR CHECKING, CORRECTION and DETECTING CODES.

Line errors occur infrequently. Even on an ordinary telephone
line the likelihood of error is only about one bit in 50,000. If
the message transmitted takes the form of, say, an English-
language text, then at the point of reception it will often be
clear (1) that an error has occurred, (2) what to do about the
error.

Figure 25.
Types of errors

If I receive a telegram that says

Yox are invited to a party at 1655 105th St.,
Philadelphia, Pa.

Then I know that there are two errors here. I know that "Yox"
should be "You", because I am familar with the language being
used. But I also know that there is no 105th Street in Philadel-
phia, so that I know the address is in error.

Here we have an example of two sorts of errors. The first sort of
error is recognizable by anyone who knows the English language,
but the second is an error of fact. Had the address referred to a
67th St., then I might not have recognized it as an error because
there is such a street in Philadelphia. In data communications
errors of fact may be unrecognizable, that is unrecoverable. Con-
textual errors, like "Yox", are more often recognizable, but it
may be impossible to correct them; however, contextual errors
are often recoverable, that is, the receiver can ask the trans-
mitter for a retransmission, or it may ask a computer attached to
it for a correction.

Errors resulting from the equipment attached to the line, such as
the modems, are much less frequent. In general one can say that\f

equipment and line error are so infrequent that the most cost-
advantageous way of treating an error is to ask for a repetition
of the transmission. If the error cannot be tolerated at all
then it may be decided to transmit each message two or more times,
whether or not an error has been noticed. In most cases the major
problem in error checking (often called error detection) is
determining if an error has occurred.

Once it has been determined that an error has occurred, there are
a number of things that can be done. We have noted the procedure
of asking for a repeat of the message. Sometimes one simply notes
the error and decides to ignore it. Later on we will see how the
structure of a message can be utilized to aid in error detection.

It is the possiblity of error that complicates data transmission.
If no error could ever occur, then in a great many cases the trans-
mission and reception equipment could be quite simple and inexpen-
sive. For remote printing, for example, we could simply have a
transmitter at one end and a receiver at the other end and commu-
nications would go in only one direction all the time. We would
also not need much in the way of intelligence in the system. Eve-
rything that was received would be printed, except for the con-
trol characters, which would tell us how to print it and on what
device. But since transmission errors can occur, it is necessary
to have at each end of the line equipment that can both send and
receive, as the print system has to have some way to tell the
transmitter that it should at least repeat an erroneous message.

When large masses of data are being transmitted very fast, one
often also needs a computer on each end of the line to encode the
message for transmission in such a way that it is possible to de-
tect an error when it has occurred.

T_6.1 Parity, Blocking, Buffering

Let us consider an 8-bit binary code. Every character represented
in it has either an even or an odd number of 1's. If an 8-bit cha
&_ racter, has an even number of 1's, then we say, that it has even
parity, if it has an odd number of 1's, then it has odd parity.
Note that some systems count the number of zero's, rather than
1's, to determine parity, but this is not important for our purpo-
ses. The point is to understand that whatever parity system is
used, that system must be used throughout the discussion.
\f

A very simple way to use parity checking to check for error would
be to use an 8-bit character and use one bit for parity checking.
If this is done, then the last bit would be set to 1 if the cha-
racter had odd parity, and to 0 if the character had even parity.
This would be done before transmission by a parity generator.
Then all characters received should have even parity. If the re-
ceiver and the computer to which it was attached noticed a cha-
racter with odd parity, then they would know that an error had
occurred. One problem with this simple system is that if two er-
rors occurred in the character during the transmission, the pari-
ty check would not find the error.

For this reason information is usually transmitted in blocks.
That is, the characters are grouped into blocks of, say, 256
bytes. Parity bytes can be added to each block so that we could,
in the simplest case, first receive the block and check that the
first bits of each character when taken all together had even
parity. That would require that before transmission the com-
puter attached to the transmitter had set up the parity bytes in
such a way that if all the first bits of each data character came
out together with odd parity, then the first bit of the parity
bytes would be set to 1, otherwise to 0 etc.

Parity checking can be done for bytes, blocks or parts of blocks,
and it can be variously arranged so that any desired degree of
error-free transmission can be attained, except for totally error-
free transmission. Naturally, the fewer errors that a system al-
lows, the more expensive the system must be. In the extreme case
some systems have the requirement that a message be automatical-
ly retransmitted hundreds of times to ensure accuracy but this
is found mostly on military systems.

Though error inside a computer is very infrequent, it can occur
as the machine transfers data around, within itself. Thus compu-
ters also use parity checking and the blocking of data, particu-
larly in communicating with their peripheral equipment. Histori-
cally, a place was provided in computers where errors could be
checked for during the transfer of data between the computer and
its peripherals. These places were called buffers. Today, when
computers are so many thousands of times faster than their pe-
ripherals, the buffers serve the second purpose of providing a
place where data can be collected into blocks for fast transfer
between the computer and its peripherals. In data transmission,
therefore, we often find two buffers in both of the machines that
are communicating with one another. One buffer in the sender will\f

hold the character or block that has just been sent, in case its
transmission has to be repeated, and the other buffer will hold
the next character or block to be sent. On the receiving end the-
re are also two buffers. One holds the last character or block
that has been sent, and the other holds the current character or
block to be sent, in case it must be retransmitted. This characte-
ristic of double buffering speeds up transmission considerably.
Character buffers are used for slow transmission, and block buf-
fers for fast transmission.

In data transmission it is often convenient to make the block
size equal the size of some buffer, but one must often be content
to have the buffer size be some multiple of the size of the
blocks being transmitted. Block length, therefore, is determined
by the total configuration of the system of which the data
transmission is a part.

A block is said to consist of one or more records. Record size
relates to the medium that is used for input or output. For
example, systems depending on punched cards may use a record size
of 80 characters, corresponding to the number of characters that
can be punched on card. Then the block size will usually be some
multiple of 80. For systems running on tape or disc the block
size will usually be larger; often the determining factor will be
the maximum buffer size of a computer in the total system. Record
length here will depend on other factors in the total system.

In sum, the number of records in a block and the length of each
record are determined by total system requirements. One can say
that today the concepts of blocks and records are less than high-
ly descriptive, but in general block size depends on the error
checking procedures used in a system, and record size depends on
the sort of peripherals used. At any rate the block size must
equal some multiple of the record size for normal systems.

The first step in error control is to find ways of detecting er-
rors. There are several methods used to detect errors. The most
commonly used are Vertical Redundancy Check (VRC), Longitudinal
Redundancy Check (LRC, and Cyclic Redundancy Check (CRC).\f

T_ Block of eight
x 8 bit bytes

Parity bits for each byte = Vertical redundancy checking

0 0 1 1 0 1 0 10

0 1 0 0 1 1 0 0 1

0 1 1 0 0 1 0 1 0

1 1 1 0 1 1 1 0 0
Vertical
8 bit1 1 0 0 0 0 0 1 1 Parity byte =
characterslongitudinal
or bytes 1 1 0 1 0 0 1 1 1 redundancy checking

0 0 0 0 1 1 1 0 1

1 1 0 1 1 1 0 1 0

0 0 1 1 0 0 1 1 0

Figure 26.
&_ T_w_o_-_w_a_y_ _p_a_r_i_t_y_ _c_h_e_c_k_i_n_g_(showing even parity)

VRC - parityof each character may be either odd or even, meaning
that the sum of the number of ones will always be odd or even de-
pending upon which code is used. To satisfy the odd (even) parity
requirement, a bit is simply inserted in the check position for
each of the characters to change the sum of the one bits to an
odd (even) number. If an odd (even) number of bits are either
lost or added the character will be in error.

LRC - each transmitting and receiving terminal generates a separa-
te count of one bits for each of the bit positions of the code.
Characters are grouped into blocks of data. Each block is ended
with an End-of-Block (EOB) character, the LRC character generated
at the transmitter is sent to the receiver. The transmitted LRC
is compared at the receiving station with the LRC generated by
the receiver. If they are equal the next block of information is
transmitted.If they are unequal, the terminal will retransmit the
block in an attempt to automatically correct the error.

CRC - divides all the serialized bits in a block by a predetermi-
ned binary number. The remainder of this division is the check\f

character which is sent and compared with the check character
obtained in similar fashion at the receiving terminal.
Echoplex - the receiving station returns the data to the trans-
mitting station where it is compared with the transmitted data.
If a comparison isn't made the block is retransmitted.

Once the errors have been detected, the question arises: What
should the system do about them? Unfortunately, most of our sy-
stems today must be linked together using lines that were not
designed with the transmission of computer data in mind. The error
rate is part of the price paid for using lines intended for some-
thing else. When communication lines are constructed especially
for data, much lower error rates will be achieved.

It is generally desirable that the system should take some auto-
matic action to correct the fault.

In a typical transmission system, a VRC and LRC checks parity. To
detect errors, redundancy is built into the transmitted messages.
At the end of each block, the receiving station sends a signal to
the transmitting station saying the block has been received cor-
rectly or whether an error has been detected. If any error is
found, the block is retransmitted. If transmission of the same
block is attempted several times and is still incorrect, the
equipment will stop and notify its operator by some means.

In order to govern the automatic retransmission of information in
which an error has been detected, a number of special (control)
characters are used. For example, the codes ACK, NAK, CAN and
DEL.

The ACK code is used by the receiver to acknowledge to the trans-
mitter that a block of data has been correctly received. The NAK
code is used by the receiving terminal to give a negative acknow-
ledgement to the transmitter that a block of data has been re-
ceived with an error. When the transmitter sends a block of data,
it usually waits for the return of an ACK or NAK before is sends
the next message. If ACK is received, it proceeds normally; if a
NAK is received, it resends the block in error.

If the transmitter itself detects an error in a message on which
transmission has already begun, it will send a CAN character to
cancel the message.

It is possible that the control characters themselves or end-of-\f

transmission characters could be invalidated by a noise error.
Noise is an electrical disturbance on the line. If this happens,
then there is a danger that a complete message might be lost or
two messages inadvertently joined together. To prevent these
errors, and odd-even count may be kept on the records transmit-
ted.

If an odd-numbered block of data does not follows an even-number-
ed block, then the block following the last correct block is re-
transmitted. The two different ACK signals may be sent as ACK 0
and ACK 1.

The simplest method is a scheme with positive acknowledgements
only. The receiver just ignores incorrect messages. The transmit-
ter sends a message and waits for acknowledgement of its correct
receipt. If no acknowledgement is received after a specified
time. It retransmits the message.

One method of detecting errors does not use a code at all. In-
stead, all the bits received are returned to the transmitter
where the bits are checked for accuracy. If an error is detected
the message is retransmitted.

T_ Summary

Although the procedures of error detection and checking may seem
&_ complex they are when brought down to basics really quite simple
rules. However, modern computers and data communications equip-
ment have to be error free so that we may rely on our systems to
carry out tasks in an accurate manner.\f

F_ 7 SYNCHRONOUS AND ASYNCHRONOUS TRANSMISSION

Data transmission can either be synchronous or asynchronous. With
synchronous transmission, characters are sent in a continuous
stream. When no data is being transmitted, the SYN character is
used as a fill character to keep the transmitter and receiver in
synchronization, i.e. transmitting and receiving at the same
speed.
T_

Figure 27.
&_ Asynchronous transmission

Start/Stop (asynchronous) transmission is usually used on keyboard
devices which do not have a buffer and on which the operator
sends characters along the line at random when she presses the
keys. Start and stop bits are necessary between each character.

Asynchronous machines are less expensive than synchronous machi-
nes. Many asynchronous transmissions are card-to-card, paper-
tape-to-printer, card-to-computer. Although the character stream
does not have the pauses between characters, a teletypewriter
transmission has.

When machines transmit to each other continuously, SYNCHRONOUS
transmission can give the most efficient line utilization.
The continuous stream of characters of this type is divided into
blocks. All the bits are transmitted at regular intervals.
\f

Figure 28.
Synchronous transmission

The synchronization of the transmitting and the receiving machi-
nes on many systems is controlled by sending a synchronization
pattern or character at the start of the block. If this were not
done, the receiving device would not be able to tell which bit
was the first in a character, which was the second and so on.

