From Batch to Real-Time—My Journey into Sri Lanka’s Computing Frontier (1980–1982)

From WHO to Weather: My Early Years Applied Computing (The Data General S/140 is an exhibit item now at the University of Colombo School of Computing Museum) (Photo Archive Library (c) Turrance Nandasara)

Initial Appointment — WHO "Choice of Medicine" (1980–1981)

When I first stepped into the world of computing in January 1980, I had no idea how quickly things would evolve. Just after my graduation convocation on 19 December 1979, I was appointed as a Research Assistant at the Department of Pharmacology, Faculty of Medicine, University of Colombo. This was for the World Health Organisation’s project on “Choice of Medicine in Sri Lanka”, running from 7 January 1980 to 31 January 1981.

My job? To apply mathematics and statistics to real-world health data—designing surveys, crunching numbers, and writing FORTRAN IV programs to make sense of it all. First, I used an IBM System 360/Model 25 Computer for data analysis.

I was fortunate to apply my knowledge of mathematics and statistics under the guidance of Prof. N. D. W. Lionel. My responsibilities included:

• Designing and preparing questionnaires
• Collecting and tabulating survey data
• Writing FORTRAN IV programs for statistical analysis
• Reporting results to WHO through the Department of Pharmacology

The IBM 360/Model 25 mainframe was housed in the Department of Statistics and Census of Sri Lanka, where I utilised it before the University obtained its own NOVA/4 minicomputer.

---

“From punch cards to interactive terminals — a leap in computing efficiency.”

“This was my first-hand experience of using a non-batch processing computer instead of a batch processing computer for data analysis.

---

During this period, I received an all-inclusive allowance of Rs. 700, equivalent to the salary of a Temporary Assistant Lecturer. I also supervised data entry, editing, and processing using FORTRAN IV & V, Business BASIC, and Assembler on the Data General NOVA/4 minicomputer at the Department of Mathematics of the University of Colombo.

Transition to the NOVA/4 Minicomputer

Before the University acquired its own NOVA/4, I used the IBM 360/Model 25 at the Department of Census. Once the NOVA/4 was installed at the Department of Mathematics, I managed all the WHO project computing there.

The NOVA/4 minicomputer — my introduction to interactive, non-batch computing.
Image Courtesy: pdp-11.nl/dg/nova4/nova4.html

---

“A machine that changed the pace of research at the University.”

---

Back then, we relied on the IBM 360/Model 25 mainframe at the Department of the Census. It was batch processing—slow, rigid, and unforgiving. But soon, the Department of Mathematics installed the Data General NOVA/4 minicomputer, and everything changed. I had full control over this machine and later mastered the ECLIPSE S/140 as well. These systems allowed real-time interaction, and I dove into programming with FORTRAN, Business BASIC, and even Assembler.

Second Appointment — Climate and Crop Data Analysis (1981)

In February 1981, after Prof. Lionel’s dismissal, Miss Savithri Abesekera, Lecturer-in-Charge of the Statistical Unit, invited me to join another project: “Analysis of Climatic and Crop Data for Agricultural Purposes”, funded by the National Science Council of Sri Lanka.

This was a collaborative venture between:

• University of Colombo
• University of Reading, UK
• Department of Meteorology, Sri Lanka

Key collaborators included Mr E. K. Senevirathne (later awarded a PhD from the University of Reading) and Miss Ann Leaker (UK), with Prof. V. K. Samaranayake overseeing the project locally and Dr Roger Stern (University of Reading) providing international supervision.

---

We developed a central database comprising over 30 years of daily rainfall and climatic data—and in some cases, more than a century's worth—collected from 100 meteorological stations across the inland region. Each recorded report represented a crucial data point for shaping the nation's future.”

---

Looking back, those years weren’t just about mastering machines. They were about laying the foundation for interdisciplinary computing in Sri Lanka—bridging medicine, agriculture, and education through data. And I was proud to be at the heart of it.

Building the National Climatic Database

My role evolved from data collection to supervision, coordination, and computerisation of large datasets. This included:

• Data entry, editing, and verification
• Writing tabulation programs for analysis
• Using NOVA/4 and later Data General ECLIPSE S/140 minicomputers

Third Appointment — Predictors of University Performance (1981–1983)

In November 1981, I was appointed again as a Research Assistant for Prof. Samaranayake’s project on “Predictors of University Performance”. By December 1982, we had fully transitioned to the Data General ECLIPSE S/140 system.

