COMPUTING CULTURE:
THE STATE OF THE NATION

u WERSI iY OF L~GOB
,-'SHAe J

Joshua Ojo Akindele AVENI
B.Sc., PGD (Lagos), M.Math., Ph.D. (Waterloo), FCOAN, FCPN

Professor of Computer Science
University of Lagos

University of Lagos
2001



J.O.A. Ayeni, 2001

All rights reserved. No part of this publicetion may be reproduced,
. stored in a retrieval system or transmitted in any form or by any

means, electronic, mechanical, photocopying, recording
or otherwise, without the prior permission of the author.

ISSN 1119 - 4456

Printed by University of Lagos Press

INTRODUCTION
What is computing? Or let us begin with computer science. What is
computer science? That is a question many experts have endeavoured
to answer in the past and in recent times. Some have preferred to describe
what computer science is about rather than give it a definition [Hamming
(1968), Longe (1983), Denning et al. (1989)]. There is no doubt that a
proper understanding of what computer science is would enhance general
understanding of what computing discipline is all about and its culture.

The frequent discoveries involving the use of computers in the society are
enough to confuse the uninitiated mind as to what computer science is
really. Is computer science the same thing as computer technology,
computer engineering, information science. information technology,
Internet, computer studies? Not exactly.

Computer science is the body of knowledge and practices used by
computing professionals in their work. The discipline is also variously
called computer science and engineering, computing or informatics
[Ralston (2000)]

The computing profession is the people and institutions that have been
created to take care of people's needs in information processing and its
coordination through world wide communication systems. The profession
contains various specialties such as computer science, computer
engineering, software engineering, information systems, domain-specific
applications and computer systems.

Computing discipline is characterized by three recurrent themes: theory,
abstraction and design [Denning et al. (1989)].

Both computer science and computer engineering agree on fundamentals
and are concerned with all elements of information processing systems,
including the application environment, paradigms for representing
information, paradigms for processing information and tools for designing
and measuring the effectiveness of computer systems. However, design
is the pre-eminent domain of computer engineering while theory is central
to computer science. Through abstraction, both disciplines master
complexity.

1

\~ERS\ Or t\llU'
U L.\6R R \



The term information science was coined to designate an interdisciplinary
field initially concemed with the exponential growth of recorded scientific
information. Evolving from its roots in librarianship and documentation,
information science has been predominantly occupied with information
"content". The portfolios of concems of information scientists today include
issues of intellectual property, individual information privacy, equity in
information access, etc. There is agreement that, as yet, a scientific
basis for a science of information remains to be developed.

Information Technology (IT) is a term frequently applied to a broad area of
activities and technologies associated with the use of computer and
communications, but generally implying the application of computers to
storage, retrieval, processing and dissemination of data. Thus, IT profession
has an enormous scope including science, engineering, business and
broadcasting.

In some usage ofthe term, IT can substitute for computer, e.g., IT software
means computer software while IT products include computers, data
communication and data broadcasting equipment. Also IT service is any
service which results from the use of IT software over a system of IT
products for realizing value addition [GI Task Force (1998)]

The "IT family" as coined in Ralston (2000), consists of professional groups
in different sub-areas of computing discipline, some of which see
themselves as peers of computer science within the IT profession and
are hesitant to identify with computing. We now have IT specialist lawyers,
firms with IT departments, govemments having ministries of IT,universities
establishing IT faculties, IT journalists are legion and self-styled IT
professionals/consultants abound.

Information Technology may not survive as a well-defined discipline like
computer science or civil engineering largely because of its broad scope,
however, it should survive as an approach or methodology like Science or
Art. It is in this light that a faculty of Information Technology is very welcome
in the universities where there is one and it is expected that computer
centers in the universities may become Information Technology Centres
which would incorporate IT services in their activities.

2

MacMillian Oxford Dictionary defines culture as the
"sum total of the beliefs, accomplishments and behaviour
patterns of a group of people, acquired by members of
the group through social learning and transmitted from
one generation to another; intellectual and artistic content
of this".

In the first part of this lecture, we will be discussing the beliefs,
accomplishments and behaviour patterns in computing since the birth of
the discpline in the early 1940s. My own humble contribution will be
presented in the second part. The final part will consist of computing
culture in Nigeria, conclusions and recommendations.

ACCOMPLISHMENTS IN COMPUTING
These are in the areas of hardware, software, computer architecture and
Internet. 1. . "; oJ L. ",Y

\. t •.••.I •. sa R.,
HARDWARESUB~YSTEM
Hardware is the physical part of the computer and it consists of the
memories, storage devices, peripherals, routing and control circuits.

Memory
Memory or storage sizes are measured in bytes, kilobytes (1024 bytes),
megabytes (1024 kilobytes), gigabytes (1024 megabytes), terabytes
(1024 gigabytes) and petabytes (1024 terabytes). These are respectively
abbreviated as B, KB, MB, GB, TB and PB. c
Technologies used in realizing the main memory of the computer had
advanced from the non-volatile magnetic (ferrite) core (which is now
Obsolete)to the volatile Metal Oxide Semiconductor Field Effect Transistor
Dynamic Random Access Memory (MOSFET DRAM) used for main
memories in modem times. Other technologies used are MOSFET SRAM
and bipolar SRAM but MOSFET DRAM is currently the dominant
technology. By 1999, 64M bit DRAM chips were available as against the
64-bit bipolar SRAM chips used in 1960s by early computers such as
IBM system 360. It is projected that DRAM chips will contain up to 256G
bits by 2012.

The Holy Bible, King James Revised Version, contains less than 51/2
million characters. It is remarkable that the memory size of some present
day microcomputers/servers is as big as 256MB. This means that such

3



microcomputers can store in memory 40 different books each of the size
of the Holy Bible simultaneously. This is a great achievement bearing in
mind that the mainframes of the 195qs with 512K memory size cannot
accommodate a single copy of the Holy Bible in the memory all at a
time!

Storage Devices .
Storage devices cause information to be stored for later use In the
computer. Over the years storage devices have become ~ore
sophisticated and varied. Storage devices now have fast access. time,
large capacity and are low cost. A disk-based mass storage Unit now
has a capacity for up to 2TB (251 bytes) costing just 9 cents for 1MB. The
cost is even lower for DVD-RAM. These mass storage devices have up to
10ms access time (this is 100 memory accesses in one second) and
1MB/sec transfer rate.

Peripherals .
The peripherals are devices that chanqe information from one ~hyslcal
form to another and link the computer with its environment. Penpherals
include keyboard, video screens, printers and various input devices such
as mouse, button, joystick and lightpen.

A major accomplishment in development of pertpherals has t? do.with
technology requirements for multimedia computing applications.
Multimedia computing implies the integration on computer system, of
audio, video and images with the traditional types of data such as text
and numerics and is based on the evolving ability of computers to store,
transmit and convey diverse types of information. Some multimedia
computing applications are: online news, advertising, .electronic m~iI,
banking, distance learning, digital libraries and medical information
systems.

The technological requirements of a Distributed Multimedia tntorrnation
System (DMIS) concern workstation technology and include: at I~ast a
CD-ROM at the workstation for" assemblage of data for retneval,
communication protocols, bandwidth, internetworking, data storage,
application interfaces and authoring tools, information modeltinq and
retrieval. The high-resolution monitors and audio output are being used
as presentation devices while data capture devices include: scanner for
still images, microphone with digitizer for voice data, keyboard for text
and camera with digitizer for video.

4

Work is currently going on to recognize voice and handwritten data entry
and will undoubtedly lead to new devices. Personal Digital Assistants
(PDA) allow input of a constrained set of hand printed characters. Data
tablets are available for personal computers. These data tablets are
essentially "electronic yellow pads" on which the user may make sketches
and handwritten notes. Small video cameras are now available which
allow two similarly equipped network users to converse through phone-
picture.

Routing Circuits
The routing circuits (including buses and networks) direct flow of information
between various functional parts of the hardware subsystem, that is, the
main memory, the central processing unit and the storage devices. In the
mainframe and minicomputers, we have buses that are capable of
transferring and receiving data at the rate of up to 4.5MB/sec. The 32-bit
Extended Industry Standard Architecture (EISA) bus developed by the
gang of nine - Wyse, ASI, Tandy, Compaq, Hewlett-Packard, Zenith,
Olivetti, NEC and Epson has data rate of up to 33MB/sec.

SOFnNARESUBSYSTEM !
Programs that are needed to make computers perform their intended
tasks are called software. Th'ey include operating systems, programming
languages an graphical user interfaces (GUls).

A complete software system will contain, in addition to the operating
system, a set of compilers for various languages, one or more system
loaders, one or more database systems, sets of utility routines, special
and general-purpose debugging systems and generalized subsystems
for applications such as sorting and merging, mathematical programming,
engineering design, report generation, simulation graphics and so on.