Synchronous transmission can give better protection from errors
because at the end of each block is an error-checking pattern.
This pattern is selected to give the maximum protection from noise
errors on the line.

In addition, faster transmission is ensured because no START or
STOP bits are needed between characters.

There are three basic types of synchronization used in data trans-
mission. They are BIT synchronization, CHARACTER synchronization
and MESSAGE synchronization.

BIT synchronization ensures that the receiving unit knows at what
instant a bit starts and ends.

CHARACTER synchronization ensures that the receiving unit knows
which bit is which in a character; first, second, etc.

MESSAGE synchronization ensures that the receiving unit knows
which characters are starting and ending characters of messages;\f

i.e., the START-OF-HEAD (SOH), START-OF-TEXT (STX), END-OF-TEXT
(ETX), END-OF-BLOCK (EOB) and END-OF-TRANSMISSION (EOT).

T_7.1 Line Control or Data Link Control

There are many aspects concerning line control and it is a sub-
ject concerning much of the programming for communications ope-
&_ rations. It is also an expanding subject. It has to continually
meet new requirements and procedures such as code structures,
synchronization of terminals and computers and controlling them.

When transmission devices send data to each other, a variety of
control signals must pass between them to ensure that they are in
synchronization with each other. The sending machine must tell
the receiving machine when it is about to start transmitting. The
receiving machine must tell the sending machine whether it is
ready to receive. When the receiving machine detects errors, it
must notify the sending machine, and the erroneous data must be
retransmitted.

T_ The format of the basic address message is a follows:

1. Control character
2. Address of control unit
3. Address of device
&_ 4. Command to be executed.

There are two types of line control which have been widely used.

T_ These are:

Start/Stop line control
Data link control

The start/stop line control is an earlier technique used for syn-
chronized transmission. It was developed according to a multitude
of designs by various manufucturers and for different types of
communications equipment. There was little compatibility between
types of terminals and therefore numerous types of adaptors were
required to overcome these difficulties. It was necessary to sim-
plify line control with a new discipline (also known as a line
protocol) and this is called Data Link control.
\f

Data Link control is a set of operating procedures and control
characters which is used as a common language by all binary syn-
chronous communicating (BSC) equipment on all types of lines. The
data is transmitted at a constant and synchronous rate. All bits
of a character and all characters are sent at a uniform speed.

The procedures of data link control include programming for mana-
ging transmission codes, synchronization, initialization of trans-
mission, error detection, message blocking, acknowledgments and
ending transmission.

Newer techniques include IBM's Synchronous Data Link control
(SDLC), ISO's (International Standards Organisation) HDLC, (for
High Level Data Link Control).\f

F_ 8 MODEMS

When transmission is carried out through a MODEM, the signal on
the line is composed of frequencies. A MODEM is composed of two
parts: a M_O_dulator and a D_E_M_odulator. The modulator (transmitter)
is coupled with a demodulator (receiver) located at the other end
of the line. A source device at the sending station sends a digi-
tal signal to the modem where it is converted to frequencies re-
presenting the "1"'s and "0"'s.

The "1" or mark is at a lower frequency than the "0" or space. At
the receiving station at the other end of the line the frequencies
are reconverted to a digital representation again by the demodu-
lator.

A typical communications system would consist of an input/output
station, modulator, transmission link, demodulator and a compu-
ter. The input can be anything from a teletypewriter to a compu-
ter. The modulator and demodulator (modem) interface the input/
output equipment with the transmission link. The modulator's
function is to make input signals suitable for transmission. The
demodulator converts the transmitted signals to their original
state before sending them on.

With the ability of the customer to buy and install his own
choice of modems on certain lines, low speed data is more econo-
mical. This choice of equipment allows a more efficient use of a
leased line. The user can now buy or rent equipment which drives
a number of channels from a single, leased voice-frequency circuit

Speed of transmission is one system characteristic which influen-
ces the number of channels which can be derived from a single
circuit. Direct computer-to-computer operations run at extremely
high rates of transmission.

Lower transmission speeds can use narrower bandwidths and less
ideal circuit conditions. Modems at rates of up to 9600 bits per
second have been carried on equalized voice circuits. High speed
exchange requires wideband transmission links which minimize
noise and distortion. Such equalization raises the cost of ser-
vices.\f

F_ 9 TERMINALS

A terminal is basically an input station or an output station but
usually it is both. It may be connected to a host computer system
as well as to other terminals on one communication line or through
a network of lines. There may be more than one host computer to
which the terminal is connected.

A basic terminal for input only could be a device such as a key-
board or a card reader. An output terminal is often a type of
printer though it may also be a display terminal like those used
in airline reservation offices.

Figure 29
Airline reservation terminal

However, more commonly a terminal is a combination of an alphanu-
meric display (or CRT i.e. a cathode ray tube similar to a TV
screen also called a VDU for Visual Display Unit), a keyboard and
a printer.

T_\f

Figure 30.
RC 822 Alphanumeric Display/keyboard

Figure 31.
RC 866 Terminal Matrix Printer \f

F_ A terminal is configured to handle the volume of data and the ty-
pes of operations required by the user at its site.A variety of
models of each type of equipment have been developed to accommo-
date the diversity of user requirements. RC offer a range of mo-
dels of alphanumeric display/keyboards, printers, card readers,
multiplexers and concentrators and other equipment.

Multiplexers are devices that can send several messages along one
line simultaneously.

Concentrators are devices which can receive data from a number of
input devices and arrange it into logical blocks for easy
transmission.

Both multiplexers and concentrators are often used to obtaingrea-
ter efficiencies in the use of transmission lines.

The equipment for a terminal is then selected by the user and the
RC company to fit the job accordingly and with sufficient capaci-
ty to cater for a growth in current volumes of activity.

Further capacity to handle increased future workloads is taken ca-
re of by the modularity of design in all RC data processing equip-
ment. This means that capacities and speeds can be increased in a
configuration (i.e. a group of devices making up a computer sys-
tem or a terminal) by adding units of, say, alphanumeric display/
keyboards, increasing the internal program storage of a concen-
trator or adding a faster printer or printers, etc.

Similarly, types of transmission lines are chosen according to
the job requirements of the terminal.

Moreover, the units in a terminal installation need not be limi-
ted to a single location at a user's premises since equipment can
be placed where it is most needed. e.g. data entry for orders for
goods to be shipped can be entered on a terminal in the sales
office and delivery notes and packing slips can be printed on a
terminal in the warehouse. \f

A few examples of common terminal configurations are described
below.

Figure 32 shows a simple terminal configuration employing a sing-
le full-duplex line.

Alphanumeric
Display/keybord

MODEM

300-1200 bps
full-duplex

Figure 32.
Single line terminal
&_
Figure 33 shows a terminal with 2 lines which will handle a
greater flow of data. Output from the computer to the printer can
occur simultaneously with typing in at the display/keyboard.

T_ Alphanumeric
Display/keyboard
MODEM

300-1200 bps
full duplex

Printer
MODEM
1200 bps half duplex
Figure 33.
&_ Two line terminal

Figure 33 differs from figure 34 only in that the printer is
shared by several keybords thereby utilizing the printer's
capacity more fully.\f

In figure 34, output from the computer to the printer and typing
in at the keyboards can occur simultaneously.

T_ Alphanumeric
Display/keyboards

MODEM

300 bps
Full duplex

PrinterMODEM

1200 bps
half/full duplex

&_
Figure 34.
Multiple keyboard terminal

The configuration of figure 34 is similar to that of figure 35
but transmission costs are considerably reduced by the introduc-
tion of a concentrator.
T_
Alphanumeric
Display/keyboards

CONCENTRATORMODEM

1200/4800 bps
half duplex

Printer

&_
Figure 35.
Multiple keyboard terminal with concentrator

A user may need only a terminal and not require the capacity of a
computer as well. In which case he can avail himself of the servi
ces of a service bureau by a transmission link. Chapter 10 deals
with this topic and RC NET.

T_9.1 Serial and Parallel Transmission

Most terminals are connected together through lines in such a way
&_ that each bit is transmitted along the same path as any other
bit. In this case each bit has its place in a stream of bits.
This is called serial trasmission. Other terminals are so connec-
ted that there is a separate path for each bit in a byte. Thus
the whole byte can be transmitted at once. This is parallel trans-
mission. Most of the time data communications utilizes serial
transmission, while the connection between a terminal and its
associated equipment is made via parallel transmission. Such
associated equipment might be, for example, a card reader, line
printer, or computer. Thus communications over a few feet are
usually parallel, and communications over long distances are
usually serial.

Parallel transmission does not necessarily require more than one
pair of lines. Multiplexing can be used to allow parallel trans-
mission of an entire byte along a single line. Whether or not
this is done depends on the sort of modem used, and of course
parallel transmission that is performed in this way generally
requires rather more expensive modems.

When computers are used in data communications systems as termi-
nals, they most always have concentrators between them and the
transmission line. There are many kinds of concentrators. But
when many terminals are connected by telephone with a central
computer the equipment the lines go into is most often called a
concentrator. The purpose of a concentrator is to arrange
incoming data in such a way that is can be fed into the computer
at a rate compatible with the computer's internal speed. The
concentrator, then, allows the computer to work at its own speed
continuosly, without having to wait for data coming in at lower
speeds.

T_ A terminal that either only receives or only sends is said to be
simplex, and a terminal operating this way is said to work in
simplex mode. Such terminals obviously need be connected by only
two wires. Thus, they are two-wire terminals. But not all two-
wire terminals are simplex.

&_Two-wire terminals are capable of being designed so as to allow
transmission to proceed in both directions. Such terminals are
said to be half-duplex. These terminals cannot transmit in both
directions simultaneously. Each time the direction of transmis-
sion changes, they must wait for some milleseconds to allow the
transmission line to turn aorund. \f

Four-wire terminals are capable of operating in full-duplex mode,
which means that they can receive and transmit simultaneously.

The terms used above are sometimes subject to misinterpretation.
For example, in some countries there is another definition of
"simplex". Finally, the number of wires no longer is strictly
attached to the terminals operating mode. It is important to be
aware that the person you are talking with may not use the terms
as defined above, but the above definitions are becoming stan-
dard. At any rate the terms "two-wire" and "four-wire" are less
useful than "simplex", "half-duplex", and "full-duplex".

T_
9.2 Transparent, Transparency

Transparancy is a word that is used in at least two different
&_ ways. It is also a word that is becoming very popular, so that it
is certain to take on additional meanings in the future. In gene-
ral it is best to avoid using the word, as few people are quite
sure of what it means.

The first meaning is when a person is describing a total data
transmission system, particularly including the part that is
operated by the local telephone company, one can often hear
expressions such as, "The system is transparent to the structure
of the telephone network". For example, in many countries where
population centers are far apart and separated by rough terrain,
the local telephone company may use microwave links in certain
parts of its telephone systems. But this fact is supposed not to
be noticeable by the rest of the telephone system, for it is
designed to perform exactly as though telephone lines had been
used throughout.

A second meaning of the word refers to data that is received by a
terminal as pure strings of binary data. When in the transparent
(or transparency) mode, the terminal will make no attempt to in-
terpret this data. It will simply take it in as is. This is use-
ful for transmitting programs written as binary data, for compu-
ters that communicate directly with each other, and for communi-
cating with IBM terminals or with terminals that are simulating
IBM terminals, i.e. another make of terminal that is programmed
to operate in a similar manner to an IBM terminal.