One major reason for selecting this configuration was to run internationally recognised statistical software:

• GLIM (Generalised Linear Interactive Modelling)
• MINITAB (Statistics Package Developed at the Pennsylvania State University)
• SPSS (Statistical Package for the Social Sciences)

These tools became invaluable for both the Climatic and Crop Data project and other consultancy services offered by the Statistical Unit.

The Data General ECLIPSE S/140 — enabling advanced statistical software at the University. (Video Archive Library (c) Turrance Nandasara)

---

“From raw data to refined insight — powered by new computing capabilities.”

---

Closing Reflection

These early years were more than just technical milestones — they were formative experiences that shaped my career in computing, statistics, and interdisciplinary research. Each project deepened my understanding of how data, technology, and collaboration can transform national research capacity.

How the University of Colombo Entered the Minicomputer Age

May 1980. Room C8 wasn’t ready yet — but history rarely waits for the paint to dry.

The small space where the IBM 029 card punch had once clattered to life was about to welcome a very different sound. Engineers unboxed a Data General NOVA/4 minicomputer, the first of its kind in Sri Lanka, loaned to the Department of Mathematics while the permanent system was being prepared.

First encounter with the future — seated at the NOVA/4 terminal, 1980, as the minicomputer era arrived at the University of Colombo. (Photo Archive Library (c) Turrance Nandasara)

At the time, few in Sri Lanka had even seen a minicomputer in person. This was the first hands‑on experience for both students and staff, marking the start of a cultural and technological shift within the Department of Mathematics.

It had taken years to get here. In the 1970s, Sri Lanka’s strict import restrictions meant that even pocket calculators were banned. As neighbouring countries leapt ahead in computing, local academics had to make do with scarce, outdated tools. That changed after 1977, when a new open‑economy policy allowed technology to flow in. IBM, WANG, ICL, HP — all began appearing on Sri Lankan soil. But Data General? Not yet.

Negotiations stretched across continents. Data Serve Pvt. Ltd., a telecommunications firm, became the sole local agent for Data General after an agreement with the US company. Dr. Roger Stern’s proposal to Professor P. W. Epasinghe sealed the deal: the University would get a system powerful enough to run world‑class statistical software like SPSS, GLIM, and MINITAB.

Two generations in one frame — the temporary NOVA/4 loaner and the permanent Eclipse S/140 that redefined our computing capacity, 1982. (Photo Archive Library (c) Turrance Nandasara)

The NOVA/4 had kept the department moving while negotiations and shipping delays tested everyone’s patience. The arrival of the Eclipse S/140 signaled the transition from borrowed capability to full independence.

The permanent machine, a Data General Eclipse S/140, arrived in 1981. Its specs were serious for the time:

• 256KB RAM
• Dual 25MB disc drives
• Time‑sharing and multi‑user capability
• A magnetic tape drive
• Eight terminals (three in the original order, plus five added thanks to extra funding)

Embracing the future—Dr. Roger Stern at the University of Reading, along with staff from the Department of Mathematics, explores sharing and statistical software on the Data General ECLIPSE S/140, technology once considered unattainable.(Photo Archive Library (c) Turrance Nandasara)

For many, this was their first encounter with SPSS, GLIM, and MINITAB. Training sessions turned into informal collaborations, as faculty discovered what was possible with shared access to a powerful machine.

The funding came from multiple allies: the Netherlands Universities Foundation for International Cooperation (NUFFIC), the University Grants Commission, and equipment votes from the Faculty of Science and the Department of Sociology.

By April 1982, Room C8 was transformed into the Statistical Laboratory — no longer defined by the click of keypunches, but by the quiet hum of multi‑user computing.

Room C8 transformed — from a keypunch corner to a fully‑fledged multi‑terminal computer room, April 1982.(Photo Archive Library (c) Turrance Nandasara)

The hum of multiple terminals replaced the clatter of keypunches. Ergonomic furniture, improved lighting, and carefully planned wiring made C8 not just functional but a showcase of modern academic computing in Sri Lanka.

---

“We moved from working around limitations to working through them — with the right tools at last.”

---

From Loan to Legacy: How Data General Arrived at Colombo

Setting the Scene – Colombo’s Computing Gap in 1980

The introduction of the NOVA/4 revolutionised. The author employs the NOVA/4 Mini Computer for the "Choice of Medicine Project" of the Faculty of Medicine, University of Colombo. (Photo Archive Library (c) Turrance Nandasara)

In the mid‑1970s, computing at the University of Colombo was still a side note. Students learned FORTRAN on paper, imagining the machine’s inner workings.

By the late ’70s, a single Computer Programming course unit existed, but there was one big problem: no computer on campus.

---

“We were learning to program without ever touching a machine.”

---

Generous offers of mainframe time from the Department of Census and Statistics and the State Engineering Corporation helped, but demand far outstripped availability.

