Softcover reprint ofthe hardcover 1st edition
1986978-0-333-39083-2
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission. No paragraph of
this publication may be reproduced, copied or transmitted save with
written permission or in accordance with the provisions of the
Copyright Act 1956 (as amended). Any person who does any
unauthorised act in relation to this publication may be liable to
criminal prosecution and civil claims for damages.
First published 1986 by THE MACMILLAN PRESS LID London and
Basingstoke
Associated companies in Auckland, Delhi, Dublin, Gaborone, Hamburg,
Harare, Hong Kong, Johannesbur~, Kuala Lumpur, Lagos, Manzini,
Melbourne, Mexico City, Narrobi, New York, Singapore, Tokyo.
British Library Cataloguing in Publication Data
Longley, Dennis Macmillan dictionary of personal computing and
communications. 1. Microcomputers - Dictionaries I. Title 11.
Shain, Michael OO~.16'03'21 QA 76.5 ISBN 978-0-333-42170-3 ISBN
978-1-349-18341-8 (eBook) DOI 10.1007/978-1-349-18341-8
The paperback edition of this book is sold subject to the condition
that It shall not, by way of trade of otherwise, be lent, resold,
hired out, or otherwise circulated without the publisher's prior
consent in any form ofbinding or cover other than that in which it
is published and without a similar condition including this
condition being imposed on the subsequent purchaser.
abbreviated addressing In programming, a process that enables a
user to employ an address having fewer characters than the full
address. lt provides a faster me ans of pro cessing data because
the shorter address requires less time to decode. See
ADDRESS
ING.
ABEND In operations , an ABnormal END of a task prior to its
completion on a mainframe because of an error. A micro computer in
emulation mode should be programmed to handle this in a controlled
manner. See TERMINAL EMULATION .
abort In operations, to terminate, in a controlled manner, a
processing activity in a computer system because it is impossible
or undesirable for the activity to proceed.
A
abscissa In mathematics, the horizontal axis of a coordinate graph.
Compare ORDINATE. See CARTESIAN COORDINATES, co ORDINATE
GRAPH.
absolute address In programming, (1) an address in a computer
language that identi fies a storage location or a device without
the use of any intermediate reference, (2) an address that is
permanently assigned by the machine designer to a storage location,
(3) a pattern of characters that identifies a unique storage
location or device without further modification. See ADDRESS.
absolute assembler In programming, a specific type of assembly
language program designed to produce binary programs con taining
only absolute addresses and address references. See ABSOLUTE
ADDRESS, ASSEM
BLING.
absolute code In programming, a code that uses computer
instructions with absolute addresses. See ABSOLUTE ADDRESS.
absolute loader In programming, a routine that reads a computer
program into main memory, beginning at the assembled origin. See
ASSEMBLING, LOADER , MAIN MEMORY.
absolute value In mathematics, the value of a number regardless of
a prefixed plus or minus sign, i.e. the absolute value of -5 is
5.
abstract data type In data structures, a data type that is defined
solely in terms of the operations that can be performed on objects
of that type and the range of values that it can take, without
regard to the method of representation of the value. See
TYPE.
AC See CURRENT.
ALTERNATING
acceleration potential In electronics, the voltage between the
cathode and the face of the tube which attracts the beam of focused
electrons, causing them to impinge on the phosphor dots. See CRT,
PHOSPHOR DOTS.
acceptance testing In operations, aseries of tests designed to
demonstrate the functional capabilities of a new computer system.
lt is usually conducted by the manufacturer to show the customer
that the system is in working order.
ACCESS US Army Automated Catalog of Computer Equipment and Software
Systems.
access In programming, the manner in which files or data sets are
referred to by the computer. See DIRECT ACCESS, RANDOM
ACCESS, SEQUENTIAL ACCESS.
access arm In backing storage, a mech anical device in a disk
drive that positions the reading and writing mechanisms. See DISK
DRIVE, HEAD.
2 access barred
access barred In data communications, a data faciJity that permits
a terminal instal lation to make outgoing, or receive incom ing,
calls but not both.
access control (1) In computer networks, the control of system
usage, imposed by hardware, software and administrative con trols.
Such controls include system monitor ing, user identification,
ensuring data in tegrity, recording system access and changes and
methods for granting user access. See HARDWARE, SOFTWARE. (2) In
databases, the control of the use of the database information.
Access to a particular user may be restricted both in terms of the
data items that may be made available and the operations that may
be performed on it, e.g. read, upqate etc.
access line In data communications, a tele communication line that
continuously con nects a remote station to a DSE. A tele phone
number is associated with such lines. See DSE.
access time In backing storage, the time interval between the
instant that data is requested from the storage device to the
instant that it is delivered to the CPU, and vice versa. The
average access time for magnetic disk systems is usually based upon
the assumption that the read/write heads will have to move over
one-third of the maxi mum number of tracks in a random seek
operation. Compare SEEKS PER HOUR. See CPU, MAGNETIC DISK,
READ/WRITE HEAD,
TRACK.
accidental destruction In data security, the unintentional
overwriting or deletion of data, e.g. by faulty hardware or
software. Backup is needed for recovery. See BACKUP
COPY.
accordion fold In pnntmg, a method of folding paper in which each
fold is in the opposite direction to the previous one. A printer
can be fed with accordion-folded paper without continuous operator
inter vention.
accumulator In architecture, a device that functions as a holding
register for arithmetic, logical and input/output operations. In
some CPUs, data words fetched from memory are loaded into the
accumulator and words to be stored into memory are first loaded
into this register. See CPU, INPUT/OUTPUT, REGISTER.
accuracy In mathematics, the degree of exactness of an
approximation or measure ment. It denotes the absolute quaJity of
the result with respect to its true value, as compared with
precision which is concerned with the amount of detail used in
specifying a result. Thus a two-digit result may be more accurate
than an incorrect three-digit result, but it will be less precise.
See PRECISION.
Aces In computer languages, a continuous simulation language. See
CONTINUOUS SIMU
LATION.
achromatic color In computer graphics, an intermediate gray level
in the monochro matic gray scale. See GRAY SCALE.
AC IA See ASYNCHRONOUS COMMUNICA-
TIONS INTERFACE ADAPTOR.
access time
Memory cost/performance trade off.
0.0001
acknowledge character In data com munications, a character
transmitted by a station as an affirmative response to the station
with wh ich the connection has been set up. Compare NEGATIVE
ACKNOWLEDGE
MENT. See ACKNOWLEDGEMENT, STATION.
acknowledgement In data communica tions, the transmission by a
receiver of acknowledge characters as a response to a sender. See
AFFIRMATIVE ACKNOWLEDGE
MENT, NEGATIVE ACKNOWLEDGEMENT.
ACM See ASSOCIATION FOR COMPUTING MACHINERY.
acoustic coupler In data communications, a device to interface an
item of equipment, producing digital signals, to a telephone
network. Sound transducers in the acoustic coupler produce sound
tones corresponding to the digital signals; a telephone handset is
placed in contact with the sound transducers so that these tones
can be input into the telephone network. See MODEM, TRANS
DUCER.
action message In operations, a message issued because of a
condition that requires an operator response.
activation In computer networks, the process by which a component
of anode is made ready to perform the functions for which it was
designed. See NODE.
aetive device In e1ectronics, a circuit that contains an amplifier
providing gain. Com pare PASSIVE DEVICE. See AMPLIFIER,
GAIN.
active file In data structures, a permanent or temporary file,
having an expiration date that is later than the job date. See
JOB.
active state In microelectronics, the digital state that causes a
given action to occur. It may be either the high state or low
state, depending on the circuit and pin in question.
aetive wire concentrator In computer net works, cabinets with star
connection to individual nodes. In some local area net works, they
are connected in a ring, thus providing a combined ring/star
configur ation. See LOCAL AREA NETWORK, NODE, RING, STAR.
aetivity In data processing, the percentage of records in a file
that are processed in a run. See VOLATILITY.
Ada 3
activity loading In data processing, a method of storing records on
a file in which the most frequently processed records can be
located most readily. See RECORD.
activity ratio In data processing, the ratio of the number of
records in a file that are in use to the total number of records in
that file. See RECORD.
ACTSU US Association of Computer Time Sharing Users.
aetual data transfer rate In data com munications, the average
number of bits, characters or blocks, per unit time, trans ferred
from a data source and received by a data sink. See DATA SINK,
SOURCE.
aetuator In hardware, a device that is capable of mechanical action
under the control of a signal. See INDUSTRIAL ROBOTS.
ACU See AUTOMATIC CALLING UNIT.
Ada A trademark, in programming, a language named after Augusta
Ada, Coun tess of Lovelace who assisted Babbage.
(a) Main applications. Ada was developed for use in computer
control and communica tion systems, where instruments or systems
are monitored or governed by a program. Typical uses include
factory production lines, data recording in laboratories, naviga
ti on al systems, networking and interfacing of multiple
processors. Ada is aimed at installations with a long lifespan
where software modification and maintenance are of major
concern.
(b) Originated in the US Department of Defense in the late
1970s.