The general connotation of "transparent", then, is that someting
is allowed to pass through unchanged as glass is said to be trans-
parent to light.\f

T_ 9.3 Terminal Simulation

Most terminals and computers are so designed as to work best in
connection with other terminals or computers designed by the same
&_ manufacturer. On the other hand, the manufacturer's own terminal
or computer may not be a proper choice for the system user. Thus,
there are many terminals available which are able to behave in
such a way that they can be made to look like the terminal or com-
puter that the system requires. This is done in several ways.
First, a manufacturer may construct his terminals or computers in
such a way as to be plug-compatible with a certain machine or
family of machines. In theory this means that you can literally
plug the machine into the system with which it is said to be
plug-compatible. In practice this doesn't usually work out quite
as advertised.

A second method is to copy the original manufacturer's specifica-
tions so closely as to make you own hardware virtually indistin-
guishable from that of the other manufacturer. Needless to say,
the other manufacturer, usually IBM, has taken certain precautions
against this practice, so that here too, the replacement equip-
ment does not always work as promised.

The best way to allow oneself freedom in the selection of termi-
nals is through terminal simulation. In this case the terminal
must have a small computer in it, allowing it to be programmed to
make it behave exactly like the terminal to be replaced, or bet-
ter. There are three advantages to using a terminal of one's own
choice with simulation capabilities. First, the terminal of choi-
ce may have capabilities that the other terminal did not have and/
or it may be less expensive. Secondly, the terminal of choice may
have to communicate with equipment manufactured by different
companies so that with a simple change of program it can simulate
more than one terminal without needing special hardware. Finally,
such a terminal can be used in large communications systems where
at various times various pieces of equipment have various
functions.
For example, one might have a number of small inexpensive termi-
nals sending data into a Datapoint terminal for initial collec-
tion and rough editing. There might then be many such Datapoint
systems feeding into one or more RC 3600 minicomputers where the
incoming data could be cleaned, edited, converted to various
output media, and routed to one or more computers. At various\f

times the Datapoint systems might be operating as stand-alone
terminals. Simiarly, there might be times when the RC 3600 would
be performing off-line functions. The RC 3600 might, for example,
have the capability of communicating with an IBM 370 computer and
a CDC 6600 computer, and might have to send different data to
each. Through terminal simulation programs the RC 3600's would
have the capacity to perform all these functions at a reasonable
price. If at each stage one had to use only terminals supplied by
the main central computer manufacturer, then one would need a
variety of terminals, many of which would have to be very
expensive and elaborate.

In terminal simulation a computer program is used to make one
machine look like a different machine to the other machines that
are contained in other parts of the overall system.

CDCICL IBM
T_ computercomputercomputer

CDC 200 UT ICL 7020 IBM HASPIBM
computer

IBM RES

RC 3600IBM 2780IBM
computer

IBM 3780IBM
computer
Univac NTRRC 3600
Siemens 840
&_
UnivacRC 8000Siemens
computercomputer computer

Figure 36.
RC 3600 Minicomputersimulates other makes of terminals\f

tegning indsættes

Figure 37
Terminal simulation

F_ 10 NETWORK CONCEPTS

Regnecentralen's product lines consist of equipment modules that
can be assembled into special-purpose installations or can be
used as nodes in data nets. A data net is a network of data trans-
mission lines linking terminals and computers in a data communi-
cations system. A node is an end point of any branch of a network,
or a junction common to two or more branches of a network.

One can begin with the RC 3600 product line. An RC 3600 finds its
main off-line uses in the areas of data entry or data collection
and data conversion. An RC 3600 consists of a single or dual cen-
tral processor and an array of peripheral devices. Off-line ope-
rations are performed when the equipment is not connected to the
main computer, e.g. printing reports from a tape that was prepa-
red by the computer. Data entry is the keying of source data to
record transactions etc. for the computer. Data collection is
gathering data from various points of entry. Data conversion is
converting e.g. punched cards data to magnetic tape.

Figure 38.
RC 3600 Data Entry System

There is no basic difference between a data conversion function\f

and a data entry function. Data entry is viewed as data conversion
with a number of simple data entry terminals as the major input
devices and a disc as the major output device. An editing process
intervenes between input and output for error correction and sor-
ting of the data. The data entry devices may be local or remote.

A data entry or data conversion system can, moreover, function as
a node in a data network, because the RC 3600 can also act as a
terminal

T_
RC 3600Mainframe

Mainframe

Mainframe
RC 3600 Terminal

Figure 39.
RC 3600 terminal

A medium or large-scale central computer is often referred to as
a 'mainframe' computer. Smaller computers that are connected to
it and operate to support the mainframe computer and relieve it
of input/output from peripheral devices and transmission of data
are called 'front-end' computers.
Further, an RC 3600 can act as a front-end to any mainframe to
help it deal with communications coming in from any other
terminal. \f

T_ RC 3600 Mainframe

RC 3600

Terminal

Figure 40.
RC 3600 front-end computer

RC 3600 can control communications equipment other than terminals.
For example, it can control telex networks. And because the RC
3600 can act as a data conversion system in general, it can also
act as a general peripheral controller for any mainframe.

RC also produces a mainframe, the RC 8000 computer. And RC
produces a specialized communications controller, the RC 3500
which is excellent, for example, at supervising telephone
networks.

For off-line functions that require minicomputer capacity,
Regnecentralen produces the RC 6000, which is an RC 3600 plus
additional minicomputer capability.

These four product lines give Regnecentralen the capability of
providing any sort of node in a data network, from a simple
keyboard data entry terminal to the whole network itself.

T_10.1 Distributed Processing

Computers were extended to communications when it was realized
&_ that the five basic device/functions did not have to be placed in\f

the same physical location in order to work together. The first
function to be decentralized was the input function. The decentra-
lization of the output function followed soon after. The decentra-
lization of the console function was realized only for computers
operating in environments where it was not safe or convenient for
human operators to be placed. In the last few years, however, a
very revolutionary development has taken place, and that is the
decentralization of the mainframe function. That is, the tasks of
the mainframe were separated from one another and were then cap-
able of being placed in different devices in locations remote
from one another. This is called d_i_s_t_r_i_b_u_t_e_d_ _p_r_o_c_e_s_s_i_n_g_.

The first mainframe function to be decentralized was the memory
function. It became possible to use fast discs in such a way that
they appeared to the user to be no different from central proces-
sor memory. Once this function could be decentralized, then it
could be replaced in almost any location.

The second mainframe function to be decentralized was the control
of peripheral equipment. This function was decentralized out to
special peripheral controllers.

Finally, even the computational function of the mainframe was de-
centralized, first by separating out the various types of compu-
tational tasks the mainframe had to perform and shifting them
around among the various processors of which the "mainframe" (now
a logical, rather than a physical, entity) was considered to
consist. It was soon realized that these devices, too, could be
located "anywhere".

This ultimate distribution of the data processing situation gave
rise to more intelligent terminals and peripherals, so that the
final result of the historical process was to blur the formerly
clear lines between "mainframe", "peripheral", "terminal", "com-
munications controller", etc. A truly distributed data processing
network is said to consist of "nodes" and paths between nodes.
The nodes can be of various sorts, and some nodes can have dif-
ferent functions at different times. A node can be as small as a
simple data entry device or as complicated as a whole computer
installation with hundreds of terminals. A node can also be used
outside its network for "off-line" functions which can range from
ordinary typing tasks for a typewriter detached from its modem to
independent operation for a large computer installation.

T_10.2 Packet Switching

The concept of distributed processing has given most spectacular
&_ results in the rise of the whole new industry of data entry and
in the "packet switching" networks typified by international air-
line reservation systems, and now being extended to banking net-
works.

In packet switching each message is parcelled into a "packet"
consisting of some data and a note telling where it is supposed
to go. This packet is sent from node to node around the network,
with its optimum path to its destination being recalculated from
time to time as traffic conditions in the network change, so that
a packet can travel many thousands of miles in just a fraction of
a second to its destination, where it is combined with the other
packets in its message and output.

Data entry system allows input to be collected and edited in
various ways and then communicated to a central site for proces-
sing. The most popular data entry systems use video devices ( dis-
play keyboards) for entering data and a central terminal for re-
gulating the data entry process and performing the communications
tasks.

The success of packet switching and data entry systems have given
the impetus for developing a variety of so-called "datanet"
systems.

T_
10.3 RC Net and Network Control Program

The introduction of large central computers made it possible to
&_ perform EDP tasks for several independent users simultaneously.
For remotely placed users terminal communication facilities were
developed. A number of different communication protocols were the
result but they all had in common that the central computer alone
was monitoring the system. This hierachial structure has several
drawbacks:

The central computeris heavily loaded by the numerous communica-
tion methods, the transmission lines are poorly utilized, and the
users may need several types of terminals for different central
computers. \f

The way to solve these problems and at the same time gain a num-
ber of advantageous facilities for your current systems with
built in security for future extensions is to adopt the RC NET
data network system.

Figure 41.
&_ RC Net data network

Within the RC NET network concept an installation with data pro-
cessing capacity is termed a "Host". Associated to each Host is a
"Node" and the Nodes are interconnected by communication lines to
form a network.

RC NET is a packet switching network system by which the Host
computers can communicate. A message from one Host to another is
split up into a number of data packets, which are transmitted by
the network. The main function of the Nodes is to keep track of
the Hosts currently connected and their location in the network.
The data packets are forwarded from Node to Node until the Node
connected to the receiving Host is reached.

High Level Data Link Control protocol is used on the communica-
tion lines which connects the Nodes, but other protocols can be
adopted for certain lines if it is convenient.

A Host in RC NET could for instance be an RC 8000 System, an IBM
system or an RC 3600 Terminal Concentrator. In the RC 8000 System
the General Device Controller is inherently implemented as a Node
by means of the Network Control Program (NCP). This program
resides partly in the Internal Store and partly in the Device
Controller. When an IBM computer system or a terminal system is
connected to the RC NET the RC 3600 Minicomputer is used as
network interface.\f

F_ 11 APPLICATION SOFTWARE

The RC company offers its customer a very wide choice of systems
in its application software. There are programs available to
handle the problems encountered in data processing for manufac-
turing industries, financial and banking organizations, govern-
ment and educational institutions, public utilities, transporta-
tion and commercial enterprises.
T_
RC has had many years experience in operating service bureaus.
Many customers take advantage of these services by sending their
data to the bureau for processing by a line from their terminal,
by mail or courier service.

Regardless of the diversity or size of the customer's system
requirements in hardware, software or services the RC company has
the system solutions.

GLOSSARY

ACK 0, ACK 1
(AFFIRMATIVE ACKNOWLEDGMENT)
These replies (sequences in Binary Synchronous Communi-
cations) indicate that the previous transmission block
was accepted by the receiver and that it is ready to
accept the next block of the transmission. Use of ACK 0
and ACK 1 alternately provides sequential checking
control for a series of replies. ACK 0 is also an
affirmative (ready to receive) reply to a terminal in
point-to-point operation.

ALGOL Algol is an acronym for A L G Orithmic Language. It is
a problem oriented high level programming language for
mathematical and scientific use, in which the source
program provides a means of defining algorithms as a
series of statements and declarations having a general
resemblance to algebraic formulae and English senten-
ces.

ALGORITHM A series instructions or procedural steps for the solu-
tion of a specific problem.

ANALOG COMPUTER
A computer that operates with numbers represented by
directly measurable quantities such as voltages or ro-
tations.

ANALOG SIGNALS
The signals upon which an analog computer operates.

ATTENUATION
To lessen the amount or force of - see attenuator.

ATTENUATOR
A device for attenuating especially for reducing the
amplitude of an electrical signal without appreciable
distortion.

ASCII American Standard Code for Information Interchange.
This is a seven-bit-plus-parity code established by the
American National Standards Institute (formerly Ameri-
can Standards Association) to achieve compatibility\f

between data services. Also called USASCII.

ASYNCHRONOUS TRANSMISSION
Transmission in which time intervals between transmit-
ted characters may be of unequal length. Transmission
is controlled by start and stop elements at the be-
ginning and end of each character. Also called Start-
Stop transmission.

AMPLITUDE MODULATION (AM)
A methode of transmission whereby the amplitude of the
carrier wave is modified in accordance with the ampli-
tude of the signal wave.