The Statistical Unit Steps Up

By 1976, the Statistical Unit was teaching, consulting, and researching in Applied Statistics. It expanded courses, introduced Statistics as a full subject, and launched a Postgraduate Diploma.

Then came a breakthrough: through UK ODA funding, the Department received HP‑25 programmable calculators and the HP 9825A desktop minicomputer — complete with card reader and printer‑plotter.

Why the HP 9825A Mattered

• 6.8 KB RAM (expandable to 31.4 KB)

• FORTRAN, BASIC, and HP’s own HPL language

• Real‑time calculations on a QWERTY keyboard

• First true in‑house computing at Colombo

The Short‑Lived HP Era

By 1979, the HP 9825A was gone — shipped back to the UK for repairs and never returned. Without local servicing, it had become unreliable.

---

“We rolled back into the no‑computer era.”

---

From mid‑1979 to mid‑1980, Colombo was again without even a programmable calculator.

The Data General Proposal

In April 1980, Dr. Roger Stern wrote to Prof. P. W. Epasinghe:

“After Sam’s research in England and ours here, I feel that a Data General Computer System is the best option for us.”

The preferred choice: Data General Eclipse S/140 — with 25 MB disk, 3 VDUs, dot‑matrix printer, magnetic tape, and FORTRAN V, Business BASIC, and SPSS. Price tag: £101,950.

The Loan That Changed Everything

Before the purchase, the Department secured a one‑year loan of a Data General NOVA/4 in mid‑1981. Compact yet powerful, it introduced staff and students to interactive, multi‑user minicomputing — a leap beyond batch‑processed mainframes.

By December 1981, the University had agreed to buy the Eclipse S/140. The NOVA/4 stayed in service until the new system was installed in April 1982.

Daily Use & Impact

The NOVA/4 transformed daily work. Students and staff could now key in programs directly at terminals, see results instantly, and explore applications in statistics, mathematics, and beyond.

The NOVA/4 console and terminals provided a more immediate and interactive computing experience than the batch-processed IBM 360.

The Transition – 1981

The loan period proved one thing: the University needed its own dedicated system. Committees reviewed options, funding sources were identified, and negotiations began.

Funding came from:

• Netherlands Universities Foundation for International Cooperation (NUFFIC)

• University Grants Commission (UGC)

• Faculty of Science Equipment Vote

The decision: purchase a Data General Eclipse S/140 with:

• Time‑sharing, multi‑user interactive capabilities

• 25 MB disc storage

• 128 KB RAM

• Magnetic tape drive

• Eight terminals

One major goal was to run third‑party mathematical and statistical software such as GLIM, MINITAB, and SPSS — all successfully implemented.

In 1981, the University of Colombo acquired its own Data General system — a leap from a temporary loan to a permanent computing asset. (Photo Archive Library (c) Turrance Nandasara)

My Personal Contribution

After graduating in January 1980, I joined the Faculty of Medicine as a Research Assistant, working with Prof. N. D. W. Lionel on WHO projects. Initially, all my programming was done on punched cards for the IBM 360 at the Department of Census and Statistics.

Later, the arrival of the NOVA/4 at the Department of Mathematics changed everything — no more endless card punching. I could enter survey data directly at the console and run programs in FORTRAN IV & V, Business BASIC, and Assembler.

---

“The NOVA/4 made my computing journey easier than ever before.”

---

After Prof. Lionel’s passing, I joined the Statistical Unit as a Technical Assistant under Prof. V. K. Samaranayake. There, I worked on major projects, including the Analysis of Climatic and Crop Data for Agricultural Purposes — building a central database of over 30 years of daily rainfall and climate data from more than 100 stations across Sri Lanka.

Legacy

The arrival of Data General systems at the University of Colombo marked a turning point — from paper‑based programming and remote compilation to interactive, on‑site computing.

It laid the foundation for:

• The University’s expansion into full‑fledged Computer Science teaching and research

• The Statistical Unit’s emergence as a national consultancy resource

  ---

“From loan to legacy, the Data General era transformed Colombo’s computing story.”

---

Racing the Clock: My First Big Computing Challenge on the IBM 360

Thurston Road (today - Kumaranathunge Munidasa Mawatha), University of Colombo, late 1970s — the place where my thesis journey began. (Photo Archive Library (c) Turrance Nandasara)

From the Colombo Campus to the Department of Census and Statistics, each location marked a step in a race against time.

It was the late 1970s, and computing in Sri Lanka was still in its infancy. Mainframes were rare, expensive, and intimidating machines—giants that lived in air‑conditioned rooms, guarded by operators in white coats. As a final‑year student, I never imagined I’d get to work on one. But fate—and an ambitious thesis—changed that.