Early operating systems were control programs that were a few thousand
bytes long. They scheduled jobs, drove peripheral devices and kept track
of system resources usage for"billing purposes. Modern operating
systems are much larger, ranging from hundreds of thousands of bytes
for personal computers, e.g. MS-DOS, XENIX etc. to tens of millions of
bytes for main-frames, e.g. Multics, MVS, UNIX etc. and hundreds of
millions of bytes for some servers, e.g. Microsoft Windows NT. In addition
to managing processors, memory and dozens of inpuUoutput devices,
modern operating systems also provide interprocess communications,

5



file and directory systems, data transfer over local networks, command
languages and graphical user interfaces for invoking and controlling
programs. An operating system "does not only perform allocation of
computing resources among processes contending forthem, it also serves
as an extended machine that provides a powerful programming
environment.

Up till about 1969, it was generally assumed that the purchase or rental
of computer hardware system entitled the customer to all general-purpose
software produced by the manufacturer for that computer at no cost. In
June, 1969, IBM announced that it was introducing a new policy of not
supplying computer software along with the computers sold or rented
with the exception of essential operating system software. Almost all
other hardware manufacturers followed the lead Of IBM and unbundled
their software products. We now have several organizations which
specialize in producing software for use with computers even though
they do not manufacture hardware. Microsoft, Oracle, Magic software
Systems, Dbase Inc. are examples, just to mention a few. Bill Gates,
founder of Microsoft was rated the world's richest man (The Guardian,
September 23,2000) in year 2000 with an estimated asset of $63 billion.
What is Nigeria's gross domestic product in 2000? The 1999 estimate
was $110..5 billion! [The World (2000)].

COMPUTER ARCHITECTURE AND INTERNET
Computer architecture has to do with the hardware structure of computer
systems, the attributes of their various parts and how these parts are
interconnected. One important achievement in this area is the issue of
compatibility between computers. A family of computers that share a
common system architecture allows a software written for any model in
the series to operate correctly. This is realized' by allowing various
members to share the same instruction set, arithmetic registers, input/
output methods, memory addressing mechanisms, etc' In certain
instances, new members of the series have extended the architecture in
an upward compatible way, that is, software written for the older series
can run on the newer ones, but not always vice-versa.

The Internet is the interconnection of computer networks world-wide and
its reallsatlon can be considered the greatest achievement in the
computing profession in recent years. It has made the world to become
a global village. It allows information to be passed from one network and

6

communication media to another in digital packets. The physical layer of
the Internet connects users on telephone, satellite and cable TV network;
local area networks (LANs) and wide area networks (WANs).

The Internet Protocol (IP) is used to route the digital packets in an efficient
and generally reliable manner. The Transmission Control Protocol (TCP)
manages the transmission and receipt of the entire messages between
two points on the network using the IP packets. The application protocols
such as Email, World Wide Web (or simply Web), Search engines e.g.
Yahoo!, e-commerce, intranets and extra nets are built on the TCP.

BEHAVIOUR PATTERNS
Pattern is a model orway in which something happens, moves, develops
oris arranged. Behaviouris a way of acting or functioning. Thus, we can
infer from the above that behaviour pattern in computing is the way things
are happening in the various sectors of computing discipline and/or how
the various sectors have developed or are developing over time. The
relevant sectors are the hardware and software sectors·iJ IV R 11 U l

LIBR RHARDWARE SECTOR
Until about 1968, most computers were mainframes that were expensive
and required considerable floor space, electric power and air-conditioning
and so were used as shared systems. However, due to the rapid advances
in computer technologies, especially conductor circuitry, by the late
1960s, a genre of small size machines termed minicomputers was
produced. The machines occupied only the space of an office desk, and
were inexpensive enough to be used forthe solution of a single problem,
such as the control of a particular industrial process as well as a variety
of shared uses.

By about 1973, further rapid advances in technology led to another class
of even smaller, cheaper systems based on the microprocessor - a
complete CPU contained on a single or very few semiconductor chips.
This method of manufacturing eliminated most hardwiring and separated
component manufacture of the CPU with substantial improvements not
only in cost but also in reliability. This great advance in technology was
matched by corresponding advances in semi-conductor memory that
allowed compact, high-capacity main storage to be offered at a price
hundreds of times cheaper per unit of stored information than was available
only a decade earlier.

7



By 1990, several million microprocessor-based systems, called personal
computers or workstations were in common use in homes, offices and
businesses. Each gave to an individual user computational power and
main storage that had been available in shared form only from large
expensive systems just a few years earlier. Furthermore, the progress in
technology that has given us the microprocessor has also enhanced the
cost-effectiveness of the large minicomputer and the general-purpose
mainframe systems.

SOFWARE SECTOR
From the outset, the development of software has been directed toward
the apparently contradictory, but in fact complementary goals of bringing
the computer closer to the user while keeping the user at a distance. The
first goal had involved the creation of programming languages and systems
to facilitate the development of the application that make the computer
useful. The second goal had included the operating systems that oversee
these applications and manage the hardware and software resources on
which they draw. The third and present goal has been in interposing
layers for constructing large software efficiently.

Programming Languages
We can define a language as a particular system of words, symbols,
sig ns etc. used for conveying information. Then such system as used by
the English people is English language and the one used by the computer
professionals to direct the operations of the computer is a Programming
language.

Development of languages started as early as 1945 with the design of a
high level programming language called PLANKALKUL by Konrad Zuse
[Bauer (1972)] in Germany but was never implemented. Autocode, the
first working high level programming language was developed by Alick
Glennie in 1952. However, the first operational compiler was developed
by Lanning and Zierter for mathernatical computations and implemented
on MIT whirlwind computer in 1954 while John Backus' Fortran compiler
was most widely used in mid - 50s.

Early languages maintained dichotomy between scientific calculations
and data processing. Early examples are Fortran (1956) for scientific
calculations and Cobol (1960) for business data processing. With the
advent of interactive computing, languages like Joss (1963) and Basic
(1964) were developed for effective use in an interactive environment.

8

Between mid-1960s and late 1970s, languages were developed for
multipurpose (or general) processing. These include PU1 (1964), Algol
68 (1968), Pascal (1971) and C (1973) for system programming. More
powerful languages suitable for embedded computer systems (i.e.
systems which involve computer and other equipment) were developed in
the 1980s and 1990s. These include Ada (1979), C++ (1980) and Java
(1995).

Standardization of programming lanquaqes started in the early 1960s
with the American National Standards Institute (ANSI) coming up with
the first Fortran standard X3.9 in 1966. Standardization aims at decreasing
the amount of conversion required to move a program from one environment
to another. Subcommittee 22 (SC 22) of ISOIIEC (International
Organization for standardization / International Electro-technical
Commission) Joint Technical Committee 1 (JTC 1) is an international
body which is responsible lor international standardization of languages
for information technology. It has working groups for various languages
like Ada, APL, C, C++, Cobol, Java, etc. in different countries.

The following language concepts have endured: formal method for defining
language syntax (BNF: Backus- Naur'- Form), block structure (AlgOl),
data typing, record structure and separation in a program of data and
procedural aspects (Cobol), class concept (Simula) which has led to
modern concepts of data .abstraction and Object-oriented programming,
strong data typing mechanism (Pascal) and conceptual autonomy (Java).
Java includes in its design the specifications for the virtual machine on
which it runs.

E'f A ~O
Operating Systems L B AR .
In the early days of computing, the programmer tended to operate the
computer alone. He would mount the tape, put punch card into card
readers, etc. This resulted in inefficient use of the programmer's time.
Also, the computer processor was having too much idle time since it has
to wait for the programmer to give instruction to continue each time it
paused. About 1956, a simple operating system was developed jointly
by General Motors and North American Aviation for the IBM 704. This
system provided facilities for sequencing from one job to another, input!
output control system, calling compiler, loading jobs, etc.

9



By the 1960s, large and powerful batch operating systems, e.g., OS/360
for IBM computers were developed. They provided facilities for
multiprogramming.

By early 1970s, UNIX became widely accepted as a model of small but
powerful interactive operating system designed to facilitate the work of
the programr(ler. .

With the arrival of interactive systems, more people, many of whom are
non-computer specialists became interested in using computer. Hence,
attention was tumed to providing suitable user interface. Earlier, interfaces
consisted of jumper wires in patch boards, punch cards, prepared off-line
print-outs etc. Today,' the user interfaces include keyboards, mice,
graphical displays, etc. .

The rapid expansion of memory and processing speed dedicated to single,
stand-alone use but with opportunity to access other computers in a
network led to the development of operating systems anchored on graphical
user interfaces (GUls) or WIMP (Windows-Icons-Menus-Pointers), the
latest of these being MicrosofLWindows 2000. The use of mouse - driven
graphical user interface was popularized by the introduction of Apple
Computer's Macintosh in 1984 having been inspired by earlier research
at XEROX PARCo

Software Development
Programme is a list of planned activities to be carried out. Ina similar
vein, software or computer program is a list of instructions/activities to be
executed/performed on a computer. [The act of putting together a
program(me) is called programming].

In the past, programs were thought of as instruction sequences. Software
production then was on ad-hoc basis employing subroutines or
subprograms. Constructing large software was very expensive in spite of
the relatively inefficient programs that were produced.