(c) Facilities. Ada is essentially derived from Pascal. This entry
will therefore con centrate upon those additional features that
are unique to it. Heavy emphasis is placed upon program
readability; it is considered that it is more important for the
program to be easily understood than rapidly written. In many
applications the cost of maintenance can be an order of magnitude
greater than that of the original production. The langu age is
therefore rigidly structured and data typing is strongly enforced;
even simple operations require a substantial block of code. Machine
independence is assisted by the modular nature of Ada, ensuring
that
4 ADAPSO
implementation details are confined to a few specialized interface
packages.
An unusual feature of this language lies in its provision for
programmable exception handling. Whereas most languages assurne the
correctness of finished programs, and exceptional conditions are
detected and handled by the operating system, Ada recognizes that
the nature of realtime opera tion, hardware communication and
large sections of code render total reliability unattainable, and
so allows for the software detection and correction of errors
without recourse to human intervention. In many control
applications, e.g. those involving heavy machinery, it is vital
that the advent of a run time fault does not cause the software to
cease to operate.
Concurrent programming is supported to permit the control of
parallel processes, through the declaration of tasks whose
execution is interleaved. Tasks may co m municate, usually by
means of a rendezvous, and nondeterministic operation is permitted.
Ada programs may not be necessarily sup ported by an operating
system in the host computer, and they must therefore take command
of the necessary input/output and housekeeping operations. Hence a
special ized support environment for developing and testing code
has been investigated for the language.
In Ada the major units of program struc ture are the package and
the procedure. Packages may be separately compiled but where this
is done the package must come in two parts: the specification part,
which defines the package interface; and the body or implementation
part. The purpose of the separation is to permit the compiler to
check for valid interface usage in the same way as languages such
as Pascal do within a single compilation unit. The package
constructed thus assists in the correct implementation of very
large programs and provides a frame work for the creation of
abstract data types.
(d) Syntax. Ada is a block-structured language that centers upon
the nesting of statement sequences, successive statements being
separated by semicolons. A substan tial amount of data and control
declarations are required by the compiler, and, as for Pascal, the
precise rules are described by a complex set of graphie syntax
diagrams.
(e) Sampie program. The following com-
plete program declares two integers, reads them as input and prints
out the largest, having invoked the package TEXTIO to perform
input/output operations. with TEXT 10; use TEXT 10; procedure DEMO
is pragma main; P,Q: integer range 0 .. 100; begin GET(P);GET(Q);
NEW UNE; if P>Q then PUT(P); else PUT(Q); end if; end
DEMO;
(f) Comments. Ada is a relatively new language whose real worth may
not become apparent for some years . While it has many advantages
in the programming of control computers and other real time
activities, with associated ease of software maintenance, it is
highly verbose and unlikely to compete with the established
block-structured langu ages for more mundane tasks.
(g) Similar to Pascal, C, PLil. see ABSTRACT DATA TYPE, BLOCK
STRUCTURE, C, COMPILER, CONCURRENT PROGRAMMING, DATA ACQUISITION,
DATA TYPE, DECLARATION, HIS TORY OF COMPUTING, HOST COMPUTER,
HOUSE KEEPING, NEST, PARALLEL PROCESSING, PASCAL, PLI, REALTIME,
RENDEZVOUS, SOFT WARE MAINTENANCE, SYNTAX.
ADAPSO US and Canada Association of Data Processing Service
Organizations.
adaptive channel allocation In communica tions, a method of
multiplexing where chan nels are allocated according to demand
rather than on a fixed, predetermined plan. See FREQUENCY DIVISION
MULTIPLE ACCESS, MULTIPLEXING, TIME DIVISION MULTIPLE ACCESS.
adaptive routing In data communications, a routing scheme for
packets or messages in which the behavior adapts to network changes
such as line failures or variation of the traffic pattern. See
MESSAGE SWITCHING, PACKET SWITCHING, ROUTING.
adaptive systems Systems that display the ability to learn to
change, alter their state or otherwise react to a stimulus.
ADC See ANALOG TO DIGITAL CONVERTER.
ADCCP See ADVANCED DATA COMMUNICA
TIONS CONTROL PROCEDURE.
VERTER.
addend In mathematics, the operand of the addition operation, the
number added to the augend to form a sumo See AUGEND,
OPERAND.
adder In architecture, a device that forms an output resulting from
the sum of Mo or more numbers presented as inputs. See FULL
ADDER, HALF ADDER.
add-in In hardware, an expansion card that slots into a
microcomputer to provide ad ditional facilities. This is a very
simple method of enhancing a microcomputer. The boards available
allow for additional RAM, additional operating facilities,
particularly CP/M, enhanced graphics, modems, instru mentation
etc. See CP/M, EXPANSION CARD,
MODEM, RAM.
address In programming, (1) a character or group of characters that
identifies a register, a particular part of storage, or some other
data source or destination, (2) to refer to a device or an item of
data by its address. (3) In communications, the part of the
selection signals that indicates the destination of a call. (4) In
word processing, the location, identified by an address code, of a
specific section of the recording medium or storage.
addressability In computer graphics, the number of addressable
points within a speci fied display space or image space. See
DISPLAY SPACE, PICTURE ELEMENT.
address bus In buses, a unidirectional bus over wh ich digital
information is transmitted to identify either a particular memory
loca tion or a particular input/output device. Compare CONTROL
BUS, DATA BUS. See BUS,
INPUT/OUTPUT DEVICE, MICROCOMPUTER.
addressing mode 5
address fleld In programming, the specific portion of a computer
word that contains either the address of the operand or the
information necessary to derive that ad dress. See OPERAND,
WORD.
address format In programming, the arrangement of the parts of a
simple address, such as those required for identify ing a channel,
module or track on a magnetic disko See ADDRESS, MAGNETIC DISK,
TRACK.
addressing (1) In programming, the assign ment of addresses to the
instructions of a program. (2) In communications, the means whereby
the originator or control station selects the unit to which it is
going to send a message. See STATION.
addressing mode In programming, the method of specifying the
location of data during the execution of a machine code
instruction. At the simplest level the operand may contain the
actual address of the data or, in the immediate mode, the da ta
itself.
Locating the absolute address in the operand, however, is
uneconomic, inflexible and inefficient for many operations. Un
economic because with al-Mbyte address space each operand of each
instruction would need to be 20 bits long. Inflexible because there
would be no straightforward method of using the same instruction to
access successive locations in successive in vocations, e.g. in a
loop performing a table lookup. Inefficient because every data
operation would involve transfer of data between main memory and
the CPU. The major dis advantage and restriction to this mode,
however, is that the program can only be executed from one position
in memory.
With relative addressing the operand con tains a quantity that
represents a displace ment from the current value of the instruc
tion pointer. This is often used in jump instructions and has both
the advantage that the displacement requires fewer bits than the
absolute address and that the instruction will function correctly
if the program is relo cated.
Register addressing is employed to en hance the power of
instruction sets. Register addresses may be implicit with the
instruc-
6 address modification
Installed .&.ddress rnerTIOfy
AAMoo rnolher· 000'"
RAMon molher· _d
address spaces in an IBM-pe, XT and AT
tion or a particular register may be specified with a few bits of
the instruction. Data moves from one register to another are fast
and efficient because they require no traffic on the data bus. More
importantly, how ever, register addressing provides both high
flexibility in programming and facilitates memory management.
Indexed addressing provides a data address by combining the
contents of an index register with the fixed value of an operand;
this technique is essen tial to update addresses in loop
operations.
Indirect addressing implies that the memory location accessed
contains not the required information but the address of that
information; this technique provides for considerable flexibility
in programming be cause the location of the requisite informa
tion can be changed during program execu tion. 16-bit
microcomputers provide for very sophisticated addressing modes; the
Intel 8088/8086 has 25 such modes. A range of registers is employed
to facilitate seg-
mentation, program relocation and the accessing of sophisticated
data structures. For example, an address in an 8088/8086
instruction may involve the contents of segment register, a base
register, an index register and a displacement. See ABSOLUTE
ADDRESS, ADDRESS, BIT, BYTE, CPU, DATA BUS, DATA STRUCTURE,
IMMEDIATE ADDRESS, INDEXED ADDRESS, INDEX REGISTER, INDIRECT
ADDRESS, INSTRUCTION COUNTER, LOOKUP TABLE, LOOP, MACHINE CODE
INSTRUCTION, MEMORY MANAGEMENT, MEMORY ORGANIZA TION.
address modification In programming, an action that causes an
address to be alte red in a prescribed way by an arithmetic,
syntactic or logic operation. See ADDRESSING MODE.
address register In architecture, a special register used by the
CPU to store the address of data to be fetched from, or stored in,
the computer memory. See CPU, REGISTER.
address space In architecture, the nu mb er of memory cells that
may be accessed by the CPU. If an 8-bit processor employs a 2-byte
address then the address space is 65,536 cells, and each cell
normally holds 1 byte. With 16-bit microprocessors the address bus
normally has 20 lines giving an address space in excess of 1 Mbyte.
Bank-switching tech niques can be employed to reduce the
disadvantages of a sm all address space. See BANK SWITCHING, BYTE,
CPU, 8-BIT MICRO PROCESSOR, I6-BIT MICROPROCESSOR.
address track In backing storage, a track on a magnetic disk
containing the addresses of files, records etc. stored on other
tracks of the same device. See MAGNETIC DISK, TRACK.
add time In performance, the time re quired by a particular CPU to
add two multidigit numbers, not including the time taken to read
the numbers or store the result. Microcomputers are often rated by
comparing add times as a criterion of their relative speed. See
MICROCOMPUTER.