AUDIO FREQUENCIES
Frequencies which can be heard by the human ear
(usually between 15 cycles and 20,000 cycles per
second).

AUTOMATIC CALLING UNIT (ACU)
A dialing device supplied by the communications common
carrier. This device permits a business machine to au-
tomatically dial calls over the communications network.

BACK END PROCESSOR
A computer handling a data base to assist the main com-
puter.
BACKGROUND PROCESSOR
The automatic execution of a low-priority computer pro-
gram when higher priority programs are not using the
system reesources.

BACKING STORE
A store of larger capacity but slower access time than
the main memory or immediate access store of a compu-
ter. Also known as bulk store, auxiliary store, secon-
dary store.

BANDWIDTH The range of frequencies assigned to a channel system.
The difference expressed in Hertz between the highest
and lowest frequencies of a band.

BASEBAND SIGNALLING
Transmission of a signal at its original frequencies,
i.e., unmodulated. \f

BASIC RC Basic is a structured educational programming
language, implemented by A/S Regnecentralen to run on
RC 3600 and RC 7000 computers. RC Basic is extended
with COMAL (comman Algorithmic Language) which provides
advanced control facilities and structures.

BATCH PROCESSING
A method of processing data in which transactions are
collected and prepared for input to the computer for
processing as a single unit. There may be some delay
between the occurence of original events and the even-
tual processing of the transactions. Contrasted with
real time processing in which transactions are dealt
with as they arise and are automatically applied to fi-
les held in a direct access storage device.

BATCH PROCESSING MODE
In real time systems there is usually some aspect of
the data processing work that does not require to be
handled on a real time basis. Transactions falling in
this category may be batched on a daily/weekly/monthly
basis and be accepted into the system for processing
against sequential files. Thus, at certain periods of
relatively low activity, a real time system may operate
in batch processing mode; alternatively batch process-
ing jobs may form background processing work to real
time operations in a multiprogramming environment.

BAUD A unit of signalling speed equal to the number of dis-
crete conditions or signal events per second. In asyn-
chronous transmission, the unit of signalling speed
corresponding to one unit interval per second; that is,
if the duration of the unit interval is 20 millesec-
onds, the signalling speed is 50 Baud. Baud is the same
as "bits per second" only if each signal event repre-
sents exactly one bit.

BAUDOT CODE
A code for the transmission of data in which five bits
represent one character. It is named for Emile Baudot, a
pioneer in printing telegraphy. The name is usually
applied to the code used in many teleprinter systems,
which was first used by Murray, a contemporay of
Baudot.
\f

BINARY DIGIT (BIT)
In the binary notation either of the characters 0 or 1.
"Bit" is the commonly used abbreviation for Binary Di-
git.

BINARY SYNCHRONOUS COMMUNICATIONS(BSC)
A uniform discipline, using a defined set of control
characters and control character sequences, for syn-
chronized transmission of binary coded data between
stations in a data communications system. Also called
BISYNC.

BIT Abbreviation for BINARY DIGIT.

BIT TRANSFER RATE
The number of bits transferred per unit time, usually
expressed in Bits Per Second (BPS).

BLOCK A group of digits transmitted as a unit, over which a
coding procedure is usually applied for synchronization
or error control purposes. See also: Packet.

BUFFER Generally used as a means of temporarily storing data
when information is being transmitted from one unit to
another; e.g. between a central processor and its in-
put/output peripheral units. The purpose of a buffer is
to compensate for the different speeds at which the
units can handle data. Sometimes a buffer may be per-
manent feature of a peripheral unit (e.g. as in a buf-
fered printer) and in other systems internal memory
areas may be assigned temporarily to act as buffers for
particular units.

CCITT Comit Consultatif Internationale de Telegraphie et Te-
lephonie. An international consultative committee that
sets international communications usage standards.

CDC Control Data Corporation.A U.S. based manufacturer of
computer equipment. The trade name on its products.

CHANNEL 1. A path along which information flows. When all ele-
ments of a digit are sent in parellel, a channel is
made up of parallel paths. 2. A paper tape channel is a
longitudinal row in which code holes may be punched in
paper tape. 3. The part of a store accessible to a\f

reading station.

CHIP Intergrated circuits on a single silicone chip that is
usually only a few millimeters square in size.

COMMON CARRIER
In data communications, a public utility company re-
cognized as having a vested interest and responsibility
in furnishing communications services to the general
public. The public telephone company.

COMPATIBILITY
Compatibility is a term applied to both hardware and
software systems to describe the ease with which a com-
puter program running on one machine may be made to run
on another machine. Hardware compatibility is achieved
through similarity of instruction sets (or the ability
to simulate similarity of instruction sets), whereas
software compatibility deals with the use of a language
that can be translated into the (perhaps very different)
instruction sets of several machines.

CONCENTRATOR
A communications device that provides communications
capability between many low speed, usually asynchronous
channels and one or more high speed, usually synchro-
nous channels. Usually different speeds, codes, and
protocols can be accommodated on the low-speed side.
The low-speed channels usually operate in contention
requiring buffering. The concentrator may have the ca-
pability to be polled by a computer, and may in turn
poll terminals.

CONFIGURATION
The general term given to a computer system, usually
used to indicate the physical units of the system.

CONVERSATIONAL
Pertaining to a mode of processing that involves step-
by-step interaction between the user at a terminal by
means of keyboard and display and a computer. See also:
Interactive.

CORE MEMORY
A computer memory or store composed of magnetic cores.

CORE STORAGE
A type of memory composed of magnetic cores, in which
data is held in binary form by means of the property of
cores of retaining a positive or negative charge. The
pattern of charges serves to represent the coded data.

CROSS TALK
The unwanted transfer of energy from one circuit, cal-
led the disturbing circuit, to another circuit, called
the disturbed circuit.

CYCLIC REDUNDANCY CHECK (CRC)
An error detection scheme in which the check character
is generated by taking the remainder after dividing all
the serialized bits in a block of data by a predeter-
mined binary number.

DATA A general expression used to described any group of num-
bers, alphabetic characters or symbols which denote any
conditions, value or state, e.g. all values and descrip-
tive data operated on by a computer program but not the
program itself. The word data is used as a collective
noun and is usually accompanied by a single verb: 'data
are' may be pedantically correct but is arkward to say
and therefore arkward to understand. Data is sometimes
contrasted with information, which is said to result
from the processing of data, so that information de-
rives from the assembly, analysis or summarizing of
data into a meaningful form.

DATA BASE 1. The entire collection of information available to a
computer system. 2. A structured collection of infor-
mation as an entity or collection of related files
treated as an entity.

DATA CONCENTRATION
Collection of data at an intermediate point from se-
veral low and medium-speed lines for retransmission
across high-speed lines.

DATA COLLECTION
The act of bringing data from one or more points to a
central point.
\f

DATA COMMUNICATION
The interchange of data messages from one point to
another over communications channels. See also: Data
Transmission.

DATA COMMUNICATION EQUIPMENT (DCE)
The equipment that provides the functions required to
establish maintain, and terminate a connection, the
signal conversion, and coding required for communica-
tion between data terminal equipment and data circuit.
The data communication equipment may or may not be an
integral part of a computer; e.g., a modem. See also:
Terminal Installation, Data Link.

DATA COMMUNICATIONS SYSTEM
A data processing system with data transmission capa-
bilities.

DATA LINK An assembly of terminal installations and the inter-
connecting circuits operating according to a particular
method that permits information to be exchanged between
terminal installations.
NOTE: The method of operations is defined by particular
transmission codes, transmission modes, direction, and
control.

DATA NET A data communication exchange for controlling the
transfer of messages to and from remote terminals and a
central computer.

DATA PROCESSING
The operations performed on data usually by automatic
equipment, in order to derive information or to achieve
order among files. A data processing system may incor-
porate clerical functions and ancillary machine opera-
tions as well as all arithmetic and logical operations
performed by a computer.

DATAPOINT A U.S. based corporation and manufacture of computer
equipment. The trade name on its products.

DATA SET A device which connects a data processing machine to a
telephone or telegraph communication line. For example,
a telephone data set converts digital signals to tones
for transmission over a speech quality circuit.
\f

DATA TRANSMISSION
Pertaining to the automatic transfer of data from one
computer system to another, or to and from a central
computer and distant data collection points. The data
may be transferred by special equipment using either te-
legraph or telephone circuits, or by radio link. The
speed of transmission is largely governed by the charac-
teristics of the data transmission line or channels.
The sending of data from one place for reception else-
where. Compare: Data Communication.

DECENTRALIZED (COMPUTER) NETWORK
A computer network, where some of the network control
functions are distrubted over several network nodes.

DEMODULATION
The process of retrieving an original signal from a mo-
dulation carrier wave. This technique is used in data
sets to make communication signals compatible with com-
puter signals.

DIAL-UP LINE
A communications circuit that is established by a swit-
ched circuit connection.

DIRECT DISTANCE DIALING (DDD)
A telephone exchange service which enables a user to
directly dial telephones outside his local area without
operator assistance.

DIRECT MEMORY ACCESS (DMA)
A facility that permits I/O transfer directly into or
out of memory without passing through the processor's
general registers; either performed independently of
the processor or on a cycle-stealing basis.

DISC DRIVE OR DISC STORAGE MODULE
A unit which houses the spindle on which a disc pack is
mounted.

DISC PACKS
Magnetic disc storage units. Usually exchangeable on
the disc storage drive or module on which they are
mounted.
\f

DISTRIBUTED PROCESSING
A network in which, usually smaller, processing is di-
stributed among a number of processors at various nodes.
They can draw on files in the database of the central
mainframe computer. Distributed processing can be com-
pared with centralized processing where all the proces-
sing for a network is carried out by the mainframe com-
puter.

DUPLEX Simultaneous two-way independent transmission in both
directions. Also referred to as full-duplex.

EBCDIC Extended Binary Coded-Decimal Interchange Code.
(Pronounced EB-SID-DICK). An 8- bit character code used
pri marily in IBM equipment. The code provides for 256
different bit patterns.

EDIT To arrange data into the format required for subsequent
processing. Editing may involve deletion of data not
required, conversion of fields to a machine format (e.g.,
value fields converted to binary) and preparation of
data for subsequent output, e.g., zero-suppression.

EDP Acronym for electronic data processing.

EOT (END of TRANSMISSION)
Indicates the end of a transmission, which may include
one or more messages, and resets all terminals on the
line control mode (unless it erroneously occurs within
a transmission block).

E.O.F. ACRONYM FOR END-OF-FILE
Error checking code. A general term for all error detec-
tion codes and error correction codes.

ERROR CODE
The identification of a particular error by means of a
character code. The error code can be printed out as in-
formation that an error has occurred or can be associa-
ted with the erroneous item of data in store so that
the data may be ignored or dealt with in a specific
manner when subsequently processed.

ERROR CORRECTING CODE
An error detecting code designed so that it is in some
cases possible to recognize not only that an error has
occurred, but also what the correct code should havebeen.\f

ERROR CORRECTION ROUTINE
A routine designed to detect and correct errors on fi-
les of data.

ERROR DETECTING CODE
A code in which the representation of each character is
constructed according to specific rules. Certain combi-
nations of the elements out of which the set of charac-
ters is constructed will not conform to these rules;
such combinations are known as forbidden characters and
can be recognized and rejected as errors if they occur
in a message.

ERROR DETECTION ROUTINE
A routine designed to check data items for validity and
to detect errors. See also: Validity Check.

ERROR DIAGNOSTICS
The checking of source language statements for errors
during compilation and the printing of error messages
identifying the errors made.

ERROR INTERRUPTS
An interrupt which occurs as a result of a program or
hardware error, causing a message to be printed indi-
cating the errror condition, and the suspension of the
program in which the error has occurred.

ERROR LIST
A list produced by a compiler indicating incorrect or
invalid instructions in a source program.

ERROR MESSAGE
A message output by program indicating the incidence
and type of error which has occurred.