Following Edsger Dijkstra's letter to me Editor ofthe Communications of
the ACM in March, 1968 entitled "Goto Statement Considered Harmful"
[Dijkstra (1968)], research was intensified in constructing programs using
a metho.dology called Structured Programming (SP). This methodology
avoids using "goto" 1:1 designing programs as much as possible. In its

10

place, control structures were used. Bohm (1966) actually shows we
can construct such programs by using control structures to link parts of
a program together logically. Such programs are believed to be more
readable and maintainable. They are also more likely to be correct, as
parts of the program can be shown correct by systematic program
verification. The acceptance of structured programming philosophy by
professionals is reflected in the wide use of structured languages in
teaching programming and language design in the profession.

Since the 1980s, however, object-oriented-programming approach has
become a prevailing form of software design. It packages data and
operations into objects that have simple interfaces through which they
interact. The approach is intended to manage complexity and facilitate
software reuse by clearly separating interfaces from the implementation
details.

UNIVE SIT or LA '0~
BELIEFS LtBRA v
Let us recognize two broad classes of beliefs in computing - those beliefs
which translate into professional norm ortradition (spiritual beliefs) and
those that model or shape future activities in the profession in some way
(technological). We expatiate further.

NORMS
Computer System Development Life Cycle (SDLC)
Conventionally, computer - based systems are implemented in many
stages depending on the complexity of the system. These stages include:

Investigation and detailed study of existing system;
Analysis and design;
System development (programming);

Testing; /
Change - over;
Maintenance and Review.

The above stated stages constitute a life-cycle of the computer-ba d
system in that the process can be repeated all over again to obtain a w
system ifthe existing system, at the review stage, is found to have be me
inefficient, inadequate or problematic. However, the new appro ch to
developing computer-based systems using rapid application deve pment
tools like Microsoft Visual Basic, Magic, Oracle, etc. cause t above

11 /



stated stages to be applied repeatedly at every stage of the new approach
which generally consists of 3 stages viz: analysis, prototyping and
deployment especially at the prototyping stage even before the final product
is deployed. One obvious advantage of the new approach to developing
applications is the greater end-user involvement in the development of
the new system.

Cost-benefit Analysis
Also, it is conventional in computing to base the justification for using
computers to solve orpanizationat problems on the versatility of the
computer rather than on cost minimization. The use of computer requires
substantial investment in hardware, software and manpower (humanware).
Thus, the gain in using computer to solve problems is not necessarily in
the amount of naira or dollars that might be saved in getting problems
solved, it is rather in its cost-effectiveness, that is, in getting value for
additional money being spent. However, results are usually obtained
quickly and timely and as accurately as possible when computers are
used.

Creativity
Computer professionals have tendency to be creative just like an artist.
Sometimes this feat is achieved during hacking. Hacking is a programming
style in which programs are created in an unconventional (non-systematic)
manner. Hackers believe that programs should be built "straight from
your mind" [Turkle (1984)]. They prize concise, efficient, elegant and
even tricky code. Although more conventional programmers frown at this
programming style, many hackers-developed projects such as operating
system Linux and text-editor Emacs are premier examples of the
programming craft. Many of the founding fathers-of modern computing
are former hackers - Bill Gates of Microsoft and Steve Wozniak of Apple
Computer fame are prime examples. On the dark side of hackers are
those who create programs called virus (and worms) to damage computer
systems. Love Bug infected 45 million computers in twenty countries

ithin 24 hours in 2000! [Ralston (2000)].

T HNOLOGICAL BELIEFS
He are some predictions made as a result of technological developments

. in th field of computing in recent times.

12

Software Market
Component software is designed to solve programming problems in small,
modular bits, rather than in large, functional programs. Eventually, it is
possible and likely that we might see components becoming an alternate
software market or even a replacement software market.

100th Edition of Encyclopedia in 2768
The first edition of Encyclopaedia of Computer Science was released in
1976, the second edition in 1983, the third edition in 1993 and the fourth
edition in 2000 - an average of 8 years in between editions. It is believed
that the 1OOth edition will be released in 2768.

7th Generation Computers to be announced in 2013
In 1991, Japan announced the coming of the fifth generation computers
based on Artificial Intelligence. In 1997, Japan abandoned fifth generation
computers and announced the coming of the 6th generation computers
based on Neural networks. It is believed that Japan would abandon the
6th generation computers in 2013 and announce the coming of the T"
generation computers.

Intel to release Pentium XCI in 2095
In 1993, Intel Corporation announced Pentium chip (Intel 80586) which
was, however, flawed because for certain operands, quotients of floatinq-
point divisions were accurate only to 3 decimals rather than the normal
16. Pentium II (Intel 80686) was announced in 1997. Pentium III was
announced in 1999. It is believed that Pentium XCI (1991) will be announced
in 2095.

Do you believe all these?

Other predictions
Windows 9.5K will be released in 9500
Work will begin on Y10K problem in 9999 all over the
world.

It is now expedient for me to go to the second part of the lectun

13



CONTRIBUTIONS
The focus of my research overthe years had been in the involvement of
computers in operations research activities, program compilation, games,
information systems and artificial intelligence. I would now delve briefly
into what my contributions are in these areas.

OPERATIONS RESEARCH (OR) TECHNIQUES
OR methodology came into limelight after World War II (1945) when the
British recorded some successes in the use of the method. The
Americans decided to capitalize the approach by setting up a department
in the Ministry of Defense to carry out research in this area. G.B. Dantzig
first came up with the Simplex Method to solve linear programming problem
in 1947. Linear programming is the first mathematical model that was
applied to solve problems of the firm under various constraints such as
scarcity of funds, materials and manpower; uncertainty, etc. Similar models
that have been researched into over the years and applied include
Forecasting, Inventory Control, Dynamic Programming, Queuing Theory,
Game Theory, Simulation and so on. All these models are usually referred
to as operations research models.

Search Methods
A large proportion of our day-to-day activities at home, at work, on the
computer, etc. involves searching and sorting. In fact, solution to any
problem requires one form of search or another. The search can be carried
out with optimization algorithm or a heuristic algorithm. In the former
case, the best possible results under the given circumstances are obtained
while in the latter case good solutions are found with significant
computational savings.

In certain situations, one has reliable information about the solution to a
problem. It is expedient for someone to exploit such information to an
advantage in any search strategy. In a research project supported in part
by the National Research Council of Canada under Grant A4124, Ayeni
and Magazine (1979) develope.d a heuristic algorithm that makes use of
a prior information in finding the maximum at location x* of a single-
variable unimodal function u(x). The functional form of u(x) is not known,
but an experiment can be performed for any x'T[a,b] and u(x) determined.
Alternatively, u(x) may be given but is too cumbersome to take advantage
of its structure (other than unimodality).

14

Let us represent the prior information as a probability distribution overthe
interval to be searched. Define SN= (X1' x2' ••• xN) as any sequential
search strategy with N experiments and ?N (G) as the class of such
strategies that make use of a given prior information, G.

Let LN(SN)be the interval of uncertainty which remains when strategy SN
is used. We wish therefore to find S*Nsuch that

E(S*J = E(LN (S*N» = min E(LN(SN» RC
SE?N(G)

When no information is given, the objective is typically to find that strategy
which minimizes the largest of any possible interval. Kiefer (1953) found
this strategy and called it the Fibonacci search. The strategy does not
make use of any given prior knowledge. This is a drawback of a sort.

Fine (1966) showed that if Gx• (x) is uniformly distributed, the Fibonacci
search minimizes the accepted interval of uncertainty. Shapiro (1974a,
1974b) found optimal strategies by solving a large non-linear programming
problem under a restricted class of strategies. For instance, when N =
20, his problem requires over 10,000 variables!

Ayeni (1977) found a strategy that performed the Fibonacci search on
the cumulative probability distribution Gx• and called it Fibonatile search.

The search has the same expected interval of uncertainty as the Fibonacci
search. The drawback of this strategy is that it considers the prior
information exclusively, ignoring the lengths of the possible intervals.

Ayeni and Magazine (1979) showed tnat the optimal solution lies in
between the Fibonacci and Fibonatile points. Ayeni (1977) has shown
that this is true if G* is convex, concave or dGx• (x) I dx is symmetric
about the midpoint of the interval. The observation is stated as a conjecture
that is true in general.
Based on the above observation, the problem can be stated as:

Find qj : where

qj = [(1-~) ifFN+ bj ~.\. (ifF N)] (b-a)
i = , ,2, , FN.1 ; O~ ~j ~ 1

The heuristic search strategy simplifies the problem by assuming ~i =
1,2, .... , FN.1are stationary i.e., we wish to find

15



qjH= [(1.- PH)i/FN+ pHG-1x' (i/FN)] (b-a)
i = 1,2, , FN-1' 0 ~ pH~ 1

The proposed solution SNHis simply to place xjHat qjHwhenever the
Fibonacci search places its experiments at point

(b - a) i/F N

The objective function becomes
F -1

N
E(SNH)= a (qj+1H- qt) G (qj+1H)- G (qjH)

i=O
which is clearly less than or equal to (b - a/F N)since b'' = 0 is feasible
and corresponds to the Fibonacci search.