---

“The survey involved 2,000 houses, hundreds of variables, and thousands of data points — far beyond the reach of any calculator.”

---

My project carried a long, serious title: Multiple Regression Analysis for a Survey on Housing Conditions in Sri Lanka. It was supervised by Dr. Seneviratne and guided by Prof. V. K. Samaranayake. The survey itself was massive: every detail of each house was recorded — from roof and wall materials to household size, income sources, and even the surrounding environment.

IBM System/360 Model 25 — similar to the machine installed at the Department of Census and Statistics in the 1970s, that transformed my thesis from a calculator exercise into a computing project — and my career along with it.

Did You Know? — Computing in Sri Lanka, Late 1970s

• IBM 360 Arrival – The IBM System/360 series was introduced globally in 1964, but Sri Lanka’s first units arrived much later. One was installed at the Department of Census and Statistics, and later, in other government departments.

• Air‑Conditioned Sanctuaries – Mainframes required strict climate control; even a small temperature rise could cause system failures.

• Punch Card Culture – Each card held exactly 80 characters. Dropping a deck could mean hours of painstaking re‑sorting — or starting over from scratch.

• Limited Access – Machine time was precious. At the University of Colombo, students could submit punched cards for compilation free of charge, but had no direct access to operate the computer. Special arrangements — often at odd hours — were needed for hands‑on use.

• The Human Factor – Operators, programmers, and supervisors formed close‑knit communities. Mentorship, collaboration, and trust were as important as the hardware itself.

The Access Problem

The Department of Census and Statistics housed an IBM 360 Model 30 — among the most powerful machines in Sri Lanka at the time. But there was a catch: students could only use it for learning purposes by submitting their FORTRAN IV code as a deck of punched cards for compilation.

Any other work was charged at commercial rates — about Rs. 750 per hour, roughly a month’s salary for an Assistant Lecturer.

---

“Rs. 750 an hour — more than I could dream of paying, and impossible to estimate how many hours I’d need.”

---

When I explained my situation to Prof. V. K. Samaranayake, he devised a brilliant workaround: the Department of Mathematics would take on my project as a consultancy assignment, funded from its own consultancy budget. My supervisor, Dr. Kevin Seneviratne, took full responsibility for making it happen.

It was an extraordinary gesture of support — one that found its way into the 1979 Annual Report of the Statical Unit of the Department of Mathematics and remains etched in my memory as a turning point.

Each card is a single line of FORTRAN IV — precision, patience, and the constant fear of dropping the deck. Programming was as much a physical craft as an intellectual one.

Punch Cards and Patience

With access secured, the real work began. In those days, programming meant first writing your code on paper, then punching it into stiff rectangular cards using an IBM Model 029 Card Punch Machine available at the Department of Mathematics. My program, written in FORTRAN IV, eventually grew to about 1,500 lines — spread across more than 5,000 cards, including the dataset.

---

“No one got a large program right the first time — debugging meant re‑punching cards again and again.”

---

The process was slow and unforgiving. You would send your carefully ordered stack of cards to the mainframe centre at the Department of Census and Statistics, wait for the compilation, and then receive a thick bundle of continuous‑form paper with your program listing and error messages. Sometimes, the error reports alone ran to more than 200 pages.

Prof. V. K. Samaranayake and Dr. Kevin Seneviratne — mentors whose support opened the mainframe’s doors and made the improbable possible. (Photo Archive Library (c) Turrance Nandasara)

Two Months of Work… and a Friday Surprise

The process of computing dragged on for more than two months. Finally, on the very last working day before my thesis deadline, I got the complete results. It was a Friday. The thesis had to be submitted by 4:00 p.m. on Monday — or it would be rejected outright.

“Two days to analyse, write, chart, type, bind — and deliver. Alone, it was impossible.”

Kurunegala — where my thesis was bound in hard cover before the final dash to Colombo. A quiet provincial town that became the last checkpoint in a two‑month sprint.

Calling in Reinforcements

I called a school friend who had just started his first job as a bank manager. Without hesitation, he stepped in to help. He arranged a typist, an artist to draw charts, and a vacant room at the bank to serve as our “thesis war room.” We worked through the weekend, fuelled by urgency and determination.

The Final Dash

By Monday morning, the thesis was complete. I travelled to Kurunegala to have it bound, then caught the bus to Colombo — 120 kilometres away. The journey took three and a half hours. I arrived just in time to hand it over to the University administration, minutes before the deadline.

---

“Minutes to spare — and two months of work finally in the hands of the examiners.”