ERROR RANGE
1.The range of values for an item of data which will
cause an error condition if the item falls within the
range. 2. In data transmission, the ratio of the total
number of transmission errors to the total volume of
data transmitted. \f

ERROR REPORT
A list of error conditions generated during the exe-
cution of a particular program, e.g., errors caused by
incorrect or unmatched data.

ERROR ROUTINE
A routine which is entered whenever an error is detec-
ted. An error routine may output an error message,
attempt to correct the error, repeat the process which
caused it or perform any other required action.

ERROR TAPE
A magnetic tape onto which errors are written for sub-
sequent listing and analysis.

ERROR CHARACTER
A character which indicates that the following charac-
ter belongs to a different character set from the pre-
ceding characters.

ERROR PARITY CHECK
A parity check in which the number of ones (or zeros)
in a group of binary digits is expected to be even. Con-
trasted with odd parity check.

EXCEPTION PRINCIPLE SYSTEM
A computer system designed so that normally only situa-
tions deviating from expected standards are reported,
while results falling within expected limits are not
reported.

EXCEPTION REPORTING
Related to an information system in which the exception
principle system is used.

EXECUTION TIME
The time taken to complete the cycle of events required
to perform an instruction.
\f

EXTERNAL STORE
A backing store which is under the control of, but not
necessarily permanently connected to, a central proces-
sor, and which can hold data or programs in a form
acceptable to it, e.g. magnetic tape, direct access
store. Also known as external memory.

EQUALIZATION
Compensation for the increase of attenuation with fre-
quency. Its purpose is to produce a flat frequency re-
sponse.

ETX (END OF TEXT)
Indicates the end of a message. If multiple transmis-
sion blocks are contained in a message in BSC systems,
ETX terminates the last block of the message. (ETB is
used to terminate preceding blocks). The block check
character is sent immediately following ETX. ETX re-
quires a reply indicating the receiving terminal's
status.

FEASIBILITY STUDIES
A feasibility study is research into the possiblitity
of developing a solution to a problem. In computer
terms this may mean placing an order for the appropria-
te configuration and the research may be primarily an
appraisal of the current situation of hardware and
software, leading to the choice of equipment. It may
also be an assessment of whether a particular area of a
company's activities should utilize a computer already
used by the company.

FEED To cause data to be entered into a computer for process-
ing; a device for so doing.

FIELD A subdivision of a record containing a unit of information.
For example, a payroll record might have the following
fields: employee number, gross pay, deductions, net pay.\f

FIELD LENGTH
The size of a field, in terms of the units in which the
record is composed, e.g. on a punched card record,
field length is measured in card columns, and a magne-
tic tape record the field length may be measured in
characters or words.

FILE An organized collection of records. The relationship
between records on a file may be that of common pur-
pose, format or data source and the records may or may
not be sequenced.

FIXED POINT ARTIHMETIC
The performing of arithmetical calculations without re-
gard to the position of the radix point, treating the
numbers as integers for the purpose of calculation. The
relative position of the point has to be controlled
during calculations.

FLOATING POINT ARITHMETIC
Arithmetical calculations based on floating point num-
bers. In floating point arithmetic the position of the
decimal point does not depend on the relative position
of the digits in the number determined the absolute
value of the number. The use of floating point arith-
metic means that numbers can be stored more economical-
ly and in wider ranges of magnitudes, and calculations
can be performed to consistent relative degrees of ac-
curacy.

FORECASTING
The problem of planning for the future is met in many
situations, and in endeavouring to make the best possib-
le plan some assumption or forecast of future condi-
tions has to be made. The best forecast will be based
on the projection and analysis of past results viewed
in the light of experience. There have developed, in
recent years, several statistical methods of making
forecasts mathematically based on past performance and
whereas no one can pretend that these methods will al-
ways give an exact forecast there can be little doubt
that the correct use of such methods can greatly impro-
ve planning.The use of digital computer commercially\f

has greatly increased the value of statistical fore-
casting methods, for when used together they enable
many forecasts to be made very quickly, where once
armies of sta tisticians would have been needed. Such
methods are of great value in such areas as market
planning, inventory control, personnel planning, etc.,
where often forecasts of many items have to be made.

FOREGROUND PROCESSING
High-priority processing, usually resulting from real-
time entries given precedence by means of interrupt,
over lower priority "background" processing.

FORMAT The predetermined arrangement of data, e.g. the layout
of a printed document, the arrangement of the parts of
a computer instruction, the arrangement of data in a
record.

FORM FEED The mechanical system of positioning continuous sta-
tionery in a printing device.

FORTRAN Fortran is an acronum for FORmular TRANslation. It is a
problem oriented high level programming language for
scientific and mathematical use, in which the source
program is written using a combination of algebraic
formulae and English statements of a standard but read-
able form.

FREQUENCY DIVISION MULTIPLEXING (FDM)
Dividing the available transmission frequency range
into narrower bands each of which is used for a sepa-
rate channel.

FREQUENCY MODULATION (FM)
A method of transmission whereby the frequency of the
carrier wave is changed to correspond to changes in the
signal wave.

FRONT END PROCESSOR
A communications computer associated with a host com-
puter. It may perform line control, message handling,
code conversion, error control and applications func-
tions such as control and operation of special-purpose
terminals.
\f

FREQUENCYThe rate of repetition of a periodically recurring sig-
nal usually measured in cycles per second (cps) kilo-
cycles per second (1kc = 1,000cps) or megacycles per
second (1mcs = 1000kcs).

FREQUENCY BAND
The range within which the frequency of a signal may be
allowed to vary.

FULL DUPLEX
See: Duplex

FULLY CONNECTED NETWORK
A network in which each node is directly connected with
every other node.

HALF DUPLEX
A circuit designed for transmission in either direction
but not both directions simultaneously.

HALF DUPLEX CHANNEL
A channel providing for transmission in both directions
but not simultaneously.

HARD COPY A document in a form suitable for human beings to read
produced at the same time as information is produced in
a language suitable for a machine.

HARDWARE The physical units making up a computer system - the
apparatus as opposed to the programs. Contrasted with
software.

HASP Houston Automattic Spooling Program. An IBM 360/370 OS
software front-end which performs job spooling and con-
trols communications between local and remote proces-
sors and Remote Job Entry (RJE) terminals.

HDLC A data communications protocal and line discipline.
High-level Data Link Control according to CCITT recom-
mendations for the interconnection of computers and dif-
ferent kinds of communication lines. It supports full
duplex synchronous communication.

HEAD An electromagnet used to read, record or erase
polarized spots on a magnetic medium such as magnetic
tape, magnetic disc or magnetic drum. Examples are read
head, write head, read/write head. \f

HEADER The control information prefixed in a message text,
e.g., source or destination code, priority, or message
type. Syn: Heading Leader.

HERTZ A unit frequency equal to one cycle per second. Cycles
are referred to as Hertz in honor of the experimenter
Heinrich Hertz. Abbreviated Hz.

HETEROGENEOUS (COMPUTER) NETWORK
A network of dissimilar host computers, such as those
of different manufacturers. Compare: Homogeneous Net-
work.

HIERARCHICAL (COMPUTER) NETWORK
A computer network, in which processing and control
functions are performed at several levels by computers
specially suited for the functions performed, e.g., in
factory or laboratory automation.

HIGH LEVEL LANGUAGE
A language in which each instruction or statement cor-
responds to several machine code instructions. High
level languages allow users to write in a notation with
which they are familiar (e.g. FORTRAN in a mathematical
notation, COBOL in English) rather than a language ori-
ented to the machine code of a computer. Contrasted
with low level language.

HOLLERITH CODE
A punched card code invented by Dr. Herman Hollerith in
1888 in which the top three positions in a card column
have a zoning significance so that a combination of a
hole in the top position (known as Y-position) plus a
hole in the fourth position would have a different sig-
nificance from a combination of a hole punched in the
second position ( know as X-position) plus a hole in
the fourth position. The third position (know as O) gi-
ves another zone and it is thus possible to code all
twenty-six alhpabetic characters and the ten numerals
0-9 in the twelve punching positions of a card.

HOMOGENEOUS (COMPUTER) NETWORK
A network of similar host computers such as those of
one model of one manufacturer. Compare: Heterogeneous
(Computer) Network.
\f

HOST COMPUTER
A computer attached to a network providing primarily
services such as computation, data base access or spe-
cial programs or programming language.

HOST INTERFACE
The interface between a communications processor and a
host computer.

IBM International Business Machines. A U.S. based corpora-
tion and manufacturer of computer equipment. The trade
name on its products.

ICL International Computers Limited. A U.K. based corpora-
tion and manufacturerr of computer equipment. The trade
mark on its products.

IN-PLANT SYSTEM
Relating to a communications system for handling data
automatically within a particular building or group of
buildings, e.g. a factory.

INPUT The processof transferring data, or program instruc-
tions, into memory from some peripheral unit. Sometimes
used to denote the data itself, sometimes to denote the
signal applied to a circuit or device. Also used as a
verb.

INPUT/OUTPUT BUFFERS
Areas of memory assigned to receive data transmitted to
or from a peripheral unit. The use of buffer areas en-
ables a number of peripheral units to be activated si-
multaneously at full speed while data is processed with-
in the central processor. On earlier machines, areas of
storage were often allocated permanently for this pur-
pose, but it is now usual to permit the programmer to
specify the locations required according to the charac-
teristics of his program.

INPUT/OUTPUT CHANNEL
A communication channel for transmitting data to and
from a central computer.
\f

INPUT STATION
In an in-plant communications system input staions may
be situated at various locations within a building to
enable personnel to input data directly into the sys-
tems as transactions or events occur. This enables fi-
les to be immediately updated, and if necessary excep-
tion reports (see also: Exception Principle System) can
be generate immediately for management.

INQUIRY AND COMMUNICATIONS SYSTEMS
Pertaining to computer systems in which central files
are maintained from data input from various sources
using data transmission equipment or in-plant networks.
Inquiries may be addressed into the system from remote
terminals, immediately producing response from the
central system.

INQUIRY DISPLAY TERMINAL
A device which consists of a keyboard and a cathode ray
tube display unit. Inquiries are specified to the com-
puter by means of messages typed on the keyboard, and
results are displayed on the cathode ray tube.

INSTALLATION OF COMPUTERS
It would be wrong to consider the process of installing
a computer as the installation of the equipment alone;
more properly one should consider the development of
all the resources necessary for the data processing
centre as a whole including the hardware, software, the
buildings, air conditioning equipment, office furniture
and specialized cabinets and furniture for storing and
handling input/output media. The recruitment and train-
ing of new staff and the education of existing person-
nel must also be an integral part of any installation
plan, and careful thought must be given to the timing
of these activities.

INSTRUCTION
The part of a computer program which tells the computer
what function to perform at that stage. Instructions
are usually examined by a special unit, sometimes known
as a program controller, which interprets each instruc-
tion and initiates the actions specified. An instruc-
tion consists of series of characters subdivided into
groups which represents code commands to the computer.
An operation such as add or subtract may be specified,

along with one or more addresses which specify the lo-
cations of operands to be used at that step.

INTEGER A whole number, i.e. one that does not contain a
fractional component.

INTEGRATED CIRCUIT
A circuit in which all the components are chemically
formed upon a single piece of semiconductor material.
Computers using integrated circuits are said to be
third generation, as contrasted with first generation
machines using thermionic valves and second generation
machines using transistors.

INTERACTIVE
Pertaining to exchange of information and control
between a user and a computer process, or between com-
puter processes. See also: Conversational.

INTERFACE 1. A shared boundary defined by common physical inter-
connection characteristics, signal characteristics, and
meanings of interchanged signals. 2. A device or equip-
ment making possible interoperation between two sys-
tems, e.g., a hardware component or a common storage
register. 3. A shared logical boundary between two
software components.

INTERNAL STORE
A term used generally as a synonym for immediate access
store; specifically a store forming part of the main
memory of a computer as distinct from a backing store.