ADIS Automatic System.
Data Interchange
adjacency In character recognition, a con dition in which the
character spacing refer ence lines, of two consecutively printed
characters on the same line, are separated by less than a specified
distance.
adjacent channel In communications, the next channel, or the one in
elose proximity, either physically or electrically, to the one in
current use. See CHANNEL.
adjust In word processing, an editing feature in which the system
automatically adjusts the right-hand margin for insertion or
deletion of copy during playback. Word, and sometimes page,
wraparound is auto matically performed as needed. See WRAP
AROUND.
ADP See AUTOMATIC DATA PROCESSING.
ADSR In man-machine interfaces, Attack Decay Sustain Release; the
shape of the envelope that modulates the amplitude of a periodic
waveform such as a sawtooth or sine wave. The sound generator chip
of a microcomputer can be programmed to pro duce music by varying
the ADSR and the modulating waveform. See MODULATION,
MUSIC SYNTHESIZER, SAWTOOTH, SINE WAVE.
~ ~.l!- Sustain ~ . " 0.6 Decay
ADSR Typical amplitude envelope shape for a musical note.
advanced data communications control pro cedure In data
communications, pertaining to the operation of a data link using an
advanced (SDLC, HDLC) protocol. See HDLC, PROTOCOL, SDLC.
advanced data link control In data com munications, a link
protocol used in HDLC and SDLC systems. See HDLC, SDLC.
AECT US Association for Educational Communication and
Technology.
air gap 7
affine c10sure In mathematics, a repre sentation of infinity in
which the positive and negative infinities are considered to be at
opposite 'ends' of a line of all real numbers. Compare PROJECTIVE
CLOSURE. See INFINITY, REAL NUMBER.
affirmative acknowledgement In data communications, the replies
ACKO and ACK1 in binary synchronous transmission indicate that the
previous transmission block was accepted by the receiver and that
it is ready to accept the next block. ACKO and ACK1 sent altemately
provide sequential checking for aseries of replies. ACKO is also
used as an affirmative reply to station se1ection signal in a
multidrop circuit, or to an initialization sequence in a point to
point operation. See BINARY SYNCHRONOUS COM
MUNICATIONS, MULTIDROP CIRCUIT, POINT TO POINT.
AFIPS American Federation of Informa tion Processing Inc. A
federation founded in 1961 which ineludes the American Society for
Information Science, American Statis tical Association,
Association for Comput ing Machinery, Association for Education
Data Systems, Data Processing Manage ment Association, IEEE
Computer Society, Instrument Society of America, Society for
Computer Simulation, Society for Industrial and Applied
Mathematics, Society for In formation Display. It is the US member
of IFIP. See IFIP.
AFNOR Association Fran~aise de NOR malisation - the French
standards organisa tion.
afterglow Synonymous with PERSISTENCE.
after image In databases, a copy of a record after it has been
modified by a user, or program. If there is a system failure the
after images can be used to update the database from a previous
dump. Compare BEFORE IMAGE. See DUMP, TRANSACTION .
AI See ARTIFICIAL INTELLIGENCE.
air gap In backing storage, the very narrow gap between the two
elements of a magnetic recording or playback head. See HEAD.
8 alarm
alarm In operations, a visual or audio signal to signify that an
error has arisen or an abnormal situation exists.
ALGOL In programming, ALGOrithmic Language, an early
block-structured langu age providlng many elegant features that
were lacking in other early high-level langu ages. It has been
largely superseded by Pascal. See HIGH-LEVEL LANGUAGE, PASCAL,
PROGRAMMING.
algorithm In progamming, a finite set of well-defined rules for the
solution of a problem in a finite number of steps; for example, a
precise description of the steps involved in determining the record
with the highest value of a specified numerical attri bute. See
ATIRIBUTE, PROGRAMMING , RECORD.
algorithmic language A computer language designed for expressing
algorithms. See ALGOL.
algorithmic pattern generation In memory systems, a method of
testing RAM in which a binary pattern is written and the RAM is
subsequently read to verify that it was written correctly, and that
the writing into one area of RAM did not affect the data stored in
another section. See FIXED PATIERN TESTING, GALLOPING PATIERN
TESTING, RAM.
alias In operating systems, a Unix shell command that enables the
user to set up another name for a Unix program, e.g. the user could
replace 'ls' with the more mnem onic term 'list directory'. See
LS, SHELL, UNIX.
aliasing (1) An effect that occurs when a signal is sampled at a
rate less than twice the highest frequency present in it. When a
subsequent signal is recovered from the sampies it will not contain
the high-frequ ency component of the original signal, and it will
instead display a false low-frequency signal. See NYQUIST SAMPLING
THEOREM. (2) In computer graphics, an effect that occurs when a
computer attempts to handle detail of a diagram which exceeds the
basic resolving power of the system, e.g. the staircase effect
produced when a
low-resolution computer graphic system attempts to displaya
diagonalline. Compare ANTI-ALIASING.
aligning edge In optical character recogni tion, the edge of a
form which, in conjunc tion with the leading edge, serves to
position correctly the document that is to be scanned. See
SCAN.
alignment pin In electronics, any pin or device that will ensure
the correct mating of two components designed to be
connected.
allocate In operating systems, (1) to assign aresource, such as a
disk or a diskette file , to a specific task. See FILE, MAGNETIC
D1SK, TASK. (2) To assign memory allocations to main routines and
subroutines. See SUB-ROUTINE.
allophone In man-machine interfaces, a manifestation of a phoneme
in a speech signal. A phoneme may be acoustically different
depending upon word position and an allophone is a positional
variant of the same phoneme. See PHONEME, SPEECH SYNTHESIZER.
Aloha In computer networks, a packet switched system at the
University of Hawaii which uses radio broadcast techniques.
alphabet (1) An ordered set of all the letters and associated marks
used in a language or work. (2) An ordered set of letters used in a
code language, e.g. the 128 characters of the ASCII alphabet. See
ASCII.
alpha beta technique In artificial intelli gence, a technique used
in game-playing routines to determine the best set of moves for a
given player. The player will pick the set of moves to maximize his
score while the adversary will always attempt to select moves that
will minimize his losses. The successive set of moves can be
represented by a tree structure, one player having the choice of
branches from one level and the adversary the choice at the next
level. The alpha beta technique eliminates subtrees to be searched
from the tree, on the basis that both players using optimal
strategies would never employ such subtrees, thus reducing
the effort of searching for optimum moves. See TREE
STRUCTURE.
alphabetic character set A character set that contains letters but
not digits. The set may contain control characters, special
characters and the space character. Compare ALPHANUMERIC. See
CONTROL CHARACTER.
a1phabetic shirt In peripherals, a control for selecting the
alphabetic character set in an alphanumeric keyboard printer.
alphabetic string A character string con sisting of letters from
the same alphabet. See STRING.
a1phageometric In computer graphics, a standard in which the codes
can instruct the terminal to produce line drawings, fill areas with
color etc. in addition to normal charac ter display modes. Compare
ALPHAMOSAIC. See NAPLPS, VIDEOTEX.
alphamosaic In computer graphics, a standard in which the codes
determine the alphanumeric character or mosaic pattern to be
displayed in a character space. Compare ALPHAGEOMETRIC. See
ALPHANUMERIC, CHAR ACTER SPACE, VIDEOTEX.
a1phanumeric Pertaining to a character set that contains letters,
digits and usually other characters, e.g. punctuation marks.
a1phanumeric character set A character set that contains both
letters and digits and may contain control characters, special
charac ters and the space character. See ALPHA NUMERIC.
a1phanumeric data Data represented by letters and digits and
perhaps special charac ters and the space character. See ALPHANU
MERIC.
a1phanumeric display device Synony- mous with CHARACTER DISPLAY
DEVICE.
a1phanumeric keyboard In videotex, a keyboard used for ente ring
letters, numbers and special characters . It is required by IPs for
creating frames. Users with this key board can send messages via
electronic mail or fill response frames with alphanumeric
Alvey 9
information. Users who are only equipped with numeric keypads enter
such informa tion in a format determined by a predefined menu
selection. See ALPHANUMERIC, ELEC TRONIC MAlL, FRAME, IP, KEYPAD,
RESPONSE FRAME.
alphanumerics mode In videotex, the dis play mode in which the
display characters are those of the alphanumerics set. Compare
GRAPHICS MODE. See DISPLAY MODE.
alphanumerics set In videotex, the set of 96 display characters
comprising all the alpha numerics characters. See DISPLAY CHAR
ACTER.
alphaphotographic In videotex, a method of displaying alphanumeric
characters and picture quality graphics from individually
transmitted and stored picture elements. See PICTURE PRESTEL.
alpha testing In programming, the in house testing of a package by
a software house prior to beta testing. There are a wide variety of
conditions under which a software package is used, and the testing
phase of product development can be as much as three times as
expensive as the development phase. Compare BETA TESTING. See SOFT
WARE HOUSE.