The different measures were used to measure the performance of the
heuristic search viz:

D(SNH,SN*) = E (S/) - E (SN*)
_____ ' __ X 100%

E (SN*) .
is used to measure how go.od the solution is compared to the optimal
solution and

D (Fibo, SNH)= (b-a)/F N- E (SNH)
X100%-------

(b-a)/F N

is used to measure how well the prior information is used, in other words,
how well the algorithm performs when compared with Fibonacci search.
Some results are hereby presented.

Ayeni and Magazine (1982) show that when some priorin!ormation is
given, if the cost of experimentation is con~ider~d, a myoP.lc rule (o~e-
step-look-ahead) stopping strategy is not optimal for sequential strategies
that utilize the prior information.

In search of a best oossio'escneoule to work with, Aimufua and Ayeni
(1996) designed a heuristic algorithm for improving a u~iversity ~ecture
timetable. Penalties are assigned to periods depending on courses
scheduled for that period. A heuristic approach to minimize the sum of
penalties between any two periods in the entire timetable is ad~pted.
The algorithm terminates when no further improvement ca~ be.achl~ved
with the sum of the penalties following two or more consecutive iterations.

16

The last iteration is taken as the desired improved timetable. Analysis of
the algorithm shows that the computational order is O(n2) where n is the
number of courses on the time-table.

IJU 'OM RS
Mathematical Programming. L1BR R •
Mathematical Programming forms a large class of resource allocation
strategies that are widely used in Operations Research. Mathematical
Programming models include: linear programming, integer programming,
goal programming, geometric programming, dynamic programming,
quadratic programming and non-linear programming. Each model has its
own basic assumptions and hence is applicable to problems where those
assumptions hold reasonably. Linear programming model is most basic
and its assumptions include proportionality, e.g., if one unit provides N5
contribution to profit, 7 units should provide N35 contribution; determinism,
i.e., all parameters of the model are known constants; etc.

Linear Programming problem can be solved efficiently using any variant
of simplex method discovered by Dantzig (1963). Ayeni (1983) has shown
that application of non-linear programming model could be preferable to
linear programming model (even though it could be more computationally
difficult to solve) in some production systems. For comparison, Ayeni
used a separable programming strategy that employs adaptive procedure
of the Golden Section search as we have in Wagner (1975) to solve an
hypothetical production problem.

InAyeni (1980), a code for a computer algOrithm for solving a transportation
problem in Algol language was presented. The algorithm is a special
form of Linear Programming model. It uses the stepping-stone method to
determine the variable to leave the basis and one to enter the basis.
Ayeni (1988a) also coded the alqorithm in Fortran (WATFIV) on IBM 3701
145 computer. The program was capable of solving a transportation problem
that involves distribution of products from 20b origins (sources) to 200
destinations (markets).

Simulation
Determining the best time-slice (or optimal phase length) for a road junction
traffic control system depends on among other things: optimal number of
control phases to serve the junction and arrival and service patterns of
vehicles that enter and leave the junction. In a work to investigate these
factors, Ayeni (1988b) modeled the number of phases required as a colour

17



problem as suggested by Aho (1983). A closed-form mathematical
solution forthe problem does not exist for now.

The arrival pattern distribution is poisson and the service distribution is
exponential. Parameters of these distributions are to be obta~ned from
the field. A computer simulation model was implemented In BASIC
programming language to determine the optimal phase length (time-slice)
of a traffic light control system for a double T-junction.

The T-junction is typical of 3-access road junctions such as ~he o~e at
the AP petrol station and the one at the Bookshop at the University of
Lagos here. Three phases are adequate to control the junction. Data was
collected for estimation of the value of the appropriate parameter and
was processed. The simulation results indicate that the control system
gives better performance when the knowledge (prior informati~n) of~he
arrival distribution of vehicles into junction is taken into consideration
while allocating time to each phase than when it is ignored. It is also
found that the average queue length in the system is directly proportional
to the total phase length. Details of this simulation implementation are
contained in Ayeni (1990).

Game Theory
Von Neumann and Morgenstem (1947) provided the theory of zero-sum
two-person games and characterized their solutions. Zero-sum two-~erson
games are games played by two persons and whaton~ p~rson g.~In~the
other loses. A solution to the game is a pair of strategies In equilibrium,
that is, a player stands to be at a disadvantage if he fails to adopt a
strategy in equilibrium.

Ayeni and Longe (1985) characterize Ayo, a game that is widely played
in West Africa. A carved wooden box (board) called OPON is usually
used to play the game. The typical size of OPON is 50 x 20 x ~ ~c. It
contains two sets of 6 fist-deep holes dug on the board. Two additional
holes (initially empty) may be dug at each end of the board for storing
seeds won by each player as game proceeds. At the start of the game,
48 seeds of the plant Caesalpinia crista [igi ayo; see Odeleye (1977)] are
distributed in fours into the 12 holes. Each player sits comfortably close
to a set of 6 holes (called active houses in this lecture) facing the other
player. This set of holes or houses is said to belong to him.

18

There are many versions of the game. We limit our discussion to Ayoyayo
in this lecture. A player starts the game by selecting one of the active
houses (holes) in the set that belongs to him. The seeds in the selected
house are distributed anti-clockwise one seed into each next active house.
The distribution may continue up to and through the opponent's (other
player's) active houses and even back to the player's own active houses
and so on. However, no seed must remain in, or be distributed into the
original selected house at the completion of distribution.

For clarity of illustration, the houses had been numbered anticlockwise
from 1 to 12 and the non-active houses, I and II.

If player 1 (P) selects to deal from house numbered 5, the four seeds in
house 5 will be distributed into houses 6,7,8, and 9 in that order. If player
2 (Q) plays next as it should be, selects house 7, he will distribute the 5
seeds that are now in house 7 anticlockwise into houses 8,9,10,11 and
12. At this juncture, the OPON is transformed into a new state.

A player collects (wins) some seeds from OPON if, at the completion of
distributing the seeds from a house, he ends up at a house or at a set of
adjacent houses in the opponent's row that contains 2 or 3 seeds each
provided that the removal (or collection) of such seeds does not lead to
inadvertent stoppage of the game, that is, it does not lead to a situation
where the opponent does not have opportunity to continue with the game.
Either, all seeds won are removed completely or none of the seeds won
are removed from OPON at any stage of the game and those removed
may be deposited into the inactive house assigned to the player or if no
inactive house is assigned, such seeds are kept in one's fist for the
durationofthegame. [08"1' (,

nor vt . .;)
The game ends when any of the following events 0 AFi \J

o A player surrenders (accepts defeat)
ii) No more than 3 seeds remain on OPON
iii) A player causes inadvertent stoppage of the game
iv) Too few seeds remain on OPON, thus preventing

meaningful continuation of the game.

In the analysis of the qarne, Ayeni et al. (1985) show that using a one-
step look-ahead strategy (considered a myopic decision) to determine
which house to deal from, the resulting pay-off matrix for the game exists

19



and has a solution which is either a saddle point obtainable using minimax
criterion or a mixed strategy obtainable using simplex method to solve a
resulting linear programming problem. Even in a situation where odu or
ik3reconcept is considered, it is shown that the solution methodology
forthe resulting payoff matrix remains valid.

Ayeni et. at. (1990) describe a Pascal implementation of the Ayo software
as described above on an Apple II microcomputer. Also the work of Longe
and Ayeni (1991) discuss some practical difficulties encountered in
designing software for playing Ayo game. Some of these difficulties include
intermixing text and graphical modes, strategic selection and efficient
algorithm for configuring OPON.

PROGRAM COMPILATION
Human beings communicate instructions (that are to be obeyed) to
computer (and vice versa) and computers communicate instructions to
each other through one programming language or another. Spoken
languages cannot be used directly because computers do not understand
spoken languages directly. The programming language instructions
(usually called proqrarn) are coded in binary digits (bits). Communicating
instructions in coded bits are not natural to human beings neither is it
easy or efficient to do so. To overcome this difficulty, programming
languages that resemble spoken languages or natural languages are
developed and are used to communicate instructions to and fro computers.
A problem thus arises in that the instructions are generally not in coded
bits.

To resolve the issue, compilers and interpreters that translate these
programming language (usually called high level or very high level
programming language) instructions to coded bits that can be understood
directly by computers are desiqned. The process of translating or
converting these language instructions to coded bits is called compilation.
The first task-of a computer in trying to obey (execute) instructions is to
know what the instructions are. If the instructions are passed to the
computer in a written form, the task is called scanning process and it is
performed by the scanner or lexical analyzer.

Ayeni (1982) researches into problems that are usually encountered by
the lexical analyzer and proffer solutions. It is shown that the lexical
analyzer could ignore blanks (spaces or gaps) in instructions in many

20

I R R i
situations and still be able to produce correct transformation or tokens
for the instructions. Note that if the computer does not pick the tokens
correctly, it could be obeying different instruction from what the user has
intended. Use of regular expressions combined with table look-up is
prescribed for obtaining correct tokens.