INTERRUPT A break in a program or routine caused by an external
source, which requires that control should pass tempo-
rarily to another routine; e.g. to monitor an event
which may be proceeding in parallel to take action as a
direct result of an event which has taken place. The
interrupt is made so that the original routine can be
resumed from the point at which the break occurred.

INVENTORY CONTROL
In most industries it is the practice to hold stocks to
meet demands, for there are few occasions where demand
and supply are matched closely enough to make this un-
necessary. Even when stocks are held, temporary shor-
tages are often experienced, due perhaps to a sudden
rise in demand or delay in production. \f

The theory of inventory control, or 'stock control' as
it is often called, is applicable to all types of stock
holding and aims to strike a balance between costs of
turnover, shortages, stockholding and administration.

I/O Abbreviation for input/output.

JOB A task to produce a certain processing result on a com-
puter. A job may consist of running one or more pro-
grams.

K An abbreviatiion for kilo, used to denote a thousand.

KCS A abbreviation for a thousand characters per second.

LANGUAGE In order to communicate with each other, men use
language: in the same way 'languages' of one sort or
another are used in order to communicate instructions
and commands to a computer. The unique feature which
distin guishes a computer from other man-made tools and
devices is its versatility in dealing with vastly
different problems. Some of the more commonly used
programming language are COBOL, ALGOL, FORTRAN, RPG,
ASSEMBLER and BASIC.

LEASED LINE
A line reserved for the exclusive use of a leasing cu-
stomer without interexchange switching arrangements.
Also called Private Line.

LINE 1. The portion of a circuit external to the apparatus
consisting of the conductors connecting a telegraph or
telephone set to the exchange or connecting two exchan-
ges. 2. The group of conductors on the same overhead
route in the same cable.

LINE PRINTER
A printer which prints out results from a computer one
line at a time. Line printers are usually high-speed
devices with output speeds measured in 100's or 1000's
of lines per minute.
\f

LINK 1. Any specified relationship between two nodes in a
network. 2. A communications path between two nodes. 2.
A Data Link. See also: Line.

LOAD SHARING
The disribution of a given load among several computers
on a network.

LOCAL EXCHANGE
An exchange in which subscribers' lines terminate.

LOGIN A user access procedure to a system involving
identification, access control and exchange of network
information between user and system. Syn: Logon.

LOGOUT A user exit procedure from a system often providing
usage statistics to the user. Syn: Logoff.

LONGITUDINAL REDUNDANCY CHECK (LRC)
An error checking technique based on an accumulator ex-
clusive OR of transmitted characters. An LRC character
is accumulated at both the sending and receiving stati-
ons during the transmission of a block. This accumula-
tor is called Block Check Character (BCC), and is trans-
mitted as the last character in the block. The trans-
mitted BCC is compared with the accumulated BCC charac-
ter at the receiving station for an equal condition. An
equal comparison indicates a good transmission of the
previous block.

LOW LEVEL LANGUAGE
A language in which each instruction has a single cor-
responding machine code equivalent. Also know as basic
language. Contrasted with high level language.

LRC See Longitudinal Redundancy Check.

LSI Large scale intergreation technology in microelectro-
nics where hundreds and even a few 1000 circuits are
placed on a silicon chip.

MACHINE CODE
The coding system adopted in the design of a computer
to represent the instruction repertoire of the compu-
ter. The various operations that can be performed are\f

represented by numeric function codes and all store lo-
cations are allocated numbers to enable the data stored
in such locations to be addressed. Also know as compu-
ter code, instruction code, instruction set.

MAGNETIC CORE
A small ring of ferromagnetic material which may be po-
larized by electric currents applied to wires wrapped
around it. The magnetic core is thus capable of assum-
ing two states and may be used as a switchingdevice,
or as a storage medium. These devices have been used
extensively for the memory of computers; e.g. a single
magnetic core being used to represent a single binary
digit of some item of information.

MAGNETIC CORE STORAGE
A large array of magnetic cores arranged in matrices to
form the memory of a computer. Each individual core is
capable of assuming two states and some cores may be as-
signed as storage locations for information to be held
in binary coded form, whereas others may perform swit-
ching or gating functions.

MAGNETIC DISC
A storage device consisting of a number of flat cicular
plates each coated on both surfaces with some magnetiz-
able material. A number of tracks are available on each
surface and data is read from or written to these tracks
by means of read/write heads. There may be several
heads to each surface, a particular head being allocat-
ed a specific area (or sector) on the disc. See also:
Random Access.

MAGNETIC DISC FILE
A file of data held on a magnetic disc.

MAGNETIC TAPE
Magnetic tape is a form of backing store used for com-
puters. It is usually in the form of a continuous strip
of plastic material which is coated with a magnetic o-
xide on which data may be recorded as a series of mag-
netic spots. The general dimensions of the tape are
usually 1/2in. wide and may be approximately 2,400 feet
in length. Magnetic tape is wound on a reel usually of
10 1/2in. diameter.
\f

MAINFRAME A large computer system (mainframe) performs many va-
ried tasks from payroll, invoicing, and stock control,
to analysing results from experiments and research pro-
jects. Complete companies and organisations have to
change, adapt, and reschedule working conditions to
allow the computer to perform its tasks efficiently.
Such a system would consist of a large number, perhaps
100, of devices to input information to the computer,
fast printers to print out the results, and devices to
store the results on a long term basis. Because it's
designed to operate with masses of information being
fed into it it has to have a fast cycle time in order
to execute the tasks. The system comprises a processor
unit, input/output, interface unit, lots of fixed me-
mory, either core, semiconductor, or a mixture of both,
lots of disc memory, lots of back-up storage for record-
ing information onto mangetic tape or cards, high speed
input/output peripherals, and quite an elaborate power
supply. Each section of the system is designed for high
speed to allow fast data throughput.

MAIN MEMORY
The internal memory of a computer, i.e. the immediate
access store, as distinct from any backing store that
may be available as part of the computer system.

MANAGEMENT INFORMATION SYSTEM
A system which may perform routine commercial process-
ing functions, but which is designed so that much pro-
cessing will also produce information that will be pre-
sented to management, including top management, to
assist in decision making. The implication is that the
results will be produced speedily, perhaps requiring
real time processing, to enable management to ascertain
the progress of the organization in terms of satisfying
its major objectives.

MATRIX PRINTER
Matrix or array dot-printing techniques are sometimes
referred to as "mosaic print". In these techniques,
charaters are formed by a number of pins appropriately
selected from an array or matrix (usually 5 columns by
7 rows) that strikes the paper. The pins are generally
electromagnetically or hydraulically actuated, can be
moved quickly, and attain printing speed of up to 200
lpm or more.
\f

MEGA- A million; as in 10 megacycles per second, meaning 10
million cycles per second.

MEMORY This term is usually reserved for describing the inter-
nal store of a computer, i.e. the immediate access sto-
re. In its strictest sense it refers to the storage lo-
cations that can be immediately addressed by the pro-
gram controller of the central processor, rather than
to any backing store medium such as magnetic tape or
magnetic disc storage. However, these backing stores
are sometimes referred to as memory units, as in disc-
file memory, in which case the internal storage would
be referred to as main memory. Also known as immediate
access storage, store, core store.

MESSAGE SWITCHING
A method of handling messages over communications net-
works. The entire message is transmitted to an inter-
mediate point (i.e., a switching computer), stored for
a period of time, perhaps very short, and then trans-
mitted again towards its destination. The destination
of each message is indicated by an address integral to
the message.

MICROCOMPUTER
It has the electronic circuits containing the essential
computer functions on a single printed circuit board.
It comprises a processor unit, some memory and a basic
I/O interface. It can be used on its own to perform
functions but is more often embedded in a mini or main-
frame computer system to perform functions where logic
control is more important than the power to process
large amounts of data.

MICROPROGRAM
A program stored on a single printed circuit board or
on a silicon chip.

MICROPROCESSOR
Comprises the three fundamental sections that make up a
processor using semiconductor 'large scale intergration'
manufacturing techniques. It comprises an arithmetic
logic unit, I/O interface and memory transfer on a sing-
le silicon chip. Usually used in electronic equipment
to implement control and logic functions.
\f

MICROSECOND
One millionth of a second, expressed Usec.

MICROWAVE Pertaining to data communications systems in which ul-
tra-high-frequency waveforms are used to transmit voice
or data messages.

MILLESECOND
One thousandth of a second

MINICOMPUTER
Evolutionary development in semiconductor electronics
gave us the minicomputer. A system consisting of a pro-
cessor unit, some memory, an interface I/O unit and a
power supply. Unlike the mainframe, the minicomputer is
often sold without storage, peripherals, and large ca-
pacity fixed memory because the processor does not de-
mand special high speed or complex techniques for inter-
facing thus allowing the customer to pick and choose
from a wide variety of equipment to make up the comple-
te computer system.

MODEM Modulator-Demodulator. A device that modulates signals
transmitted over communications cicuits. Syn: Data Set.

MODULAR A method of constructing a hardware or software system
using standard compatible units. In this way a wide ran-
ge of configurations can be built up with combinations
of the standard units.

MODULARITY
The condition exhibited by any hardware or software sys-
tem that permits the subsequent expansion of the system
by the addition of standard modular units.

MODULATION
A technique used in radio, telegraphic and telephonic
communication systems, in which data signals are used
to modify either the amplitude or frequency of a car-
rier wave by means of modems (modulator/demodulator).
The carrier wave is of a suitable frequency for trans-
mitting over a specified channel, and therefore carries
with it the data signals which mormally would not be
capable of transmission over the cicuit concerned.\f

MODULATION CODE
A coded signal used to modulate the frequency or
amplitude of a carrier wave.

MODULATOR A device which superimposes a data signal on a carrier
wave according to a predetermined method.

MODULE A hardware device or software item which, as a standard
unit, forms part of a modular system.

MSI Medium scale integration technology in electronics
where up to 100 circuits are placed on a silicon chip.

MULTILEAVING
A technique which allows simultaneous bidirectional
communications traffic; e.g., output from a previous
remote batch job may be received while a new job is
being transmitted.

MULTIPLEX 1. A multiplex system involves the transfer of data
from several comparatively slow-speed storage devices
over a series of channels to a fast central storage de-
vice which continually scans and accepts data from each
channel in turn. The fast storage device is able to
service the channels without any part of the system
being delayed. 2. To transmit a number of messages con-
currently over the same circuit. A division of a trans-
mission facility into two or more channels.

MULTIPLEXER
A device used for multiplexing. It may or may not be a
stored program computer.

MULTIPLEXING
1. Simultaneous transmission of several messages over a
single communication channel, usually by modulating a
carrier wave in such a way that separate signals are
transmitted using particular frequencies within the
full bandwidth of the channel. 2. Pertaining to any
system in which a single device is used for many pur-
poses.

MULTIPLEX CHANNEL
A channel allowing the interleaving of many simulta-
neous transmissions in both directions. \f

MULTIPLICAND
One of the factors used in multiplication: a quantity
which is multiplied by another.

MULTI-POINT LINE
A single communications line to which more than one ter-
minal is attached. Use of this type of line normally
requires some kind of polling mechanism, addressing
each terminal with a unique ID. Also called: "Multi-
Drop".

MULTIPROCESSING SYSTEM
A computer system which contains two or more intercon-
nected processors, each with its own arithmetic and
logical units and each capable of independent opera-
tion. See also: Multiprogramming, for which a single
processing unit is used on a time sharing basis to
operate several programs simultaneously.

MULTIPROCESSOR
A central processor containing two or more independent
arithmetic units and their associated control logic.

MULTIPROCESSOR INTERLEAVING
A technique for allocating memory areas to the diffe-
rent processors within a multiprocessing system. The
store is subdivided into modules which are referenced
as even or odd, and the addressing structure for the
locations within any module remains as in the standard
machine code. In this way a number of modules are al-
located to each processor to avoid interaction between
programs being run simultaneously.

