I always enjoy your contributions, they have a roundness and completeness to them that is satisfying, rather like a chocolate eclair and it is not until the last bite is swallowed that one really begins to wonder what was in the middle.
The sad fact is that the Atlas Computer Laboratory died because it had not perpetuated itself or established a good reason for continued existence.
Your article lucidly explains the special national facility that the Laboratory provided in its early days. At the time it housed the biggest machine in the world and the only machine in the country capable of undertaking certain types or work.
However, it lost its supreme position as other manufacturers produced more powerful machines. With the advent of the ICL 1906A it accepted a role which undoubtedly contributed to the development of this machine, but in so doing the laboratory's strength was sapped and its position eroded.
By 1972 its role had declined. Forgive me for taking a crude view of the situation, but at that time it was providing two services. It was running a little research work for Universities and Research Councils, much of which could not be accommodated on the local machine. The reason for this was often said to be that the worker could not convince his own University that the time was justified. The laboratory also acted as a booking agency for 20% of the Rutherford IBM 195.
At this point in time, quite rightly, people were asking the question as to whether the cost of the facility could be justified. I believe Jack was well aware of the position and was actively endeavouring to make the case for a big new machine to return the laboratory to its former glory but this was bedevilled by the money situation, and the promises of the big machine of British manufacture. The question of Jack's replacement and the movement to Daresbury were all indefinite and ill-defined because of the underlying problem of identifying a clear cut and acceptable role for the laboratory.
As times became harder and even the physicists were reduced to having to make do with just that which they could demonstrate they needed, so Jack's attempt to obtain a new big machine was finally discarded.
There were those who strongly believed that the country could and should afford a special national facility for the biggest, the fastest, the best of something in computers. But the problem was to identify the something.
You mentioned two possible contenders, the National Interactive Computing Campus and the Federal Computing Campus but neither of these were sufficiently special and convincing to carry an overall majority. There was another possibility discussed at the time, it was to put the proposed South West and the Oxford big new range machines together in the Atlas laboratory. It had merit as it would have saved building costs at the separate sites, and it would have provided a unique and potentially exceedingly powerful twin processor system. Both University groups concerned like any enlightened parent, were prepared to extol the virtue of all sorts of modern ideas - except where their own daughter was concerned!
In mourning the Atlas laboratory you highlight the inability of those currently connected with Computer Research to dream big, and to have the conviction to go after their dreams.
We all know that computer hardware has already come down so much in price that the problem is no longer a question of finding suitable equipment, but a question of using it. In the commercial world we are just beginning to see some advances in the techniques of using computers. How much we would like to see an Atlas laboratory pushing forward the boundaries of computing. It is unlikely that anyone University (one possible exception) could or would take on the work. Generally speaking to do anything significant in computing you need considerable resources and this means manpower, the sort of manpower that is not available to research in Universities. One man and a boy may be able to think about computing, and lecture about computing, but they can no longer write a useful programme that has not been written many times before. This is something which those responsible for research are reluctant to accept.
What have we in mind as tasks worth undertaking at an Atlas Laboratory
Another Phoenix!
Yours faithfully,
Alan Duncan