The compiler (or interpreter), too, is a program written in a low or high
level language. Early languages such as Fortran, Cobol and ALGOL have
been improved upon by designing completely new languages such as
Pascal, C, C++, PU1, Ada, Java, etc. that feature new concepts in
programming languages. Thanks to automatic generators, compilers for
these new languages did not take too long to realize.

Instead of designing completely new languages, existing languages can
be improved upon by extending their capabilities. In this wise, the existing
compilers are still used to compile programs written in the "extended
languages". However, there is the need to preprocess the "extended
features" to enable the existing compilers understand the "added" features.
Uwadia and Ayeni (1988) extend Fortran IV language by incorporating
features for writing structured programs and inpuUoutput statements.that
are not in Fortran IV or ANSI Fortran languages. The extended language
is nicknamed LAGFOR. A preprocessor is written that pre-compiles these
statements that are written in the "added" features and the output of the
preprocessor is then compiled along with other source statements which
do not require preprocessing by the compiler of the basic language (using
the existing language compiler). Preprocessing offers much benefit with
relatively little effort [Kemighan (1976)].

One other difficult problem we have in compilation process is code
generation. It is the code generated that is actually executed on the
computer. It has not been easy to write general code generator because
the machines/computers for which codes are to be generated have different
instruction sets. The instruction sets determine what code to generate
for which instruction that is being processed on the computer.

One approach to solving this problem Is to develop a program called
code generatorto accept machine descriptions as input and produce as
output, code pattern tables. Since these tables are machine-specific,
the generation of code for a new target machine would require describing
the new machine and producing a new set of tables. Catell (1982), Graham
(1982), Henry (1984) and Aho (1985) adopt this approach.

21



Uwadia (1990) observed that some of these machines have similardesign
and this similarity can be taken advantage of to further simplify the
retargeting process. Uwadia and Ayeni (1994) reported a methodology
for obtaining a nucleus set from the instruction sets of Zilog80, Intel 8080
and Motorola 6502.

INFORMATION SYSTEMS
Reynolds (1988) defines information system as a special class of system
whose components are people, procedure and equipment that work
interdependently under some means of control to process data and to
provide information to users.

Mallach (1994) identifies various types of information systems. They
include transaction processing systems (TPS), management information
systems (MIS), and decision support systems (DSS). Others are: office
information systems (OIS), executive information systems (EIS), expert
systems (ES) and strategic information systems (SIS). Sprague (1980)
in his pioneering work on DSS observed that DSS specially focus on
features that make them easy to use by non-computer specialists in an
interactive mode. They also emphasize flexibility and adaptability to
accommodate changes in the environment and decision-making approach
of the user.

In the research carried out by Ayeni (1994) on the feasibility of a micro-
based prototype students' administration system with partial supporttrorn
the Central Research Committee, it was found preferable to design the
system as a distributed system that can be implemented on a network
of microcomputers [Ayeni and Uwadia (1996)]

A prototype implementation of the system as an MIS is documented in
Ayeni (1994).

A faculty subsystem forthe students' administration system in a university
environment is implemented in Dbase IV in a windows environment as a
decision support system (DSS) and is reported in Ayeni and Bamgbose
(1999). The software nicknamed SAS allows the following activities to be
carried out on the computer among others:

22

set parameters e.g. grading system,
degree classification etc.
view available courses in departments
and programmes
enter results into system
generate reports e.g. mark sheets,
broad sheet etc.

ARTIFICIAL INTELUGENCE
The Encyclopedia of computer science [Ralston, (2000)] defines Artificial
Intelligence (AI) as "a field of computer science and engineering that is
concerned with the computational understanding of what is commonly
called intelligent behaviour and with the creation of artifacts that exhibit
such behaviour". Further exposition on AI reveals that the field can be
viewed in the directions of computational psychology, computational
philosophy and machine intelligence.

Machine intelligence is concerned with specific tasks and how to extend
the frontier of what can be programmed on computers (i.e., knowledge)
about the task. This is what reasoning is. For example, using explicitly
stored knowledge to produce additional explicit knowledge is reasoning.

Njoku (2000) in his doctoral thesis under my supervision, has been
concerned with the task of developing expertise in playing Jonda, an
indigenous game that has rich features for artificial intelligence research.
He has studied the game for the purpose of experimentation in machine
intelligence. The expert system programmed in PROLOG language is a
predictive model that rationally and dynamically selects slot to deal from
without cause to backtracking. Njoku and Ayeni (2000) have established
that in playing the game the degree of competitiveness increases slowly
as the board size is increased and tends to a limit.

COMPUTING CULTURE IN NIGERIA
Acting Vice-Chancellor Sir, I have presented to this august audience
computing culture and what my humble contributions had been in its
various aspects. It has been shown that computing culture entails great
accomplishments in the provision and usage of sophisticated data
processing and data communication equipment with great potential in
improving socio-economic conditions in a society. Standards are being
set in products and professional practice to ensure continued relevance

23



and retention of the culture. We now want to address the state of our
nation in this field.

Education
The Universal Basic Education scheme of the Federal government is yet
to fully come on board. The scheme is a conscious effort to increase
literacy level (which is presently 57.1 % [The world (2000)], UNESCO's
figure is 42%, Federal government's 52%) ofthe average Nigerian. Few
years back, the government planned computer literacy programme that
was not executed in a large scope for one reason or another.

Basic Infrastructures
Electricity is a scarce commodity in Nigeria in most homes and offices.
At any rate, the Federal Government of Nigeria has promised abundant
electricity supply for the nation by the end of 2001.

Some of the telephone exchanges are still analogue, 'e.q., Somolu
exchange, Mainland Ebute Metta Exchange etc. The telephone services
in the areas covered by these public exchanges are at best in fits and
starts just like the supply of electricity in most parts of the country as at
now. The International Telecommunication Union (lTU) prescribes one
telephone to 101)people. Nigeria with a population of over 120m has a
little over 400,000 telephone subscribers (700,000 installed capacity).
This makes one telephone available to about 300 people. A new (revised)
National Telecommunications Policy was approved late 2000. Perhaps
this would usher in many additional fixed and mobile lines in 2001 and
beyond.

Our tertiary institutions still have very much less number of students in
science-based programmes than in non-science ones. Even though there
is Federal government poticythat stipulates 60:40 admission ratio in
favour of science/technology-related disciplines, the Joint Admissions
and Matriculation Board (JAMB) to date hardly observes this ratio. The
1998 JAMB admission statistics of candidates shows only 43% for
science/technology-related programmes [Ogunleye (1999)]. The fear is
that even if our country becomes highly computer literate, majority of us
might end up being at best technology users. To really keep up with the
world in the area of information technology or any other technology for
that matter, our country should pay special attention to science.

Though Nigeria has been connected to the information super highway-
the Internet, the facility is yet to be widely subscribed to.

Stake holders
The Federal government is yet to come up with an IT policy. Personally,
I don't think this is a serious setback for now since the basic infrastructures
are not there yet. Govemment can still be doing things that can enhance
activities in computing. What I consider a serious problem is the lack of
priority given to the Ministry of Science and Technology (MST). If Nigeria
wants to be part of the information age in the twenty-first century, priority
attention must be given to IT infrastructures, products and services. In
budget 2001, only 0.3% of the total national budget (recurrent and capital)
is allocated to the ministry.

Computer Hardware and Software
The number of computers in Nigeria is no doubt small. While South Africa
with population of about 45m has about 25m computers, Nigeria has just
over half a million computers for its over 120m people. To make more
computers available, the organized private sector had taken the initiative
to assemble personal computers (PCs).

Since 1997, the Computer Association of Nigeria (COAN) had been
organizing a yearly Software Exhibition. Among the objectives of this
activity is 'to coordinate the process of ensuring adherence to the
Intemational software development and documentation standard viz: ISO
9000 - 3' [COAN (2000)]. I believe this is a step in the right direction. It
would on the long run ensure that software developed in Nigeria would
have acceptance in other parts of the world.

The Federal government set up Computer Professionals (Registration
Council) of Nigeria (CPN) through Decree no 49 of 1993. This body is the
sole organization that is given the responsibility to regulate the practice
of computing and register practising computer professionals in Nigeria.
Unfortunately, the govemme'nt is yet to adequately fund this body to enable
it perform its lawful duties unhindered. The Computer Association of Nigeria
(COAN) and the Information Technology (industry) Association of Nig ri
(ITAN) have been organizing activities such as seminars, workshop nd
conferences to promote acceptance of computing culture In NI Ii I.

24 25



Nigeri~'s organized private sector (OPS) is presently groaning (The
~uart!,an,!uesday, November 7, 2000) that its investment in PC assembly
ISfacing dlffic.ulties S~Chas heavy operating costs, unstable exchange
rates, delays In clearing goods, high tariffs (10%) on IT products, etc.
Unless government consciously assists this sector, nothing worthwhile
may be achieved by the operators.

low level investment in the IT sector by banks
and financial institutions
present structure, size capacities and ability of
IT companies cannot compete favourably with
the rest of the world.