MULTIPROGRAMMING
There is an extreme difference in the speeds at which a
computer handles its internal operations (performing
calculations on data in its central processor) and the
speeds at which even the fastest peripheral units used
for input and output of data operate. In the time taken
to print one line on a printer working at 1,200 lines
per minute a processor could perhaps perform something
of the order of fifty thousand additions of more. Mul-
tiprogramming is a technique developed in order to
utilize a computer more efficiently by enabling the
processor to spend a greater proportion of its time in
action and by making more use of all available peri-
pheral units. The basic principle of multiprogramming\f

is that more than one program can be present in memory
at the same time, and share the available processor ti-
me and peripheral units.

NANOSECOND
One thousand-millionth of a second or 1 billionth of a
second.

NARROWBAND CHANNELS
Sub-voice grade channels characterized by a speed range
of 100 to 200 bits per second.

NEGATIVE ACKNOWLEDGMENT (NAK)
Indicates that the previous transmission block was in
error and the receiver is ready to accept a retransmis-
sion of the erroneous block. NAK is also the "not
ready" reply to a terminal selection (multipoint) or to
an initialization sequence (line bid) in point-to-point
operation.

NETWORK 1. An interconnected or interrelated group of nodes. 2.
In connection with a disciplinary or problem oriented
qualifier, the combination of material, documentation
and human resources that are united by design to achie-
ve certain objectives, e.g., a social network, a scien-
ce information network.

NETWORK CONTROL PROGRAM
That module of an operating system in a host computer
or front end, which establishes and breaks logical con-
nections, communicating with the network on one side,
and with user processes within the host computer, on
the other side.

NETWORK SECURITY
The totality of measures taken to protect a network
from an unauthorized access, accidential or willful in-
terference with normal operations, or destruction. This
includes protection of physical facilities, software,
and personnel security. See also: Privacy.

NETWORK TOPOLOGY
The geometric arrangement of links and nodes of a net-
work.\f

NODE An end point of any branch of a network, or a juction
common to two or more branches of a network.

NOISE 1. Any disturbance affecting the characteristics of a
signal e.g. random variations in voltage, current or
frequency. 2. Errors in data generated by disturbance
in a circuit, particularly in a data transmission cir-
cuit.

NON-TRANSPARENT MODE
Transmission of characters in a defined character for-
mat, e.g., ASCII or EBCDIC, in which all defined con-
trol characters and control character sequences are
recognized and treated as such.

ODD PARITY CHECK
A parity check in which the number of ones (or zeros)
in a group of binary digits is expecteed to be odd.
Contrasted with even parity check.

OFF-LINE Pertaining to equipment or devices not under control of
the central processing unit.

ON-LINE 1. Pertaining to equipment or devices under control of
the central processing unit. 2. Pertaining to a user's
ability to interact with a computer.

OPERATING SYSTEM
Software that controls the execution of computer pro-
grams that may provide scheduling, debugging, input and
output control, accounting, storage assignment, data
management, and related service. Sometimes called Su-
pervisor, Executive, Monitor, Master Control Program
depending on the computer manufacturer.

PACKET A group of bits including data and control elements
which is switched and transmitted as a composite whole.
The data control elements and possibly error control
information are arranged in a specified format.The
basic unit of information exchanged with and within the
Packet Switch.
\f

PACKET SWITCH
A protocol within RCNET. It accepts packets from hosts
and routes them towards the host specified as recei-
ver.

PACKET SWITCHING
A data transmission process, utilizing addressed pa-
ckets, whereby a channel is occupied only for the
duration of transmission of the packet.
NOTE: In certain data communication networks the data
may be formatted into a packet or divided and then for-
matted into a number of packets (either by the data ter-
minal equipment ofr by the equipment within the net-
work) for transmission and multiplexing purposes.

PACKET TRANSPORTER
A protocol within RCNET. The protocol is implemented in
hosts. It controls the transmission of packets through
the Packet Switch and includes mechanisms to detect the
loss of a packet.

PAGE PRINTER
A printer for which the character pattern for a comple-
te page is determined before printing. Contrasted with
line printer.

PAPER TAPE
Punched paper tape has been used as an input/output me-
dium since the earlist developments of electronic digi-
tal computers. To some extent it was used initially
simply because 5-track paper tape was already in use to
meet the requirements of the telegraph service, and
paper tape readers and paper tape punches were adapted
for use as input/output units.

PAPER TAPE READER
A device which translates the information punched in
code on paper tape into machine language and transmits
the data into a central processor.

PARALLEL TRANSMISSION
Method of data transfer in which all bits of a charac-
ter or byte are transmitted simultaneously either over
separate communication lines or on different carrier
frequencies on the same communication line.
\f

PARITY BIT
1. A check bit whose value (0 or 1) depends on whether
the sum of 1 bits in the word being checked is odd or
even. If the total number of 1 bits, including the pa-
rity bit, is even, the word is known as having even
parity; if the number is odd, it has odd parity.

PARITY CHECK
A check made when data is transferred which consists in
addding up the bits in a unit of data, calculating the
parity bit required and checking the calculation parity
bit with the parity bit transferred with the data item.
This form of check will normally be performed automati-
cally by hardware.

PARITY ERROR
An error caused by incorrect parity detected as a re-
sult of a parity check.

PASSWORD A word or string of characters that is recognizable by
automatic means and that permits a user access to pro-
tected storage, files, or input or output devices.

PERIPHERAL UNITS
Machines which can operate under computer control. Pe-
ripheral equipment consists of input devices, ouput
devices and storage devices.

PHASE MODULATION (PM)
A method of transmission whereby the angle of phase of
the carrier wave is varied in accordance with the sig-
nal.

PLUG COMPATIBLE
Referring to peripheral devices, memory and other units
made by a competitive computer equipment manufacturer,
which can simply be plugged in to the computers of an-
other manufacturer and operate as if they were the
peripherals etc. made by that manufacturer.

POINT-TO-POINT CONNECTION
1. A network configuration in which a connection is
established between two, and only two, terminal instal-
lations. The connection may include switching facili-
ties. 2. A circuit connecting two points without the
use of any intermediate terminal or computer.
\f

PRINTER An output device which converts data into printed form.

PRIVACY The right of an individual to the control of informa-
tion about himself.

PRIVATE LINE
A communications line permanently leased or owned by
the user. It can also imply a line between terminal/s
and computers on the users premises or properby.

PROCESS 1. A systematic sequence of operations to produce a
specified result. 2. A set of related procedures and
data undergoing execution and manipulation by one or
more computer processing units.

PROGRAM A set of instructions composed for solving a given pro-
blem by computer.

PROGRAMMING
Programming is the process by which a set of instruc-
tions is produced for a computer to make it performa
specified activity. The activity can be anything from
the solution of a mathematical problem to the produc-
tion of a company payroll. The instructions ultimately
obeyed by the computer are the numerical codes signi-
ficant to the computer's central processor. Since a
computer cannot reason, it is entirely dependent on in-
structions supplied to it by its all too human user.

PROGRAM SHARING
The ability for several users or computers to utilize a
program at another node.

PROGRAM TESTING
Checking a program (in order to establish that it is
performing expected operations correctly and that all
errors have been discovered) by running the program on
a computer usually with test data.

PROTOCOL A formal set of conventions governing the format and
relative timing of message exchange between two com-
municating processes.

RANDOM ACCESS
Pertaining to a storage device where data or blocks of
data can be read in any particular order. In random
access devices you do not have to read from the beginn-
ing to find what you want as you do with paper tape and
magnetic tape.

RC A/S Regnecentralen
A Denmark based corporation and manufacturer of compu-
ter equipment. The trade name on its products.

RCLC Regnecentralen Link Control. A High-Level Data Link
Control protocol applied to RC equipment and networks.
It supports synchronous transmission according to CCITT
recommendations.

RC NET A computer network developed and marketed by the
Regnecentralen company. It offers a general transpor-
tation service based on packet switching technology.

READ/WRITE HEAD
An electromagnet used to read or write on a magnetic
medium such as magnetic tape or magnetic disc.

REAL TIME CLOCK
A mechanism whereby the time of day can be monitored by
a computer system. It increments automatically by one
interval every millisecond (or other fixed period). Such
a device is used in real time systems to calculate the
actual or elapsed time which passes between the occur-
rence of two events. Often called an interval timer.

REAL TIME SYSTEM
A system performing computation duriong the actual time
the related physical process transpires, so that the
results of the computation can be used in guiding the
process.

RECORD A unit of data representing a particular transaction or
a basic element of a file consisting in turn of a num-
ber of interrelated data elements.

RECORD BLOCKING
The practice of grouping records into data blocks.
\f

REMOTE JOB ENTRY
1. Submission of jobs through an input device that has
access to a computer through a communications link. 2.
The mode of operation that allows input of a batch job
by a card reader at a remote site and receipt of the
output via a line printer or card punch at a remote
site. Abbr: RJE.

RESOURCE Any means available to network users, such as computa-
tional power, programs, data files, storage capacity,
or a combination of these.

RESOURCE SHARING
The joint use of resources available on a network by a
number of dispersed users.

RESPONSE TIME
The elapsed time between the generation of the last
character of a message at a terminal and the receipt of
the first character of the reply. It includes terminal
delay, and service node delay.

SCIENTIFIC LANGUAGE
A language designed for the writing of mathematical or
scientific programs, e.g. Fortran.

SDLC Synchronous Data Link Control. A uniform discipline for
the transfer of data between terminals in a point-to-
point, multipoint, or loop arrangement, using
synchronous data transmission techniques.

SERIAL TRANSMISSION
A method of transmission in which each bit of informa-
tion is sent sequentially on a single channel rather
than simultaneously as in parallel transmission.

SIMPLEX MODE
Operation of a channel or line in one direction only
with no capability of reversing.

SIMULATION
A system, either hardware or software, designed to per-
form the simulation of some real process.
\f

SOFTWARE A set of computer programs, procedures, rules and
associated documentation concerned with the operation
of network computers, e.g., compilers, monitors,
editors, utility programs. Compare: Hardware.

SORTING Usually data which is to be processed requires to be
ordered and presented in a predetermined sequence. The
need for sorting arises after the data has been tran-
scribed onto a computer input medium, e.g. punched
cards, magnetic tape.

SOURCE DATA
Data from transactions at origin and still to be
processed.

SPOOLING The technique by which output to slow devices is placed
into queues on mass storage devices to await transmis-
sion. This allows more efficient use of the system since
programs using low-speed devices can run to completion
quickly and make room for others.

SSI Smale scale integration technology in electronics where
up to 10 circuits are placedon a silicon chip.

START OF TEXT
A communication control charater which precedes a se-
quence of characters that is to be treated as an entity
and entirely transmitted through to the ultimate de-
stination. Such a sequence is referred to as text. STX
may be used to terminate a sequence of characters
(heading) started by SOH.

START-STOP TRANSMISSION
Asynchronous transmission in which a group of code ele-
ments corresponding to a character signal is preceded
by a start element and is followed by a stop element.

STATIONERY
Stationery is used in computer systems to record re-
sults produced on a printer or typewriter. Most prin-
ters make use of continuous forms, driven through the
printer by means of spocket holes at each edge. Pages
are separated from each other by perforations. Form
designs may be preprinted on each page, the computer
supplying the variable information, e.g. invoice\f

details. Multi-part sets of stationery enable several
copies to be printed of each page, either by means of
interleaved carbon paper or specially treated chemical
surfaces.

STATION An independently-controllable configuration of data
terminal equipment from or to which messages are trans-
mitted on a data link. Mainly synchronous with Terminal
Installation.

STORAGE Synonymous with store.

SUPERVISORY PROGRAMS
Computer programs that have the primary function of
scheduling, allocating, and comtrolling system re-
sources rather than processing data to produce results.

SWITCHED LINE
A communication link for which the physical path may
vary with each usage, e.g., the dial-up telephone net-
work.

SYNCHRONOUS TRANSMISSION
Transmission in which the data characters and bits are
transmitted at a fixed rate with the transmitter and
receiver synchronized. This eliminates the need for
start-stop elements, thus providing greater efficiency.
Compare: Asynchronous Transmission.