Altair bus Synonymous with SIOO.
alternate route In communications, a secondary or backup route that
is used if normal routing is not possible.
alternate track In backing storage, a track on a magnetic disk or
other storage device, which is automatically substituted for a
damaged track. See MAGNETIC DISK, TRACK.
alternating current In electronics, electric power supply in the
form of a sine wave, normally a frequency of 60 Hz in the USA, and
50 Hz in the UK. Compare DIRECT CURRENT. See Hz.
ALU See ARITHMETIC LOGIC UNIT.
Alvey A programme, named after Mr John Alvey, of pre-competitive
research in ad-
10 AM
vanced information technology costing some BOO million over five
years. See ESPRIT.
AM See AMPLITUDE MODULATION.
ambient noise level In electronics, random, uncontrollable and
irreducible noise level at a location or circuit. See NOISE.
American National Standards Institute A body that organizes
committees formed of computer users, manufacturers etc., to develop
and publish industry standards; e.g. ANSI FORTRAN, ANSI Standard
Code for Periodical Identification.
ampere In electronics, the basic unit of electrical current.
Compare VOLT. See CURRENT.
amplification In electronics, (1) the strengthening of a weak
signal, (2) the ratio between so me measure of the output signal
and the input signal of a device. Compare ATTENUATION.
amplifier In electronics, a normally unidi rectional device that
increases the power or amplitude of an electrical signal. See
AMPLI TUDE.
amplitude In electronics, the magnitude of the greatest deviation
from the midpoint value of a periodic signal or phenomenon. See
FREQUENCY, WAVELENGTH.
amplitude modulation In communications, a form of modulation in
which the amplitude of the carrier signal is varied in accordance
with the amplitude of the modulating signal. Compare FREQUENCY
MODULATION, PHASE
MODULATION. See CARRIER, MODULATION.
analog In computing and communications, pertaining to the form of
continuously vari able physical quantities. Compare DIGITAL.
See ANALOG SIGNAL, TRANSDUCER.
analog channel In communications, a data channel in which the
amplitude of the signal transmitted can take any value between the
limits defined for the channel. Voice-grade channels are analog
channels. See ANALOG.
analog computer A device that performs mathematical functions on
continuous vari ables, usually voltages, and produces a solution
in the form of an analog signal. They were used extensively in the
study of dynamic systems and as simulators. The basic building
block of the analog computer was the operation al amplifier, used
to pro duce the mathematical functions of addition, subtraction,
integration and multiplication. Non-linear effects, e.g.
saturation, were also effected with special diode units. Unlike
digital computers they were parallel in operation, easy to program
for their dass of problems and relatively fast. They did not
benefit, however, from the rapid advances in digital technology,
and they are now obso lete except for special purpose
applications. See DIGITAL COMPUTER, HYBRID COMPUTER,
OPERATIONAL AMPLIFIER, SIMULATION.
analog flux copier Synonymous with BIT
COPIER.
analog signal A signal that varies contin uously according to the
information in trans mission, e.g. sound waves. Compare
DIGITAL
SIGNAL.
analog to digital converter In control and instrumentation, a
device that accepts a continuous analog signal and produces a
stream of digital signals corresponding to values of the analog
signal at sampling instants. The speed of conversion of the sampled
signal may vary from less than 1 microsecond to up to 100
milliseconds. The converter may operate by successive ap
proximation or integration.
The successive approximation method re quires aperiod of 1 to 25
microseconds; the converter successively approximates the digital
bits until an internally genera ted digital to analog converter
output equals the analog input. The analog input must be held
constant during the conversion period by a sampie and hold device.
This type of analog to digital converter is used for high sampling
rates where the noise rejection is not a major problem.
The integration converter is used when the analog signal is subject
to noise and high
sampling rates (i.e. greater than 1000 read ings per second) are
not required. The analog to digital conversion integrates the
analog signalover aperiod of several milli seconds, thus averaging
out the effects of noise. Compare DIGITAL TO ANALOG CON VERTER.
See COMPUTERIZED INSTRUMENTA TION, SAMPLE AND HOLD CIRCUIT.
analog transmission In communications, the transmission of
information by analog signal. See ANALOG SIGNAL.
analysis The methodical investigation of a problem, and the
separation of the problem into smaller related units for further
detailed study. Compare SYNTHESIS. See SYSTEMS ANALYSIS.
analyst A person who defines problems and develops algorithms and
procedures for their solution. See SYSTEMS ANALYSIS.
ancillary equipment In communications, equipment located in a
subscriber's pre mises, e.g. answering devices, automatic dialers,
to provide a greater utility of a communications channel for
individual sub scribers.
AND A logical operation, A AND B has the result true only if both
of the logical variables A and Bare true. The correspond ing truth
table is:
A B AANDB 000 100 o 1 0 1 1 1
Compare OR. See BOOLEAN ALGEBRA, TRUTH TABLE.
AND gate In electronics, a logic unit that produces an output
signal that is the logical AND of the input signals . Compare OR
GATE. See AND.
angstrom A unit of measurement equal to 10 to the power of minus 10
meters, i.e. one ten-millionth of a millimeter. Commonly used in
the measurement of wavelengths of light.
animation In video and computer graphics, the projection of a
continuous sequence of related images, at a speed that matches
the
ANSI-SPARC 11
human eye's inherent persistence of vision, to create a
flicker-free image. At the average cinematic screen illumination
the human eye detects no flicker for frequencies of inter mission
above apl?roximately 16 per second. American televIsion generates
images at 30 frames per second, while the European standard is 25
frames per second. Conflicts arise in the television and video
recording of films employing 24 frames per second. American
television uses three two puH down. British television transmits
the film at 25 frames per second giving a speed increase of 4
percent and a rise in the sound pitch. See PERSISTENCE OF VISION,
THREE TWO PULL DOWN.
anisochronous signal In electronics, a sig nal that is not related
to any dock and in wh ich transitions could occur at any instant.
See ANISOCHRONOUS TRANSMISSION, CLOCK.
anisochronous transmission In data com munications, a form of data
transmission, similar to asynchronous transmission, but in which
there can be variable time intervals between the bits within a
character as weil as between the characters themselves. Com pare
ISOCHRONOUS TRANSMISSION. See ASYNCHRONOUS TRANSMISSION, BIT.
A/N See ALPHANUMERIC.
annotation symbol In programming, a symbol used to add messages or
notes to a flowchart. See FLOWCHART.
anode In electronics, the positive terminal of a device. Compare
CATHODE.
ANS I See AMERICAN NATIONAL STANDARDS INSTITUTE.
ANSI MCll-1975 In buses, another designation for GPIB. See ANSI,
GPIB.
ANSI-SPARC In databases, pertaining to schemata developed as a
general standard for the description of database management
systems. The standard defines three schemata: (a) a conceptual
schema which provides a logical description of the data; (b) one or
more user schemata which represent the users' views of the data;
and (c) a physical schema which describes the mann er in which the
data is stored in the computer. See SCHEMA.
12 ANS I X3H3
ANSI X3H3 In computer graphics, the ANSI Technical Committee on
Computing Graphics Programming Languages. See ANSI.
answer back In data communications, a signal sent by a receiving
unit to the sending station for identification or to indicate it is
ready for transmission. See VOlCE ANSWER
BACK.
answering In data communications, the process of responding to a
calling station to complete the establishment of a connection
between data stations. See STATION.
answering time The time between the appearance of a signal and the
response made to it.
anti-aliasing In computer graphics, a method of disguising the
aliasing errors introduced by low-resolution graphics dis plays.
The jagged edges produced by boun dary pixels may be softened by
adjusting the shading intensities to create a smoother transition
of color changes. Compare ALiASING. See PIXEL.
anticipatory staging In operating systems, a technique in which
blocks of data are moved from one storage device to another, with a
shorter access time, in anticipation that they will be required by
the program and before the program actually requests them. Compare
DEMAND STAGING.
anticoincidence circuit In hardware, a logic circuit that only
provldes an output if dif ferent signals are received on the input
lines. Compare COINCIDENCE CIRCUIT.
Antiope The French standard for character coding and display for
videotex terminals.
APA In computer graphics, All Points Addressable. Synonymous with
PIXEL.
APD See AVALANCHE PHOTODIODE.
aperture In architecture, apart of a mask that permits retention of
the corresponding portions of data. See MASK.
APL In programming, A Programming Language.
(a) Main applications. APL is an unusual language in that it began
life as a functional notation to express mathematical algorithms
and was subsequently adapted for use as a programming language. It
is extremely con cise and has found favour mainly in mathe
matical applications, although it has also been used for tasks such
as data retrieval, teaching and simulation. As an interactive, and
interpreted, language it is more suited to one-off jobs and
experimental tasks than large program construction.
(b) Originated from IBM in the late 1960s from Iverson's
notation.
(c) Facilities. APL allows the expression of complex mathematical
functions in a very compact manner, employing a special char acter
set to this end. Whereas most langu ages can only operate directly
upon single variables, APL handles complete data struc tures
(arrays) with its primitive operations and thus offers a great
saving of effort in many situations. Indeed, the whole language is
based around the use of numeric and character arrays alone,
providing simplicity at the cost of more elegant data
structures.
APL provides a powerful range of mathe matical and logical array
functions, gener ator operations wh ich permit sophisticated array
manipulation and rather limited input/ output facilities; its
mathematical derivation gives rise to certain peculiarities such as
right to left evaluation of expressions. The pro gram constructs
are rather weak; they are centered on the GO TO statement, but
provision is made for recursive subprogram calls with parameter
passing and local vari ables.