In the draft copy of the National Report on the Millennium (Y2K) Bug
[FMST (2000a) ] prepared by the Federal Ministry of Science and
Technology (FMST) (I happen to be among the team of experts that
prepared the report), we have it that "Computers in Federal organizations
were mainly for document processing and analysis that will not have
ad~erse effe~t'~me January 1, year 2000. Federal organizations using
mainframe/mini were not many but there were serious software problems,
more than hardware as some of the customized software written in Cobol
dbase and other programming languages that required a substantial
amount in effecting correction". The general impression is that except for
some. oil .companies and very few financial houses in Nigeria, most
organizations use computers 'for document processinq, payroll, billing
and straight forward analysis that may be time-consuming. The computer
applications in many cases are meant to serve as checks on the more
convenient manual processing. The use of expert systems is not in vogue
and computer networks such as client/server systems are not available
in many organizations.

The above lessons sl.ow that as far as computing is concemed, Nigeria
is still in the swamp.

UNIVf ssn j(
CONCLUSION BRA ,
As abstract as computing culture appears to be, one should'emphasize
that it is more than just providing relevant basic infrastructures and setting
standards. It has got a lot to do with people, knowledgeable people, with
right training and perceptions. It is in this light that the government of
India [GI Taskforce (1998)] Jaunched "OPERATION KNOWLEDGE" whose
aim was to "universalize computer literacy" and also to spread the use of
computers and IT in education. Itwas planned to set up a National Council
of IT Education to initiate a 'Teach the Teachers' (3T) programme for
upgrading, on a regular basis, the IT knowledge and skills of teachers.
The overall target is "IT for all by 2008". T0 realize its vision, the government
of India has earmarked 1-3% of the budget of every Ministry or Department
for applying IT in the DepartmenUsector; this investment includes not
only the purchase of software and hardware products but also fortraining
and providing IT services.

The MAIN REPORT in the final NATIONAL REPORT ON THE
MILLENNIUM (Y2K) BUG [FMST (2000b)] summarises the state of
computing in Nigeria as at present. These are the "lessons learnt from
Y2K experience". They include the following:

very low computer awareness
very low computer literacy
ccmputerizatlon at federal, state and local
government levels is abysmally low or non-
existent
weak power infrastructure and poor
telecornmunicatlon facilities
slow capacity building in the education sector
no Information Technology legislation

How does a nation go about cultivating computing culture? We shall
consider just two cases to illustrate the approach. First, China, a
developing country, uses the concept of management support systems
to prepare its state-owned enterprises (the Federal Republic of Nigeria
recently appointed Board members to 137 of such institutions [see Sunday
Tribune, 12 November, 2000]) to achieve its desired goal of becoming a
developed country as inexpensively as possible. Second, we will look at
how some developed countries are reaping the dividends of computing
culture through the application of artificial intelligence.

"The Chinese government regards the use of computer technology as
one ofthe decisive factors in improving China's socio-econornic conditions.
Since the late 1980s, China has channeled sizeable resources to acquire
more sophisticated methods, tools, and techniques for the development
and applications of computer-based information systems (IS)" [Zhi et al.

26 27



(1997)]. In 1990, a guideline was issued which defined a management
support system (MSS) 'as 'a computer-based multi-level, multi-purpose
and multi-function IS'. Such a system can provide overall control of the
operational and managerial activitles of an organization and it can also
provide information to assist rnanaqers to make decisions at various
organizational levels". This type of system is a combination ottransactlon
processing system (TPS), management information systems (MIS) and
decision support systems (DSS).

RECOMMENDATION
It is obvious that any expected improvement in computing culture in Nigeria
would depend on what the Federal Government in particular and Nigerians
in general do or fail to do. The public and private individuals and institutions
have to be mobilized to contribute their own quota faithfully. I, therefore,
make the following recommendations:

In the mid-1980s to late 1980s, networks technology, especially LANs,
developed rapidly in China and most enterprises gradually shifted from

. purchasing application software abroad to self software development,
frequently jointly with universities, vendors and/or research institutes. To
.really get out of the swamp, companies exerted enormous efforts in the
training of its managerial staff. All functional department heads learned
to electronically access information that was not easily accessible
manually thus realizing the huge benefit brought forth by the MSS. All
hardware and software were integrated to form client/server architecture.
China now has a China Wide Web on the Internet.

1. The Federal Government should give tacit and financial
support to the Computer Professionals (Registration
councih of Nigeria (CPN) to enable it pursue, unhindered,
the objectives for which it was set up, among which is
the regulation of the practice of computing in Nigeria .
Also, it should approve without further delay, career
cadres for the computing practitioners in private and
public services to encourage professionalism in the
sector;

In the developed countries, artificial intelligence (AI) is extensively applied
to deliver services to the people. Most applications of At are of a type
called expert systems. An expert system is a computer program that
reasons using knowledge captured and embedded explicitly within the
program to solve complex problems. I give below just a tew applicatlons
of expert systems in Britain and USA [Ralston (2000)].

American Express uses Authorizer's Assistant (AA) to assist a
human authorizer with the decision to allow or disallow a particular
credit arrangement by analyzing a large amount of customer data
from the company's database files.

British Department of Social Security uses expert system in
answering the written queries of citizens concerning their pensions.

US Internal Revenue Service uses Tax payers' ASSistant Expert
System to give accurate tax information in response to telephoned
queries from taxpayers.

2. Govemment should plan towards setting up a world class
information infrastructure with an extensive spread of
fibre optic networks, sat com networks and wireless
networks for seamlessly interconnecting the local
information infrastructure (LlI), national information
infrastructure (Nil) and the global information
infrastructure (GII) to ensure a fast nation-wide onset of
the internet, extranets and intra nets. The Ministries of
Science and Technology (MST) and Communications
should be given this responsibility and to be able to
realize this project as soon' as possible, their annual
budgets should be increased substantially because of
the capital intensiveness of the IT infrastructure. It is
suggested that up to 6% ofthe nation's Gross Domestic
Product (GDP) should be allocated to them; .

3. Tariffs on IT products - hardware and software should
be zeroed until the nation can compete favourably in IT
sector internationally;

The American Airlines use MOCA expert system to schedule
hundreds of aircraft for varying types of required maintenance.

4. A national scheme like China's Management Support
System (MSS) should be put in place to enhance
computerisation of operations in Federal, State and Local

28
29



Governments-controlled enterprises. A sub-scheme that
would entail the setting up of an Artificial Intelligence
(AI)/Decision Support System (DSS) unit in each
Computer or Information Technology department of each
enterprise may be used to trigger off implementation of
comprehensive information systems in these
establishments. Enterprises should be encouraged to
create websites on the Internet;

5. Government should adopt more pragmatic approach to
enforce 60:40 stipulated ratio in favour of science/
technology - related disciplines at secondary and tertiary
levels of study in our educational system. My
recommendation is that special incentives be provided
for people in science/technology related disciplines who
are students or workers. By this way, Nigerians will be
empowered to become producers as well as consumers
of IT products and services of international standards;

9.

6. There is urgent need for IT legislation. This will give legal
backing for enforcement of aspects of it that concern
computing. The current body set up by MST under the
Minister of State to fashion out a suitable IT policy for
Nigeria should be assisted in every way possible to
complete the assignment in shortest possible time;

7. A major source of power for computers is electricity.
Regular supply of electricity should be ensured in all
parts of the country to enhance computing culture.
Telephones (wired, mobile and wireless) are required for
connectivity and computer communication. Government
should evolve economic policies that would make these
facilities available to the public and at affordable prices;

8. Information Technology is a universal tool. Non-producer
nations acquire it to provide solutions to their problems
without really having conducive environment for optimal
application of the technology. For example, lack of
consistent data could prevent the provision of correct
solutions to problems. This partially explains why some

30

highly placed people acquire computers as a status
symbol in Nigeria. I wish to recommend here that
government should assist in establishing interlinkag.e
programmes between local institutions (including ~u~hc
enterprises and educational institutions) and similar
institutions in developed countries where this technology
is optimally used. This will enable local institutions share
expertise with foreign institutions, prepare adequately
for the use of the technology and hence be in position to
improve on their performance and standards;

As a matter of urgency, and in preparation for information
technological take-off, the Federal Govemment of Nigeria
should set up, under the Presidency, a Research and
Development Centre for Information Technology (R & DC
for IT). At this centre, adequate facilities for research
and development of IT products should be provided.
Competent researchers from any part of the world
including Nigeria should be employed and tasked to
contribute meaningfully to the development of Nigeria in
the field of Information Technology.

10. Finally, Mr. Vice - Chancellor, let me come back to base.
It is not at all palatable to remark that most important
computerisable aspects of the University of Lagos
operations are not computerized in this information age-
staff control, resource control, examination processing,
admissions, etc. are still done manually. Although, we
have operations where computers are used to aid the
manual processing via some sort of computerized
transaction processing, this is not enough.