SYSTEM Any group of objects related or interacting so as to
form a unit. In data processing the objects which are
interrelated will be individuals and machines, the pur-
pose of their interacting being to achieve certain de-
fined ends concerned with the manipulation of informa-
tion, e.g. to produce a payroll. System can also be
used to refer specially to a particular interrelated
collection of machines (in this sense, the term
configuration is also used).

SYSTEMS SOFTWARE
A general term applied to all software other than app-
lications software. It includes; operating systems, mo-
nitors, utility programs, data management, network con-
trol, etc. \f

TARIFF 1. A published rate for services provided by a common
or specialized carrier. 2. The means by which regulato-
ry agencies approve such services. The tariff is a part
of a contract between customer and carrier.

TELECOMMUNICATION
The transmission and reception of data over radio cir-
cuits or transmission lines by means of electromagnetic
signals.

TELEGRAPHIC COMMUNICATION
Communication by means of signals in an on/off mode.
For example, the earliest telegraph systems were ope-
rated by a manual key which interrupted current in a
circuit to produce audible clicks for a distant ope-
rator. This techniqe was developed and the teleprinter
(and associated equipment) emerged as the principal
device for this form of communication.

TELEMETER Equipment for recording and transmitting measurements
as data to a distant location by eletromagnetic waves.

TELEX SERVICE
A Western Union world-wide teletypewriter exchange ser-
vice that uses the public telegraph network. Baudot
equipment is used.

TELEPRINTER
A device resembling a typewriter which can be connected
to a communications link to receive signals or transmit
data from and to a distant point. Incoming signals may
be received on a print unit forming part of the tele-
printer, and on some machines the incoming signals may
also be automatically recorded on paper tape. Outgoing
messages may be entered on a keyboard and, on some te-
leprinters, may be transmitted automatically by the
passing of a message tape through a paper tape reading
unit. Teleprinters are sometimes used for direct input/
output to a central processor.

TERMINAL A device or computer which may be connected to a local
or remote Host system, and for which the Host system
provides computational and data access services. Two
common types of terminals are timesharing (typically
interactive keyboard terminals) and remote batch, such
as the IBM 2780 and terminal simulation by RC 3600.\f

TERMINAL INSTALLATION
1. The totality of equipment at a user's installation
including data terminal equipment, data communication
equipment, and necessary support facilities. 2. A set
composed of data terminal, a signal converter, and pos-
sibly intermediate equipment; this set may be connected
to a data processing machine or may be part of it.

TIE LINE A private line communications channel of the type pro-
vided by communications common carriers for linking
two or more points together.

TIME-DIVISION MULTIPLEXING
A system of multiplexing in which channels are establi-
shed by connecting terminals one at a time at regular
intervals by means of an automatic distribution.

TIMER CLOCK
An electronic device used for timing events that occur
during the operation of a computer system. Situated in
the central processor, it can provide data for changing
out computer time, monitor operations to detect looping
and similar error conditions and provide time in hours
and minutes for maintaining an operating log.

TIME-SHARING
A method of operation in which a computer facility is
shared by several users for different purposes at (ap-
parently) the same time. Although the computer actually
services each user in sequence, the high speed of the
computer makes it appear that the users are handled
simultaneously.

TRANSACTION
Any event which requires a record to be generated for
processing in a data processing system. Also refers to
the record itself.

TRANSACTION DATA
A collection of characters or digits, representing one
or more events, which requires to be accepted into a
data processing system to update a master file or to
generate results. In a real time system such data may
arise at random and must be dealt with as it occurs; in
a batch processing system transactions are batched to\f

form groups which are sorted and applied to the master
files at predetermined periods.

TRANPARENT MODE
Transmission of binary data with the recognition of
most control characters suppressed. In Binary Synchro-
nous Communications, entry to an exit from the trans-
parent mode is indicated by a sequence beginning with a
special Data Link Escape (DLE) character.

TRANSFER RATE
The rate at which data may be transferred between a pe-
ripheral unit and the main memory of a computer. Depen-
dent upon the speed and operational mode of the peri-
pheral and upon the speed of the memory.

TRANSISTOR
A small solid-state semiconductor which can operate as
an amplifier or as a switching device. Transistors are
usually constructed of silicon or germanium; they are
small and light and have very fast switching speeds.

TURNAROUND TIME
1. The elapsed time between submission of a job to a
computing center and the return of results. 2. In com-
munications the actual time required to reverse the
direction of transmission from the sender to receiver
or vice versa when using a two-way alternate circuit,
e.g. a half-duplex line. Time is required by line pro-
pagation effects, modem timing and computer reaction.

UNIVAC Sperry Univac Corporation. A U.S. based corporation and
manufacturer of computer equipment. The trade name on
its own products.

UPDATE To apply transactions to a data file in order to amend,
add or delete records and thus ensure that the file re-
flects the latest situation.

USASCII See ASCII \f

VALIDITY CHECK
A check to ensure that data falls within certain pre-
scribed limits, e.g. that numerals do not appear in a
field which should have been alphabetic characters on-
ly, that a field for days of the month does not contain
a number over 31, etc.

VERTICAL REDUNDANCY CHECK (VRC)
A check or parity bit added to each character in a mes-
sage such that the number of bits in each character,
including the parity bit, is odd (odd parity) or even
(even parity).

VISUAL DISPLAY UNIT
A display unit that consist of a cathode ray tube which
is used to display characters or graphs representing da-
ta read from the main memory of a computer. A visual
display unit also incorporates facilities to key in in-
quiries so that computer files can be interrogated from
remote locations.

VOICE-GRADE CHANNEL
A channel used for speech transmission usually with an
audio frequency range of 300-to-3400 Hertz. It is also
used for transmission of analog and digital data. Up to
10,000 bits per second can be transmitted on a voice-
grade channel.

WACK (WAIT BEFORE TRANSMITTING POSITIVE ACKNOWLEDGMENT)
In Binary Synchronous Communications, this sequence is
sent by a receiving terminal to indicate that it is
temporarily not ready to receive.

WAITING TIME
The waiting time of a computer store is the interval be-
tween the moment a control unit calls for a transfer of
data to or from the store and the moment the transfer
begins. Also know as latency.

WATS Wide Area Telephone Service. A service provided by
telephone companies in the United States that permits a
customer to make calls to or from telephones in speci-
fic zones for a flat monthly charge. The monthly char-
ges are based on size of the zone instead of number of
calls. WATS may be used on measured-time or full-time
basis. \f

WHEEL PRINTER
A printer which prints characters from the rim of a
print wheel. The available characters are placed round
the rim of each wheel and there is a wheel for each
printing position.

WIDEBAND Communications channel having a bandwidth greater than
a voice-grade channel characterized by data transmis-
sion speed of 10,000-to-500,000 bits per second.

WILLIAMS TUBE
An electrostatic storage cathode ray tube which uses a
cathode ray tube with only one gun assembly. The device
was developed by F.C. Williams of the University of
Manchester and was a significant advance in the devel-
opment of digital representation and storage.

WORD A basic unit of data in a computer memory; the unit
will consist of a predetermined number of characters or
bits to be processed as an entity, i.e. a program
instruction or an element of data. In many digital
computers a fixed-word length is used, but in other
machines characters may be grouped to form words of
variable length according to the requirements of the
particular instructions to be performed.

WORD LENGTH
The size of a word, measured by the number of digits it
contains, e.g. a 24-bit word will be able to hold num-
bers in the range -2UU23DD to +2UU23DD -1.

WORD ORIENTED
A computer is said to be word oriented if the basic e-
lement of data which can be individually addressed in
store is a word. The individual bits or characters with-
in the word may be accessed by the use of certain in-
structions if required. Contrasted with character ori-
ented.

WRITE To transcribe data onto a form of store from another
form of store, e.g. transcribing data onto a magnetic
tape from the main memory of a computer. Data is 'writ-
ten to' tape rather than 'written on' tape. Contrasted
with read.
\f

WRITE HEAD
An electromagnet used to write on a magnetic medium
such as magnetic tape, magnetic disc, or magnetic drum.
Also known as recording head or writing head.

XEROGRAPHIC PRINTER
A page printer (i.e. a printer in which the character
pattern is set for a whole page before printing) using
the principle of xerography.

XEROGRAPHY
A dry copying process: the image to be copied is pro-
jected onto a plate causing an electrostatic charge to
be discharged where the light falls and retained where
the image is black. Resinous powder is then tumbled
over the plate, adhering only to the uncharged areas.
The resin is transferred to paper or other medium for
use as a printing master. \f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

1. INTRODUCTION ........................................... 1

2. CALL ................................................... 2

3. FUNCTION ............................................... 3

4. EXAMPLES ............................................... 4

5. ERROR MESSAGES ......................................... 5
\f

ii
\f

1_._ _ _ _ _ _ _ _ _I_N_T_R_O_D_U_C_T_I_O_N_ 1.

The program initamx is used to change characteristics of ter-
minals connected to RC8000 via an RC3600 Front end or an RC3600
Remote devicecontroller.

The parameters to be changed include:

- terminal type (conversion tables)
- input timer

For terminals connected via the RC3682 asynchronous multiplexer:

- number of stopbits
- parity
- character length
- bitrate

The program can run under the operation system 's' and 'sos'.
Under 'sos' the process must be included as user of the subhost
involved (ex. include 17).

The program is intended for use on installations with Basis
System Package, SW8001/1, Release 3.0, 1981.08.01 or younger
releases.
\f

F_ 2_._ _ _ _ _ _ _ _ _C_A_L_L_ 2.

M_m_m_ *
initamx <hostspec> (<hostspec> <termspec>)
P_p_p_ 1
<hostspec>::= host <hostid>
<termspec>::= terminal dklink.<no>,
M_m_m_ 1 1
(type.<termtype>) (timer.<timer>)
P_p_p_ 0 0,
M_m_m_ 1 1 1 1
(s.<stopbits>) (p.<parity>) (1.<charlength>) (r.<bitrate>)
P_p_p_ 0 0 0 0

<hostid> => devicehost id (ex. 5014)
<no> => devicehost linknumber of terminal
<channo> => channel number of input line

<termtype>::= 0..9 => the terminal kind (0 (=VDU) is default)
<timer> => timer value of input message (default : 60 sec)
<stopbits>::= 1, 2 => number of stopbits (default : 2)

<parity>::= n,o,e => parity of the line
n : no parity
o : odd parity
e : even parity (default)

<charlength>::= 5,6,7,8 => length of data excl. parity bit
(default : 7)
<bitrate>::=
40,50,75,110,134,150,200,220,300,600,1200,2400,4800,9600
=> line speed in bps (default : 2400)
\f

F_ 3_._ _ _ _ _ _ _ _ _F_U_N_C_T_I_O_N_ 3.

The program creates a link to the devicehost specified by
<hostspec> and terminals on that devicehost are initialized until
a new devicehost is selected.
\f

F_ 4_._ _ _ _ _ _ _ _ _E_X_A_M_P_L_E_S_ 4.

initamx host 5014 terminal chan.0 r.300,
host 5015 terminal driver.cx0 chan.128,
p.n 1.8 r.4800 terminal driver.cx0 chan.129.
\f

F_ 5_._ _ _ _ _ _ _ _ _E_R_R_O_R_ _M_E_S_S_A_G_E_S_ 5.

devicehost no <devicehost> not found.
- the devicehost does not exist in the network.

link error: <cause>
connecting to devicehost no <devicehost>

<cause>::= supervisor device not present /
supervisor device reserved /
no resources at jobhost /
no resources at devicehost /
timeout /
priority /
link present /
device unknown.

- an erroneous result was received from the linkup message to
host.

terminal not found.
- a terminal coroutine with the specified channel and driver does
not exist.

no link number specified.
- no link number was specified in <termspec>.

no devicehost connected.
- no devicehost was selected before a terminal initialization.
\f

F_
\f

«eof»

DataMuseum.dk

CP/M

⟦e190ffa18⟧ TextFile

Derivation

TextFile