The interactive operation of APL is one of its most important
features, the language is well suited to the rapid development of
small programs for experimentation, or single runs of data
processing. The user is allocated a temporary workspace for the
creation of programs and variable storage, wh ich can be loaded
from, or stored into, a system library. The user is also afforded
intimate control of code execution within this environment, to the
extent that there is no real concept of a main program but rather a
collection of subprograms which can be called into action,
interrupted and rearanged at will.
(d) Syntax. The syntax is straightforward and homogeneous. A piece
of code com prises a set of independent subprograms,
each composed 01 statements numbered by their order of execution. A
statement may only be a single expression, although the variety is
somewhat bewildering due to the many operators available.
(e) Typical program . An APL program for operations on the rows of
a two dimensional array would have more the appearance of a complex
mathematical expression than a statement of computer
instructions.
(f) Comments. APL may perhaps be better described as a programming
environ ment, than a language, since it is ill equipped for the
production of finished programs intended for intensive use. The
dense code allows good execution speeds but also permits the
construction of single state ments that are alm ost unintelligible
even to the highly experienced programmer. The impoverished program
and data structures also restrict its range of applications, where
as the requirement of a special character set restricts its use on
many computers. See ARRAY, DATA STRUCTURE, GO Ta, INTER PRETER,
LOCAL VARIABLE, PARAMETER PASS ING, RECURSIVE ROUTINE, SYNTAX,
SYSTEM LIBRARY, VECTOR, WORKSPACE. application interface - In
distributed pro cessing, a routine that enables a program in a
microcomputer to communicate with the microprocessor on an emulator
expansion card. This routine could be a sequence of operator
commands for automating file up loading to the host. See EXPANSION
CARD, HOST COMPUTER, SCRIPT, TERMINAL EMULA TION , UPLOAD.
application layer In data communications, the topmost layer in the
ISO Open Systems Interconnections model. The content of this layer
is left to the users, and it is expected that standard protocols
for specific indus tries will be developed. Cornpare DATA LINK
LAYER, NETWORK LAYER, PHYSICAL LAYER, PRESENTATION LAYER, SESSION
LAYER, TRANS PORT LAYER. See OPEN SYSTEMS INTERCON NECTION.
application-oriented language In pro gramming, a language that has
facilities or notations useful for solving problems in one or more
specific classes of applications, e.g. numerical analysis, business
data processing, simulation. application program In programming, a
program, usually written in-house, for a
architecture 13
specific user application, e.g. payroll. Corn pare OPERATING
SYSTEM, SOFTWARE PACKAGE, SOFTWARE TOOL. See IN-HOUSE.
architecture The specification of the re lationships between parts
of a computer system. All generations of computers prior to the
fifth generation were based upon Von Neumann architecture. This
architecture was the turning point in computer design because it
incorporated the concept of stored pro gram control, thus
decoupling the hardware design from the detailed consider ation of
algorithms relating to applications and providing the incredible
flexibility of modern computers. The traditional Von Neumann
computer contains: (a) a single computing element incorporating a
proces sor, communications and memory; (b) a linear organization
of fixed-sized memory cells; (c) a one-Ievel address space of
memory cells; (d) a low-Ievel machine language; (e) sequential
centralized control of computation; and (f) a primitive input/
output capability.
The Von Neumann architecture was the keystone of computer
development for over 40 years, but it was designed in an era when
electronic devices were expensive, large, relatively unreliable and
dissipated consider able thermal power. Moreover the computer
applications in the 1950s and 1960s were relatively straightforward
and amenable to design by conventional high-Ievellanguages. The
situation changed during the late 1970s; the software crisis can
be, at least in part, attributed to the fact that relatively simple
architectures can only be induced to perform complex processes by
the design of sophis ticated software. Moreover the massive
advances in microelectronics have provided architecture designers
with powerful, cheap, fast processors and memories. The wheel has
thus undergone a complete revolution, and it is now more economical
to design hardware modules for classes of problems.
The perceived shortcomings of current architectures are given
below.
(a) They are not equipped with the necessary functions to process
non-numeric data: sentences, symbols, speech, graphics and images.
These function as are essential for user friendly man-machine
interfaces.
(b) They are unable to provide the level of performance demanded by
a wide range of
14 archiving
and difficult to implement. (d) System performance with
conven
tional architectures appears to be elose to a plateau.
(e) It is not elear that VLSI technology can be efficiently
exploited with multimicro computer architecture.
Von Neumann architectures operate upon the cyele: control unit
sends one instruction at a time to the ALU, the ALU performs one
instruction at a time on a single piece of data extracted from
memory for that pur pose. This mechanism is very inefficient for
modern applications, e.g. controlling the multifarious aspects of a
sophisticated robot, analyzing images from TV cameras. Parallel
processing is essential for these and many other modern areas of
computer application. Attempts to develop parallel processor
architecture have been made, e.g. ICL distributed array processor
and London University College CLIP-4 machine. With the
ever-decreasing cost of processors, multiprocessor architectures
providing true parallelism, i.e. operating on multiple in
struction streams and multiple data streams, with possibly hundreds
or more processors, can be expected. Such machines could pro vide
the high degree of redundancy essential for realtime
processing.
New forms of architecture have moved away from the control flow
characteristics of Von Neumann machines. Thus in data flow
architecture, instructions are actuated when the required input
data is available and are weIl geared to parallel processing. Dis
tributed function architectures are designed with modular
facilities allowing specialized machines to perform their specific
functions, e.g. database management, computation, man-machine
interface operations, network control, with high efficiency. The
fifth gen eration of computers will necessarily em body novel
architecture to achieve their objective. See ALU, DATABASE
MANAGE
MENT SYSTEM, DATA FLOW, FIFTH-GENERATION
COMPUTER, HISTORY OF COMPUTING, INDUS
TRIAL ROBOTS, LOW-LEVEL LANGUAGE, MAN
MACHINE INTERFACE, MULTIPROCESSING,
PARALLEL PROCESSING, PROCESSOR, VLSI.
archiving In operations, the storage of backup files and associated
journals, usually
for a given period of time. See FILE, JOUR
NAL.
ARDIS US Army Research and Develop ment Information System.
area In databases, the CODASYL defi nition of an area is a named
subdivision of the addressable storage space in the data base
which may contain occurrences of records and sets, or parts of sets
of various types. See CODASYL, RECORD, SET.
area infill In computer graphics, a tech nique used with
pixel-based systems in which a defined area of the screen is filled
with a specified color or pattern. See PIXEL.
areal density In backing storage, the num ber of bits per unit
area that can be stored on a recording device. In magnetic and
optical disk systems, it is equal to the product of bits per inch
and tracks per inch. See BITS PER
INCH, MAGNETIC DISK, OPTICAL DISK STORAGE,
TRACKS PER INCH.
argument (1) Any value of an independent variable. (2) In
programming, a parameter passed between a calling program and a
subprogram or statement function. See SUB ROUTINE.
arithmetic (1) The branch of mathematics concerned with the study
of the positive real numbers and zero. (2) In programming, the
operations of addition, subtraction, multi plication and
division.
arithmetic capability In word processing, the ability of a system
to be used as a calculator or adding machine.
arithmetic instruction In programming, an instruction in which the
operation part specifies an operation that follows the rules of
arithmetic. See INSTRUCTION.
arithmetic logic unit In architecture, the unit in which
arithmetic, logic and related operations are performed. See CENTRAL
PRO
CESSING UNIT.
arithmetic mean In mathematics, the aver age value of a number of
values of a variable. It is calculated by summing all the
component values and dividing the result by the number of
values.
arithmetic overflow See OVERFLOW.
arithmetic shift In programming, (1) a shift that does not affect
the sign position, (2) a shift that is equivalent to the
multiplication of a number by a positive or negative integral power
of the radix. Compare CIRCU LAR SHIFI, LOGICAL SHlFI. See RADIX,
SHIFI.
arithmetic unit See ARITHMETIC LOGIC UNIT.
arm stealing In backing storage, the rapid movement of read/write
heads, in a multi user environment, when the system is re
sponding to a succession of enquiries from different users. See
MAGNETIC DISK, READ/ WRITE HEAD.
ASCII 15
decay envelope. See ANALOG SIGNAL, MUSIC SYNTHESIZER.
artificial intelIigence (1) The science of making computers do
things that would require intelligence if done by man. (2) That
branch of computer science that studies how to make computers
smarter. The term has been employed to cover a wide variety of
computer developments, e.g. efficient repre sentation of
knowledge, reasoning, deduc tion, problem solving and heuristic
search. These capabilities have been applied in game playing,
automatic theorem proving, automatic computer programming, robots,
machine vision, natural language systems and information
processing. See EXPERT SYSTEMS, MACHINE TRANSLATION.
ARPA Advanced Research ARU Audio Response Unit. See AUDIO
Projects RESPONSE TERMINAL. Agency. See D ARP A.