However, one is very much aware of the NUC package,
NUMIS, 'that has been imposed on the Nigerian
Universities in the name of management information
system. For all intents and purposes, NUMIS has not
made any serious impact since it was introduced over 5
years ?go on the administration of the University of La~os
especially in the areas of students' records proces~lng
and financial management among others. Th r might

31



be many reasons forthis inefficientiveness among which
may be resistance to change of procedure in some
quarters. These remarks are just observations, they are
not an indictment on anybody or group of people
associated with the development or use of the package.

However, I am totally convinced that with the autonomy
granted to the University system, a locally produced
MIS that would consider the specifics of our University's
structure and mode of operations would impact more on
the operations ofthe university. I, therefore, recommend
that the University of Lagos should set up immediately
an in-house committee to study and design a suitable
MIS forthe university. NUMIS may be used as a basis
for the study. This process will take care of all
appropriate operations in the University for
computerization.

To start with, I wish to give all the honour and glory to God through our
Lord Jesus Christ for sparing my life and for giving me good health to
witness this wonderful day and for granting me this rare opportunity to
give the Inaugural Lecture. Thank you, JESUS. ft.,

• .l"i... \

My foremost thanks go to my late parents, Mr John Olosu Ayeni and
Madam Felicia Oredola Ay~ni nee Alabi Bobade, the sapetu Okitiko of
Ikere Ekiti. When I reflect on my early days and consider my parents'
struggles financially and spiritually to. educate me (I was a little bit
stubborn) I give thanks to God fortheir lives and forgiving them courage
and ability to see me through to a level I was able to fend for myself
before they joined their ancestors in the 1970s. May their gentle souls
continue to rest at the bosom of Jesus Christ. Amen. I also want to use
this opportunity to express my sincere appreciation to my sisters, brothers
and well-wishers for their unflinching support over the years.

Furthermore, our University should embark on building
a fibre-optic network to serve as backbone for a university----
wide intranet so that staff and students can communicate
with each other on computers. The Intranet would be
connected to the Internet through a firewall. As this
project is both time and money.consuming, a quick,
less expensive but less durable alternative can be
considered as a temporary measure. The current
University telephone network - PABX Hicom300E
installed under my supervision as the Chairman of the
Telephone Committee which has capability for voice and
data communication can be hooked onto the internet
via a suitable transmission equipment. This would enable
the University community have access to the Internet
soonest. I am happy to learn that something is currently
being done along this line by the current administration.

One person I believe God used to. drag me into computing is the first
Professor of Computer Science in Nigeria, late Professor Olasupo J.
Fagbemi. I was a survey assistant on a study leave from the Ministry of
Lands and Housing, Western State of Nigeria to study Mathematics at
the University of Lagos in 1968. Professor Fagbemi was the Head of
Department of Mathematics then. Professor Fagbemi got interested in
me when he observed that I was doing very well in the computer courses
as well as other courses offered in the Department at that time. He lured
me to stay on for a yearto do Postgraduate Diploma in Computer Science
after my first degree and to join the staff of the Institute of Computer
Sciences which had just been created. He was the Dean of the Institute.
Thus, I jettisioned my contract with the Ministry of Lands and Housing. I
have no regrets whatsoever for taking this decision and for taking up
academics as a profession. I thank God for his life. Professor Fagbemi,
a brilliant scholar, an eloquent speaker, a committed academician and a
die hard man for excellence died in a motor accident on Lagos-Ibadan
Express Way (then under construction) shortly after he was retired from
the service of the University of Lagos by the Military Administration of
Muritala/Obasanjo in 1975. May his gentle soul rest in perfect peace.
Amen.

ACKNOWLEDGEMENT
Mr. Vice-Chancellor, Sir, Distinguished Ladies and Gentlemen, I would
like to recognise at this juncture some of the notable people who have
contributed to making me what I am today. If I do not mention your name
specifically, please forgive me, limited time and space given to me have
not allowed me to mention every notable person .•

I must thank brother 'Jisefinni Fasuan, a tailor by profession. God used
him to take care of me for over six months when I was a job seeker after
my Secondary School Certificate Examination in November/December

32



1964. Thank you, Sir. May God always meet you at your point of need.
Amen. In the same breath, I want to thank my good friend of over 35
years, Pastor Gabriel Adekunle Akinyede. I met Kunle at Federal Survey
School in July 1965. We completed our Basic Surveying Examination at
Oyo in May/June 1966. Both of us were posted to Akure Divisional branch
ofthe Ministry of Lands and Housing, Westem State. He arrived at Akure
before me, got accommodated and settled. He accommodated me for a
month before I secured my own place. Since then, our family friendship
had grown from strength to strength. May God continue to uphold you
and your family. Amen.

I must not fail to thank the crew at the Centre for Educational Technology
(CET) who packaged this lecture in a most professional manner and my
able assistant, Alaba Apaguntan, who word processed the original script.
Thank you all.

My thanks also go to the institutions that provide the facilities that mould
my character and knowledge right from the primary school - St. John's
Anglican Primary School, Ikere Ekiti, Ise-Emure Grammar School, lse-
Emure Ekiti; Federal Survey School, Okene; School of Surveying: Oyo,
University of Lagos; University of Waterloo; CISM, Udine; ICTP, Trieste
and C.RAI., Rende, Italy. I thank all the good people I interacted with in
these institutions at one point or another.

I also want to thank my fellow colleagues in the Department of Computer
Sciences, my students overthe years and the non-academic members
of the Department. I thank, in particular, members of my research group,
especially Dr. Charles O. Uwadia and Dr. Dele Longe. I also want to use
this opportunity to express my sincere appreciation to members of the
Council of the Computer Professionals (Registration Council) of Nigeria
(CPN) , Computer Association of Nigeria (COAN), Information Technology
(industry) Association of Nigeria (ITAN) and others not specifically
mentioned. but whose members are here present. May the.dtscipllne of
computing continue to grow from strength to strength. Thank you all.

Finally, I wish to give glory to the Almighty God for giving me the pretty
Bosede Alice nee ornotoso as a wife, a loving mother, a help-mate, a
virtuous woman, "a jewel of inestimable value" to quote the sage, Chief
Obafemi Awolowo. She is very accommodating, trusting and hardworking.
She is presently the Local Administrator of Ifako/ljaye Local Education
District of Lagos State Ministry of Education. Thank you for coping with
my character and for your moral support always. Special thanks also go
to my children - Toyosi, who is presently studying at the University of
Ottawa, Canada; Ayodeji, a graduate·of Computer Science, University of
Lagos. He assisted me along with Mr. Jide Ogundipe, a former PGD
student in my department in retrieving information from the internet in
course of preparation ofthis lecture; Olutosin, the baby ofthe family, is a
level 300 student in the Department of Business Administration, University
of Lagos. Olutosin, equally pretty as her mother, has been very caring.
She wants to know now I feed when I travel to our country home in Ikere
Ekiti. She wants to know why I work so hard always. She wants to know
more, she cares very' much. Children, you are a source of joy to me, I
love you all. May the Almighty God care for you always. Amen .'

My special thanks go to Professor Laz Ekwueme for encouraging me to
go ahead with my chosen topic. Although I put the topic to him as
"Computing as a culture", he assured me it was a worthwhile topic. Thank
you, Sir. My thanks go to my good friend, Professor Kola Ogundowole for
educating me philosophically on the meaning of "culture". Thank you.

I am greatly indebted to Professor Mike Magazine, my Ph.D. supervisor
at the University of Waterloo. He is the main architect of my research
capability. He is unavoidably absent here today but he sent his good
wishes from United States where he presently resides. Thank you, Sir, in
absentia.

Now unto the King Eternal, Immortal, Invisible, the only wise God, be
honour and glory forever and ever. Amen (1 Tim. 117)

Mr. Vice-Chancellor, Sir, Distinguished audience, thank you all for your
attention and God bless.

34 35



REFERENCES
VON NEUMANN AND MORGENSTERN (1947): Von Neumann, J. and
O. Morgenstern, THEORY OF GAMES AND ECONOMIC BEHAVIOUR

Princeton University Press, 1947.

KIEFER (1953): Kiefer, J., Sequential minimax search for a maximum.
Proc. Amer: Math. Soc., volA, pp 502-505,1953.

DANTZIG (1963) : Dantzig, G.B., LINEAR PROGRAMMING AND
EXTENSIONS, Princeton University Press, Princeton, W.J.
1963.

BOHM (1966) : Bohm, C. and Jacopini, G., Flow Diagrams, Turing
Machines, and Languages with only Two Formation Rules,
Comm .. of the ACM, 9, 5 (May), 366-371

FINE (1966) :Fine, T., Optimum search fprthe location ofthe maximum
of a unimodal function, IEEE Trans. Inform. Theory, vol. IT-
12, pp 103-111, Apr. 1966.

DIJKSTRA (1968): Dijistra, E.W., Goto statement considered Harmful,
communcetions of the ACM, vol.tt , 3 (march), 147-
148,1968.

HAMMING (1968): Hamming, RW., One Man's View ofComputer
Science, Journal of the ACM, vol 16, no. 1, January 1969,
pp.3-12.

BAUER (1972) :.Bauer, F.L., and Wossner, H., he Plankalkul of Konrad
Zuse : A forerunner of Today's Programming Languages,

Comm. of the ACM 15,678-685.

