Declan McCarthy became a recruitment consultant in 1979 and spent many years advising international technology companies on staff selection for their operations in Ireland. He was also a non-executive director of Glockenspiel, the object technology company that did much to change the orientation of the Irish software industry.
His early career included roles with Bank of Ireland and with computer vendor ICL in Poland and Ireland. It took off, however, when the Department of Finance ran aptitude tests to choose civil servants for training as systems analysts.
When I went to work in the Department of Social Welfare in 1960, the first piece of office technology that I encountered was an Adrema addressing machine. This device printed the labels that we put on the front of peoples’ allowance books. I started in the pensions section, then moved to benefits.
We worked in a large open plan office. One side of the room dealt with claims. On the other side were the people who did all the donkeywork. The big advantage of having everyone in one big pool was the sharing of experience. If a difficult issue arose, somebody in the room was usually able to suggest a precedent.
In 1964 I began studying for a BComm degree on a night course at University College Dublin, but I never completed it. After two years I was offered a place on a masters level course at the School of Public Administration. Ours was the first group to be educated in this way.
The masters course was intensive with 800 lectures in one year. The school brought in lecturers from many universities and places like Sandhurst in England and Stanford in the US helped to design the programme. What was happening here was that the civil service wanted to grow its own high fliers. We were the beneficiaries of this plan.
I returned to Social Welfare with a degree-level diploma in public administration and started to work in the scope department. This was where legal decisions were made on the insurability of employments.
All the client records were still kept in physical files and a new file was opened whenever someone applied for benefits. There was no cross-referencing of social welfare numbers with any other public service records. The only form of data storage for this information was a microfiche library. In fact, the only computer installation in the civil service was run by Revenue, which at this time was replacing an ICT machine with a new Honeywell system.
The Department of Finance was, however, developing a strategy to introduce computing into more departments. In July 1967 it held a series of aptitude tests in order to select trainee systems analysts from inside the civil service. We were given easy tests in the morning and the questions became more difficult in the afternoon. I did better as the day went on. In the end I got one of the top scores in these tests, along with Michael Kelly. We were both promoted, transferred into the Department of Finance and sent immediately to the University of London, where IFIP had initiated a ‘seminar in administrative data processing’ as part of an intergovernmental co-operation programme.
There were about 25 participants in the IFIP programme – drawn from governments all over the world that were in the early stages of computerisation. We started with a residential course that ran from July to December and covered almost as much material as a full degree in computer science. This included a lot of systems analysis and programming in multiple languages. The university gave us accommodation, but IFIP brought in all the lecturers. It was like a parade of people from different countries that had practical experience of computing in public administration.
In addition to the lectures we were offered work experience. My placement began in October 1967 at the Rijks Kantoormachine Centrale in Den Haag, where I was shown how the Netherlands public administration had computerised a variety of services. The examples included telephone billing and a chain of savings banks that was run by the Post Office.
I returned to the Department of Finance in January 1968. The department was still a long way from installing its first computers, but it had already established a Central Data Processing Service (CDPS).
Maurice O’Connell was the principal officer in charge of CDPS, supported by two assistant principals – Gerry Colgan and Michael Cullinane. The group expanded steadily by selecting more candidates through aptitude tests inside the government departments. Soon we had a separate programming section.
Our first practical exercise was the computerisation of the army payroll on an IBM System/360 model 20 that the Department of Defence had installed in Glasnevin. Then we did an information requirements study for a national forestry scheme that would provide incentives for people to plant more trees.
Another project took us into the Paymaster General’s office, which handled all the cheque payments in the civil service. They used to verify all these payments by hand, matching cashed cheques against addresses. We automated this process by introducing a checksum digit on pay cheques. This stopped anyone from photocopying their cheques and cashing them more than once.
CDPS used the Revenue’s computer for this job, working on its Honeywell system at night. The machine occupied a large computer room in O’Connell Street – a typical mainframe installation of the time. We used it other occasions as well, usually for work that involved financial oversight, such as a cheque reconciliation project for the Department of Social Welfare.
Code sharing was remarkably casual in the 1960s. The computer companies charged customers for hardware, but not for software. So programmers passed around their work quite freely.
For example, I came back from the Netherlands with one large printout of the code for a savings bank scheme and another printout with a system for telephone billing. My Dutch programmer pals said that I might as well take them home with me. I didn’t do anything with these applications, but in 1968 I applied unsuccessfully for a computer manager’s job at Posts and Telegraphs. If I had got it, I might have done something with the software.
A similar situation arose later on in Scotland. The Department of Health wanted to analyse information about hospital in-patients for epidemiological research and for its administrative planning. This project was not welcomed by some hospital managements and was slow to get moving. The department therefore sought assistance from its counterpart in Britain. After a night out with a group of Department of Health and Social Security programmers in Edinburgh, I got a copy of the software that they had used for similar purposes. I handed it over to the Medico-Social Research Board.
In 1969 the board started to implement the software that I had brought back from Edinburgh. It subsequently began to collect hospital discharge summaries, but I was not invited to the launch of its service.
I was an ideas man and bandied ideas around openly. Sometimes they were acted on. I suggested, for example, that the county councils might share computing resources. I had seen how the Dutch developed new systems in a central place and distributed them to local administrations. Gerry Colgan picked up on this idea and wrote a discussion paper. That led eventually to the formation of the LGCSB.
One of the big ideas inside the public service during those years was multi-annual budgeting. I recall an event on this topic at the Blessington Hotel, where an American expert explained its implications. All the departments sent senior managers – secretaries and assistant secretaries – to hear this speaker. I was the only systems analyst there. I ended up having a one-to-one discussion with the US expert.
Soon afterwards, however, I left the civil service. I joined the Bank of Ireland in 1970 to build management information systems on its first ICL computer. Three years later ICL hired me as a project manager. I became a troubleshooter for the company behind the Iron Curtain – first in Croatia and then in Poland. I continued working for ICL when I moved back to Dublin.
Today I probably know Poland better than I know Ireland and I can still speak a fair bit of the language.
Last edit: November 2016
© Declan McCarthy 2016