ARQ See AUTOMATIC RETRANSMISSION RE QUEST.
array In da ta structures, an ordered arrangement or pattern of
items or numbers, e.g. table of numbers. The use of arrays can
substantially reduce programming effort in the processing of
repetitive operations. For example, if the same operation is to be
performed on variables Al, A2, ... , AlOO, then the redefinition of
these variables into an array A with 100 members enables the
process to be defined as a loop operation. See LOOP, VECTOR.
array processor In architecture, a com puter system that is
designed to perform identical operations, on elements of an array,
in parallel. It may be a self-contained unit or attached to a
mainframe computer via an internal bus or input/output port. See
ARRAY, INPUT/OUTPUT PORT, PARALLEL PRO CESSING.
articulator In man-machine interfaces, a component of a music
synthesizer that causes the sound to decay according to the
characteristics of a particular instrument. The circuit effectively
multiplies the analog signal of the sound by the approximate
ASA American Standards Association; a body with groups responsible
for the estab lishment of data-processing standards.
ascender In printing, the portion of a lower case character, such
as '1', 'h', etc. that rises above the upper level of letters such
as· '0', 'a' etc. Compare DESCENDER. See LOWER CASE.
ASCII American Standard Code for Infor mation Interchange,
pronounced ASKEE. A standard data transmission code that was
introduced to achieve compatibility between data devices. It
consists of seven information bits and one parity bit for
error-checking purposes, thus allowing 128 code combin ations. Of
these, 32 are used for upper case characters and a few punctuation
marks, another group of 32 characters is used for numbers, spacing
and additional punctua tion symbols, a third group of 32
characters is assigned to lower case characters and some rarely
used punctuation symbols. The last set of 32 characters is
allocated to machine and control commands, e.g. line feed, carriage
return. See BIT, CARRIAGE CONTROL, INTERNATIONAL ALPHABET NUM BER
5, LOWER CASE, PARITY CHECKING, UPPER CASE.
> '" ~ - - C O
N T
R O
L C
H A
R A
C T
E R
ASIS American Society for Information Science.
aspect ratio In television, the ratio of the dimensions of a TV
screen, normally 4:3.
assemble In programming, (1) to translate a source program using an
assembler, (2) to in te grate subroutines into the main pro gram.
See ASSEMBLING, SOURCE PROGRAM,
SUBROUTINE, TRANSLATOR.
assembler In programming, a program that translates a source pro
gram written in a low level language to machine code. Compare
COMPILER, INTERPRETER. See ASSEMBLING,
LOW-LEVEL LANGUAGE, MACHINE CODE,
SOURCE PROGRAM, TRANSLATOR.
assembling In programming, the processes involved in translating a
low-Ievel pro gram into an object code program. The procedure
commences with a source pro gram produced with a text editor. The
source code instruc tions are generally arranged under four
columns: label, operation code, source code and comment. Individual
instructions do not necessarily contain entries in all the c?lumns,
e.g. not all instructions require elther labels, operands or
comments, and in some cases only the operation code entry is
completed. Generally, but not invariably, one source code
instruction will correspond to one machine code instruction. The
excep tions are macros, where one source code instruction is
translated into a predeter mined sequence of machine code instruc
tions, and assembly directives. The latter do not produce any
machine code instructions but simply provide commands to the
assembler, e.g. to reserve storage space for data entries.
The source code file is processed by the assembler to produce the
object code file. The assembler performs this task in two passes.
One of the functions of the assembler is to replace mnemonics in
the source code program by corresponding machine code addresses (in
absolute or relative terms). In scanning through the program,
however, the assembler will often encounter a reference to a
mnemonic be fore it is defined, e.g. in a jump to a label
instruction when the label is attached to an instruction further
down the source code
assembly directive 17
program. The first pass is therefore required to construct the
symbol table of all mnem onic references to data, instructions or
addresses. All internal references, termed local references, can be
handled at the end of the first pass. However external refer
ences, to code in other modules or to library subroutines, must be
handled in a special manner at a later stage . Moreover the
programmer mayaiso specify that certain labels are to be referenced
by external modules. Such references are dec\ared to be public, or
global, and are stored in global reference tables to be accessed in
the later linking stage of producing the executable machine
code.
In the second pass, the assembler can produce the necessary object
code file. Operation code mnemonics are replaced by their machine
code equivalents, the address ing modes for operands are
determined and the symbols for internal references are replaced by
machine code addresses. The assembler will also check the syntax of
the source code instructions and report ' errors. A listing file,
showing the source code, corresponding machine code and error mess
ages, is often genera ted during this pass.
In some cases, the second pass will produce absolute code that can
be loaded into the computer, or microprocessor system, and run. In
other cases, the object code is in the form of relocatable code.
This is the form used when the program is to be run in conjunction
with other programs, or library subroutines, and the operating
system requires the facility to decide upon the location of the
program in main memory. Relocatable code will be incom plete in
the sense that the assembler was unable to provide machine code
addresses to external references. The linker is employed to
complete the process of producing the executable machine code
program. Compare COMPILER, INTERPRETER . See ADDRESSING
MODE, ASSEMBLY D1RECTIVE, COMMENT, EX
TERNAL REFERENCE, INTERNAL REFERENCE,
REFERENCE, MACHINE CODE INSTRUCTION,
CODE, SYMBOL TABLE, SYNTAX.
assembly directive In programming, an in struction in a low-Ievel
language that teils the assembler to perform certain actions,
18 assembly language
e.g. assign memory space to a variable or constant, but does not
generate any machine code instructions. See ASSEMBLING,
MACHINE
CODE INSTRUCTION.
assembly language In programming, a language that allows a
programmer to de velop a machine code program using sym bols and
mnemonics for storage locations and operations. This language
greatly im proves the comprehension of the program and enables
modifications to be more readily incorporated. Compare HIGH-LEVEL
LANGU
AGE. See ASSEMBLING, LOW-LEVEL LANGUAGE,
MACHINE CODE, TRANSLATOR.
assembly Iisting In programming, the printed list produced by the
assembler giving details of any syntax errors, a listing of the
source program and, normally, an associated listing of the
corresponding machine code program. See ASSEMBLING, MACHINE
CODE.
assembly time In programming, the time at wh ich an assembler
translates the source program into the corresponding object code.
See ASSEMBLING, OBJECT CODE, SOURCE PRO
GRAM.
assertion (1) In programming, a boolean expression that is stated
to be true. In proof of program correctness assertions are used to
make statements about program states. See PROOF OF PROGRAM
CORRECTNESS. (2) In artificial intelligence, a hypothesis about the
problem to be solved. The likelihood of an assertion is established
by asking the user questions or, alternatively, rules may be used
to deduce the likelihood from other assertions or stored data.
Compare OBJECT.
See EXPERT SYSTEMS, RULE.
Assign In operating systems, a command commonly employed to
reassign a disk drive. H is employed when applications pro grams
specifically address one disk drive and the user wishes to employ a
different drive. See APPLICATION PROGRAM, D1SK
DRIVE.
assignment statement In programming, a statement that results in
the assignment of a value to a variable. The expression on the
right-hand side of the statement is evalu-
ated, and this value is assigned to the variable named on the
left-hand side. See EXPRESSION, STATEMENT, VARIABLE.
Association for Computing Machinery US professional computer
science organization. Hs objective is to advance all aspects of
information processing and to promote the interchange of such
techniques between computer specialists and users.
associative processor In architecture, a device using associative
storage methods, i.e. data is accessed by reading keys and
comparing their values with those that identify the item sought.
See ASSOCIATIVE
STORAGE.
associative storage In memory systems, a storage device in which
the user identifies data by apart of its conte nt rather than by
its physical location. It provides a fast method of searching for
data with certain keys. The computer system mayaiso rearrange its
storage of data without affecting the user's application programs.
See APPLICATION PRO
GRAM, KEY.
async In data communications, an abbre viation for asynchronous
communication. See ASYNCHRONOUS TRANSMISSION.
asynchronous Pertaining to actions and events that are not
correlated with some reference time. Compare SYNCHRONOUS.
asynchronous access In memory systems, a method of accessing the
dynamic RAM for refresh operation. Requests for refresh access are
genera ted at a fixed rate indepen dent of the state of the
microprocessor. Compare SYNCHRONOUS ACCESS. See DYN
AMIC MOS RAM, REFRESH CONTROL.
asynchronous communications interface adaptor In hardware, a device
performing a similar function to a UART. See UART.
asynchronous computer A computer in which each operation is
initiated as a result of a signal genera ted by the completion of
the previous operation or by the availability of the equipment
required for the next operation. Compare SYNCHRONOUS COM
PUTER.
2 3 S
1 2 3 4 S 7 8 Stop Stop
Mark Mark Mark
asyncbronous transmission
asynchronous transmission In data com munications, a form of data
transmission in which there can be variable time intervals between
characters but the bits within a character are sent with fixed time
intervals. Start and stop elements are used to indicate the
beginning and end of characters. Com pare SYNCHRONOUS
TRANSMISSION.
ATM See AUTOMATED TELLER MECHANISM.
ATN In buses, ATtentioN, a command signal on the GPIB indicating
whether the information on the data bus is to be inter preted as
an interface control message or device-dependent data. See
GPIB.
atom In programming, a value that cannot be decomposed further. In
LlSP-like langu ages, an atom is an arbitrary string of
characters. In languages operating entity attribute structures, an
atom would corre spond to the set of characters assigned to an
entity name or attribute. See ATIRIBUTE,
ENTITY, LlSP.
attacker In data security, an intruder who attempts to decipher the
ciphertext message. See CIPHERTEXT, INTRUDER.