SHAPIRO (1974a): Shapiro, RD., Optimal univariate search with unequal
block sizes under minimax and stochastic criteria, Ph.D.
dissertation, Stanford University, Pal. Alto, CA, 1974.

SHAPIRO (1974b): Shapiro, RD. and D.J. Wilde, Variable block search
strategies with prior tnforrnation, Dept of O.R., Tech. Rep.
74-17, Stanford University. 1974.

36

WAGNER (1975): Wagner, H.M., PRINCIPLES OF OPERATIONS
RESEARCH, Prentice Hall, Incor .. 1975.

KERNIGHAN (1976): Kernighan, B.W. and Plauger, P.J., SOFTWARE
TOOLS, McGrawhill Book Company, 1976.

AYENI (1977) : Ayeni, J.OA, search with prior information, Ph.D.
dissertation, University of Waterloo, Waterloo, ON, Canada,
1977.

ODELEYE (1977): Odeleye, A.O. Ayo, A POPULARYORUBA GAME,
Oxford University Press, Ibadan, 1977.

AYENI and MAGAZINE (1979): Ayeni, J.O.A. and Magazine, M.J., A
Heuristic Algorithm for sinqle-variable sequential search with
Prior Information, IEEE Trans. on Sys., Man, and
Cybemetics, vol. SMC-9, no.4, Apr. 1979 pp. 246-250.

AYENI (1980) : Ayeni, J.OA, Automatic Detection of Loop in Stepping-
Stone Method of Solving The Transportation Problem, 15th

Annual Cont. In Stat., Compo SC., O.R. and Math. Vol. 4,
OR Cairo University, Dec., 1980 pp. 1-24

SPRAGUE (1980) : Sprague, RH. Jnr., A framework forthe
Development of Decision Systems, MIS Quarterly 4, no.4,
1980 pp. 1-26.

AYENI (1982) : Ayeni J.OA, Blanks Removal from strings by a Lexical
Analyzer, Journal of Computer Science and its Applications,
voiz. no.1, 1982, pp. 47-52.

AYENI AND MAGAZINE (1982): Ayeni, J.OA and Magazine, M.J., On
Optimal Stopping Rules in Sequential Seraches, Journal of
Computer Science and its Applications, vol, 1, no.1, pp.
20-28, February, 1982.

CATELL (1982) : Cattell, R.b.G., Formalisation and automatic derivation
of code generators, UMI Research Press, Ann Arbor,
Michigan, 1982.

37



GRAHAM (1982) : Graham, S.L., Henry, RR and Schulman, RA, An
experiment in table-driven code generation, SIGPLAN
Notice~ 17,6,pp. 32-43,June, 1982.

REYNOLDS (1988) : Reynolds, G.W., INFORMATION SYSTEMS FOR
MANAGERS, West Publishing Co., 1988.

U. L. ARCHlV
UWADIA AND AYENI (1988) : Uwadia, C.O. and Ayeni, J.OA, A Lag-

for Preprocessor, JCSA, vol. 5, 1, 1988, pp. 59-68.Aho (1983) : Aho, AV., Hopcroft, T.E. and Ullman J.D., DATA
STRUCTURES AND ALGORITHMS, Addison Wesley, 1983.

AYENI (1983) : Ayeni J.OA, Mathematical Programming: A case study,
Nigerian Journal of Management Studies, vol. 2, n01, 1983,
pp.100-111.

OGUNLEYE (1999): Ogunleye, AO., SCIENCE EDUCATION IN
NIGERIA, Sunshine Int Pub. Ltd, 1999.

DENNING et.al. (1989) :Denning, P.J. (Chairman), Comer, D.E., Gries,
D. et.al., Computing as a Discipline, a report,
Communications oftheACM, vol. 32, number 1, pp. 9-23,
January

AYENI (1990) :Ayeni J.OA, A Simulation Model for Timing Road Junction
Traffic Lights, Modelling, Simulation & Cntrol, AMSE, vol
20, no.3, 1990, pp. 53-64.

LONGE (1983): Longe, O.,lfa Divination and ComputerScience,lnaugural
Lecture, University of Ibadan, Ibadan, 1983. AYENI et.al. (1990): Ayeni, J.OA, Longe, H.O.D., Uwadia, C.O. and

Job V.O., The Game Ayo on the Computer, NJS, vol. 24,
1990, pp. 144-145.HENRY (1984) : Henry, RR, Grahara-Glanvllle code generators. Ph.D.

thesis, Computer Science Division, EECS, University of
Califomia, Berkeley; CA, May, 1984. UWADIA (1990) : Uwadia, C.O., Code Retargeting Using A Nucleus Set,

Ph.D. Thesis, Dept. of Computer Sciences, University of
Lago~ Nigeria, March, 1990.TURKLE (1984) : Turkle S., THE SECOND SELF: COMPUTERS AND

:THE HUMAN SPIRIT, Simon and Schuster, New York, 1984.

Aho (1985): Aho, AV. AND Ganapathi, M., Efficient Tree Pattern
Matching: An aid to code generation, ACM Symposium on
Principles of Programming Languages, New Orleans, LA,
January, 1985.

LONGE AND AYENI (1991) : Longe, H..O.D. and Ayeni, J.OA, Ayo
Game: Some Practical Difficulties and Efficiency
Considerations, JCSA, vol.5 no.2, 1991 pp.58-72.

AYENI and LONGE (1985): Ayeni, J.OA and Longe, H.O.D., Games
People Play: Ayo,lnt.Jnr. of Game Theory, Physicaverlag,
Vienna, vol.14, issue 4, 1985,pp. 207-218.

AYENI (1994) : Ayeni, J.OA, A Micro-based prototype Students'
Administration System, Research Report no. CRS 88/04,
University of Lagos, Akoka, 1994.

AYENI (1988a): Ayeni, J.OA A Computer Program for solving
Transportation Problems, JCSA, vol. 5, 1, 1988, PP. 89-
102

MALLCH (1994): Mallach, E.G., UNDERSTANDING DECISION
SUPPORT SYSTEMS AND EXPERT SYSTEM, IRWIN,
1994.

AVENI (1988b) :Ayeni, J.OA, On Timing Traffic Lights at a Road Junction,
-Modelling, Simula(ion & Control, C, AMSE, vol.' B, no 4,
1988, pp. 24-41.

UWADIA and AYENI (1994) : Uwadia, C.O. and Ayeni, J.OA, A
methodology for obtaining the nucleus set of a group of
instruction sets, /fe Journal of Technology, OAU, vol., n01,
1994.

38 3



AIMUFUA AND AYENI (1996) : Aimufua U.E. and Ayeni, J.OA., A Time-
Table Improvement Algorithm, JSCA, vol.2, 1995/96, pp. 145-
154.

NJOKU AND AYENI (2000) : Njoku, C.N. and Ayeni, J.O.A., A theory of
Jonda, Technical Paper, Department of Computer Sciences,
University of Lagos, Nigeria, 2000.

AYENI AND UWADIA (1996): Ayeni, J.OA and Uwadia, C.O., A Micro-
based Prototype Students' Administration System, JSRD,
vol.2, 1995/96, pp. 162-171.

RALSTON (2000) : Ralston, A., Reilly, E.D. and Hemmendinger, D. (eds.),
Encyclopedia of Computer Science, Fourth Edition. Nature
Publishing Group, 2000.

ZHI et.al, (1997) : Zhi, YC., Wang, P. and Turban, E., Management Support
Systems of State-Owned Enterprises in China, Int. Jnf. Of
Information Management, vo1.17, no.4, pp. 271-285, 1997.

THE WORLD (2000): w.w.w. dCi.gov/ciaipublicationsifactbookigeos/
ni.htmI2000.

GI TASK-FORCE (1998) : Government of India IT Taskforce, Information·
Technology Action Plan, http://it-taskforce.nic.inlit-
taskforce, 1998.

AYENI AND BAMGBOSE (1999): Ayeni, J.OA and Bamgbose, A.P., A
Faculty Students' Information System, JSRD, vol4, 1999,
pp.137-154.

OGUNLEYE(1999): Ogunleye, A.O. ,SCIENCE EDUCA.TION IN
NIGERIA, Sunshine In1. Pub. (Nig) Ltd, 1999

COAN (2000) : THE COMPUTER ASSOCIATION OF NIGERIA (COAN),
NIGERIA SOFTWARE EXHIBITION-NISE 2000, Brochure,
March/April 2000.

FMST (2000a) : Federal Ministry of Science and Technology, NATIONAL
REPORTPN THE MILLENIUM (Y2K) BUG, DRAFT, vol.3,
Abuja, Nigeria, 2000

FMST (2000b) : Federal Ministry of Science and Technology, NATIONAL
REPORT ON MILLENIUM (Y2K) BUG, vol. 1, MAIN
REPORT, May, 2000.

NJOKU (2000) : Njoku, C.N., Learning System Dynamics through
Knowledge-base Updating, Experimantal Study with Jonda,
Ph.D. thesis, Department of Computer SCiences, University
of Lagos, 2000.

40 41