---

From IBM System 360/Model 25 to mentors — the legacy of a project that bridged disciplines, decades, and the gap between opportunity and determination. (Photo Archive Library (c) Turrance Nandasara)

Looking Back

That project was more than just an academic requirement — it was my initiation into the world of computing. The IBM 360, the punch cards, the endless debugging, the last‑minute race to the finish line — all of it taught me lessons I’ve carried throughout my career: persistence, creative problem‑solving, and the power of teamwork.

In an era when computing power was scarce and expensive, it wasn’t just the machine that mattered — it was the people who believed in you, and the determination to see the job through.

Learning to Code Without Computers

By Turrance Nandasara

In 1975, I entered the University of Colombo expecting to study the Arts. What unfolded instead was a journey into a bold educational experiment—the Development Studies Special Degree programme. It was a time of reform, of ambition, and of improvisation. We were students of science without laboratories, programmers without computers.

The programme was born from a simple but powerful truth: many students in Sri Lanka, especially in rural areas, had no access to science education. Dr. Jayaratne’s report laid bare the disparities, and the Development Studies degree was designed to bridge them. Its foundation year gave us mathematics, general science, and English. Then came the job-oriented streams—Statistical Services, Fisheries, Tourism. Additional degree-level courses were introduced in Taxation, Estate Management, Valuation, and Education. I was selected for the Statistical Services stream through an internal mathematics test. That moment changed the course of my life.

The pioneering third batch of Development Studies students—an experiment in interdisciplinary education.

We were taught by giants. Prof. V. K. Samaranayake, Dr. Roger Stern, Prof. P. W. Epasinghe, Dr. Saviti Abesekara, and Prof. Prof. A. D. V. de S Indraratna—each brought not just knowledge, but vision. Dr. Kevin Seneviratne’s tutorials in the C8 Lecture Room, with chalk and blackboard, remain etched in memory. Dr. Stern’s hand-drawn charts and diagrams were our version of PowerPoint—decades ahead of their time.

The C8 Lecture Room, where chalk and imagination substituted for screens and simulations.

But the most surreal part? Learning programming without ever seeing a computer. We studied FORTRAN IV through lectures alone. Prof. Samaranayake used an overhead projector and pre-written transparencies at the New Art Theatre. We imagined memory units and logic gates, wrote code on paper, and mentally compiled it. Our brains were the processors; our pens, the output devices.

Prof. Samaranayake’s teaching tools—transparencies and overhead projection—brought computing to life without machines.

Eventually, I had the chance to work on the IBM 360 Model 30 at the Department of Census and Statistics. My final-year project—Multiple Regression Analysis for Survey on Housing Condition in Sri Lanka—was supervised by Dr. Seneviratne and guided by Prof. Samaranayake. The survey involved hundreds of variables and thousands of data points, far beyond the reach of manual calculation or even advanced calculators. It demanded a proper computer programme written in FORTRAN IV, and the only viable input method was punch cards—each card representing a line of code or a data record. The sheer volume of data and the complexity of regression analysis made it clear: computing wasn’t just a subject we studied—it was a necessity we had to master.

The IBM 360 Model 30—Sri Lanka’s gateway to modern computing in the 1970s.

The punch card process itself was a lesson in precision and patience. Each card had to be typed meticulously, with no room for error. A misplaced character meant rerunning the entire batch. Debugging was a physical task—sorting, re-punching, and waiting for the machine to compile and execute. It was programming with your hands, your eyes, and your nerves.

Card Punch Machine and Punch cards—each one a fragile line of logic, typed and sorted by hand.

The Development Studies programme was discontinued in 1977, a casualty of political change. But its spirit lives on. It gave students like me a chance to dream beyond boundaries, to learn without tools, and to build futures from imagination and grit.

I share this not just as history, but as a reflection. Sometimes, the most powerful learning happens when resources are scarce but minds are open. And sometimes, the legacy of a programme isn’t in its duration—but in the lives it transformed.

Sidebar Timeline: Milestones in a Reformist Journey

Year	Milestone
1975	Entered University of Colombo expecting to study Arts; selected for Development Studies programme.
1975	Internal mathematics test led to placement in Statistical Services stream.
1976	Learned FORTRAN IV programming through lectures—no access to computers.
1976	Attended tutorials in C8 Lecture Room; learned from Prof. Samaranayake, Dr. Stern, and others.
1977	Conducted final-year project on Housing Conditions using IBM 360 Model 30 and punch cards.
1977	Development Studies programme discontinued due to political changes.
Post-1977	Continued academic journey, eventually contributing to national computing standards and education reform.