AT&T American Telephone & Telegraph Co.
attended trail printer In word processing, a trail printer that has
no paper-handling facilities and so requires operator interven
tion before and after the printing of each page. See TRAIL
PRINTER.
attention interruption In peripherals, an input/output interruption
caused by a ter minal user pressing an attention key, or its
equivalent. See INTERRUPT, INPUT/OUTPUT.
attention key In peripherals, a key on a terminal that, when
pressed, causes an input/output interruption in the processing
unit. See INPUT/OUTPUT, INTERRUPT.
attenuation In electronics and communica tions, the reduction in
strength of an elec trical signal as it passes through a circuit
or an electromagnetic wave as it propagates through a transmission
medium. Compare AMPLIFJCATION.
attribute In databases, a field that contains information ab out an
entity, e.g. in a per sonnel database 'horne address' would be an
attribute of entity employee. Compare ENTITY, RELATIONSHIP. See
DISPLAY ATIRI BUTE, FJELD.
audio cassette recorder In backing storage, a recorder designed for
use with compact cassettes.
20 audio cassette recorder interface
audio cassette recorder interface In back ing storage, an
interface unit that enables an ordinary cassette tape recorder to
be used as an input/output device for a microcomputer.
audio frequency The frequency of an audible sound wave; for normal
hearing the range of frequencies lies between 20 and 20,000 Hz. See
Hz.
audio inquiry See VOICE ANSWER BACK.
audio response terminal In peripherals, a terminal that receives
spoken information from a computer. Digitized speech in the form of
words/phrases or phonemes is accessed by the program, and converted
from digital to analog form for transmission. If the receiving unit
is a push button telephone, with multifrequncy tones, the user can
send inquiries via the push button. See PHONEME, SPEECH
SYNTHESIZER, VOICE
ANSWER BACK.
BACK.
audit traU In programming, a clerical or automated method for
tracing the trans actions affecting the contents of arecord. See
RECORD.
augend The number to which an addend is added to produce the sum in
an arithmetic operation. See ADDEND.
authentication In data security, processes that ensure everything
ab out a teleprocess ing transaction is genuine and that the
message has not been alte red or corrupted in transmission. The
parties to the trans action must identify each other reliably, know
that each message they receive comes from the other party and has
not been modified, or stored earlier and replayed, by a third
party. Acheck field can be added to the message such that the
calculation of the check field depends upon the contents of the
whole message, and the calculation involves a secret key.
Additional information is re quired in the message to guard
against replay of earlier messages; time/date fields are sometimes
employed for this purpose. Public key cryptosystems can be used for
authentication. In this case the message is
encrypted with the private key and de crypted with the public key.
Such messages can be decrypted by an attacker, but only the holder
of the private key is able to create a meaningful message. See
CRYPTOGRAPHY,
MAC, PUBLIC KEY CRYPTOSYSTEM, REPLAY,
WIRE TAPPING.
authentication of users In data communica tions, the verification
that the user at the terminal corresponds to his claimed
identity.
author The writer of books, articles, com puter software,
computer-assisted leaming packages etc.
authoring In interactive video disk, a struc tured approach to
developing all elements.of an interactive video disk program wlth
emphasis on preproduction. See PREPRODUC
TION.
authoring system In computer-aided leam ing, a computer system
capable of executing an author language. See AUTHOR LANGUAGE.
authorization In operations, the right given to a user to
communicate with or make use of a computer system or stored
data.
authorization code In operations, a code used to protect against
unauthorized access to da ta and system facilities. The code
normally consists of a user id (identification) and password. See
PASSWORD, USER ID.
author language In computer-aided leam ing, a programming language
used for de signing instructional programs for computer assisted
instruction and computer-based training systems. See
COMPUTER-ASSISTED
LEARNING, COMPUTER-BASED TRAINING.
ANSWERING.
autodialer In computing and videotex, a device that automatically
dials a prerecorded telephone number for connection to a host
computer.
Autoexec See AUTOMATIC PROGRAM EXECU
TION.
autoidentifier In computing and videotex, a device by which a
terminal automatically identifies itself to a computer.
autokeying In peripherals, the ability to record frequently used
groups of keystrokes and to play them back at designated loca
tions on the screen.
automata theory A mathematical study of the systems that receive
discrete inputs, change their internal states according to the
input, and their current states, and deliver outputs according to
their internal states and inputs. See Tu RING MACHINE.
automated teller meehanism In peri pherals, a device that provides
for cash withdrawals, payment of bills, account balance inquiries
and transfers of funds between accounts. See SELF-BANKING.
automatie Pertaining to a process or device that, under specified
conditions, functions without intervention by a human
operator.
automatie answering In data communica tions, a system in wh ich
the called station automatically responds to the calling signal;
the call may be established whether or not the called station is
attended.
automatie eaIling In communications, a machine feature that allows
astation to initiate a call automatically over a switched line. See
STATION.
automatie eaIling unit In data communica tions, a device that
enables a business machine to dial calls automatically over a
network. See AUTODIALER.
automatie earriage return In word process ing, the automatic
performance of a carriage return when the last word that will fit
onto a line of print is typed. A system that has this facility will
usually employ a buffer to hold the word currently being typed
until it can decide whether to place the word on the CUTrent line,
or to wrap it onto the next line. See CARRIAGE CONTROL,
WRAPAROUND.
automatie eentering In word processing, the automatie ability to
center a word or portion of text.
automatie data proeessing Data processing performed by computer
systems as com pared with manual systems. See ELECTRONIC
DATA PROCESSING.
automatie letter writing 21
automatie decimal alignment In word pro cessing, the feature of a
machine that enables numbers to be aligned automatically on either
side of a decimal marker.
automatie decimal tab Synonymous with AUTOMATIC DECIMAL
ALIGNMENT.
automatie dialer See AUTODIALER.
automatie dietionary (1) In machine trans lation, a database that
provides a word for word substitution from one language to another.
See TERM BANK. (2) In information retrieval, a system that
substitutes codes for words or phrases in the encoding
operation.
automatie file seleet In word processing, a facility for making a
selection from a data file based on the characters that appear in a
specified data field. For example, using a zip code field, the
system can select all the addresses with a 12480 zip code for one
letter, and print a different letter for all other codes etc.
automatie file sort In word processing, a facility for performing
sorts on files in alphabetical or other order. This feature is
useful for manipulating address lists so that changes need not be
performed in alphabet i calorder.
automatie footnote tie in In word process ing, a system that ties
a footnote to the appropriate text segment. If the text seg ment
is moved to another page or document, the foot note will travel
with it.
automatie headers/footers In word process ing, the ability to
place header/footer text at the top or bottom of each page of a
multi page document. The operator specifies the text once, and the
header/footer (usually document title, company name or con
fidentiality requirements) is automatically added during print
out.
automatie letter writing In word process ing, the ability to
produce a standard docu ment as though it were typed specially for
the recipient.
22 automatie line/paragraph numbering
automatie line/paragraph numbering In word processing, a facility
whereby the system automatica11y supplies an identifying number for
each li ne or paragraph during input for use in defining locations
during sub se quent editing. The line/paragraph numbers are
automatica11y deleted during final print out.
automatie line spacing In word processing, the ability of a printer
to perform different line spacings without the need for operator
intervention.
automatie loader In programming, a loader program implemented in a
special ROM that a110ws loading of the first re cord or sector of a
backing storage device. See BACKING STOR
AGE, LOADER, ROM.
automatie logging In word processing, a facility in wh ich a system
automatica11y records titles and log numbers with a11 documents.
Thus data can be played back separately when required. See
LOG.
automatie margin adjust In word process ing, a facility to change
margins with a single command. Line endings are adjusted with out
further intervention.
automatie message-switehing center In data communications, a
location at wh ich mess ages are automatically routed according to
the information they contain. See MESSAGE
SWITCHING.
automatie page numbering In word pro cessing, a facility to
generate automatically page numbers within documents. When text is
rearranged and page numbers change, the system can generate a new
set of correct page numbers. See AUTOMATIC PAGINATION.
automatie pagination In word processing, a facility to take a multi
page document and divide it into pages of a specified length in
terms of line numbers. Often, this feature is joined with the
capability to generate auto matica11y page numbers. See
AUTOMATIC
PAGE NUMBERING.
automatie poIling In computing and data communications, a feature
of a transmission control unit that enables it to handle ne
ga-
tive responses to polling without interrupt ing the CPU. See CPU,
POLLING.
automatie program exeeution In operating systems, a facility that
automatica11y starts a specific program every time the operating
system is loaded.
automatie programming In programming, the use of a computer to
convert a program written in a convenient language for the programm
er into a set of machine code instructions that can be executed by
the computer. See HIGH-LEVEL LANGUAGE,
TRANSLATOR.
automatie restart In operations, a facility to perform
automatically the initialization functions necessary to resurne
operation following an equipment or power failure. See
INITIALIZATION.
automatie retransmission request In data communications, a
technique to ensure accurate transmission of data. Data to be
transmitted is held in a buffer until the communication link is
ready to deal with it. The data is then despatched and a copy made
at the same time. The copy is de posited in the buffer and erased
when the sending device receives acknowledgement of correct
receipt, as verified by CRC. If the receiving device detects an
error in the data it informs the sending device, which then
retransmits the buffered copy. See CRC.
automatie tab memory In word processing, a facility of a system
that enables it to store a format of tab settings to be restored
auto matica11y to the typewriter at the time of printing.
automatie volume switehing In backing storage, access to a
sequential data set that extends across two or more volumes, and to
concatenated data sets stored on different volumes. See
CONCATENATE, SEQUENTIAL
DATA SET, VOLUME.
automatie widow adjust In word process ing, a facility that
prevents the first line of a paragraph, title or heading from being
the last line on a page. It mayaIso prevent the last line from
being the first line on a new page. See WIDOW.
automatie word wraparound In word pro cessing, the automatie
placing of a word onto the next line if it does not fit onto the
line being typed; frequently combined with the automatie carriage
return feature. Also it is used to denote systems that can wrap
words during margin adjust procedures. See AUTOMATIC CARRIAGE
RETURN.
automation The technology concerned with the design and development
of pro cesses and systems that minimize the neces sity of human
intervention in their oper ation. See FEEDBACK.
auxiliaryequipment In architecture, equip ment not under the
direct control of the CPU. See CPU.
auxiliary storage In backing storage, data storage other than the
main memory, usually with slower access, e.g. magnetic tape or
direct access devices. See D1RECT ACCESS
STORAGE DEVICE.
availability In operations, the degree to wh ich a system or
resource is ready when needed to process data. See UPTIME.
available point In computer graphics, an addressable point at which
a characteristic, such as color, intensity or on/off condition, may
be specified. See PIXEL.
azerty keyboard 23
available time Synonymous with UPTIME.
avalanche photodiode In electronics, a photodiode operated with a
high reverse voltage. Hole electron pairs are produced by incident
infrared or light energy, and these carriers are swept to the
appropriate elec trode. The electron carriers can collide with
other atoms releasing more electrons, hence increasing the
sensitivity of the device. Compare PIN PHOTODIODE. See
PHOTODIODE,
SEMICONDUCTOR.
average Synonymous with ARITHMETIC MEAN.
average access time In backing storage, the average time between
the instant of a re quest for data and its delivery from a storage
device.
average delay In communications, the average time that a caller
must wait for access to a communication facility.
AVIP See BAVIP.
A WC See ACTIVE WIRE CONCENTRATOR.
azerty keyboard A keyboard arranged as on the standard typewriters
of continental Europe, with the keys a, z, e, r, t, y on the upper
left-hand side. Compare DVORAK KEY
BOARD, MALTRON KEYBOARD, QWERTY KEY
BOARD. See KEYBOARD.
B babble In communications, the aggregate cross talk from a number
of interfering sourees. See CROSS TALK.
backend In programming, a functional unit of a compiler. See
COMPILER.
back faces In computer graphics, the facets of a three-dimensional
object that cannot be seen from the current position of the obser
ver. Deletion of current back faces can reduce the effort in
computing the screen display of the object. See CULUNG, HIDDEN
UNE.
background See BACKGROUND PROGRAM.
background color In videotex, the color filling the parts of the
character rectangle not occupied by the character itself. The
background color may be black or one of the seven display colors.
It may be changed within a row by control characters. See CHARACTER
RECTANGLE, CONTROL CHARACTER,
DISPLAY COLOR.
background ink In optical character recog nition, a type of ink
that is not detected by the scan head because of its high
reflective characteristics. It is used for print location guides,
logotypes, instructions and any other desired preprinting that
would otherwise interfere with the scan head reading.
background job In operations, a job of relatively low priority in a
multitasking environment; computer resources are only allocated to
it when they are not required for higher-priority foreground tasks.
Compare FOREGROUND PROGRAM. See MULTITASKING.
background processing (1) In operations, the execution of
lower-priority computer programs when .higher-priority programs do
not require any system resource. (2) In word processing, the
execution of an operator's request such as printing a document
while the operator is performing other tasks. See BACKGROUND
PROGRAM, PRIORITY.
24
background program In operating sys tems, a program that does not
involve interactions with a user and is run with a low priority in
a multiprogramming environ ment. Compare FOREGROUND PROGRAM. See
BATCH PROCESSING, MULTIPROGRAMMING.
background region In operations, a region in main memory to which a
background job is assigned. See BACKGROUND JOB, MAIN
MEMORY.
backing storage In peripherals, an inter mediate storage medium,
e.g. magnetic tape, magnetic disk etc., on to which data is entered
for later processing by the central computer. Ideally computers
would be pro vided with a single nonvolatile store that had random
access, extremely low access times, of the order of nanoseconds,
and massive capacity , of the order of gigabytes. However fast,
random access storage is expensive, and economical high-capacity
storage devices have slow access, of the order of milli seconds,
or even seconds, and provide only se rial or direct access. In
practice therefore, computers are provided with hierarchies of
memory: sm all volatile cache memories for nanosecond access by the
CPU; volatile random access memories with capacities of the order
of kilobytes, or more, with access times of the order of
microseconds; and nonvolatile backing stores of magnetic disk,
magnetic tape, optical digital disk, magnetic bubble etc. with
storage capacities in the hundreds of kilobytes to gigabyte range.
Data is accessed from backing storage in either serial or direct
access mode; in the first case all stored data must be read until
the required item is attained, whereas in direct access it is
possible to jump to a required block of data and the individual
data items are then read in aserial mode.
The programmer must arrange for the transfer of data to and from
backing storage in such a manner to ensure that program execution
is not unnecessarily delayed by such accesses; data that will be
required
after the computer has been switched off must be transferred to
backing storage before the end of program execution. Back ing
storage is also employed to hold soft ware. See BUBBLE MEMORY,
CACHE MEMORY,
CPU, DIRECT ACCESS, GIGABYTE, KILOBYTE,
MAGNETIC DlSK, MAGNETIC TAPE , MEGABYTE,
NANOSECOND, OPTICAL DlSK, RANDOM ACCESS ,
RANDOM ACCESS MEMORY, SERIAL ACCESS ,
VOLATILE STORAGE.
backlog In operations, application pro grams that have been
authorized but not written. See APPLICATION PROGRAM.
backplane In hardware, the connector blocks and wiring units that
provide the me ans of interconnection between a com puter and its
peripherals. It normally com prises aseries of multiway sockets
connected to the internal bus of the computer. See BUS,
MOTHERBOARD .
back porch In television, a picture signal that lies between the
trailing edge of the line sync pulse and the trailing edge of the
corresponding blanking pulse. Compare FRONT PORCH . See BLANKING
.
backspace In peripherals, to move the printing head on a printer
back one charac ter position, or the cursor on a VDU back by the
same amount. See VDU.
backspace character In printing, a control character that causes
the print or display position to move one position backward along
the line without printing or displaying any graphic character. See
CONTROL CHAR
ACTER.
backtracking In programming, a technique of searching for a goal
which allows for the possibility that a given search path may prove
to be a dead end. Algorithms using this method retreat from such
dead ends and recommence the search along a different path that was
not subjected to an earlier search. See PROLOG.
Backup In operating systems, a command commonly employed to backup
one or more files from a fixed disk to floppy disks. See FILE,
FlXED DlSK STORAGE, FLOPPY DlSK.
backupIrestore 25
backup In reliability, pertaining to a sys tem, device, file or
facility that can be brought into action in the event of a mal
function or loss of data. See BACKUP/
RESTORE.
backup copy In operations and word pro cessing, a copy of a file
or data set that is kept for reference in case the original file or
data set is destroyed. See BACKUP/RESTORE,
DATA SET, FILE.
backup diskette In backing storage, a disk ette that contains
information copied from another diskette . It is used in case the
original information is unintentionally altered or destroyed. See
BACKUP/RESTORE.
backupIrestore In backing storage, the actions involved in
transferring data from magnetic disk to tape, or disk, for backup,
and the subsequent action of restoring the data to disko Winchester
disk drives have fixed disk systems, and their data must be
periodically transferred to another medium to provide insurance
against data corruption or a drive failure , which would render the
disk data inaccessible. The available tech niques for backup are
floppy disks, magnetic tape (1/2 or 1/4 inch) and removable
disks.
Floppy disk backup suffers from the rela tively low storage
capacity of diskettes, compared with Winchester disks , and their
comparatively low data transfer rates. Back up operations should
be convenient to the operator, otherwise there exists the danger
that they will not be performed with suf ficient regularity.
Backup onto floppy disks often involves tens of minutes, and
contin uous operator attention , as new disks are inserted and
backup instructions given for the various sets of disk files.
Tape streamers using 1/2- or 114-inch magnetic tape were developed
for backup operations; such streamers are designed for the
continuous flow of data to and from a computer. They are simpler
mechanically, and hence cheaper than conventional start/ stop
magnetic tape devices. These devices require a continuous flow of
data to, or from, the read/write heads for efficient operation, and
the software and interface systems must provide this continuous
flow, or the repositioning cycles of the streamer will seriously
extend the total time of the
26 Backus Naur Form
backup operations. Start/stop magnetic tape drives are akin to the
magnetic tape trans ports used in mainframe systems and, as the
name implies, are designed to cope with more intermittent data
flows. They are more sophisticated mechanical devices, and hence
more expensive, than tape streamers, but provide for greater
flexibility in the b