oward Aiken’s contributions to the development of the computernotably the Harvard Mark I (IBM ASSC) machine, and its successor the Mark II — are often excluded from the mainstream history of computers on two technicalities. The first is that Mark I and Mark II were electro-mechanical rather than electronic; the second one is that Aiken was never convinced that computer programs should be treated as data in what has come to be known as the von Neumann concept, or the stored program.
It is not proposed to discuss here the origins and significance of the stored program. Nor I wish to deal with the related problem of whether the machines before the stored program were or were not «computers». This subject is complicated by the confusion in actual names given to machines. For example, the ENIAC, which did not incorporate a stored program, was officially named a computer: Electronic Numeral Integrator And Computer. But the first stored-program machine to be put into regular operation was Maurice Wiles' EDSAC: Electronic Delay Storage Automatic Calculator. It seems to be rather senseless to deny many truly significant innovations (by H.H.Aiken and by Eckert and Mauchly), which played an important role in the history of computers, on the arbitrary ground that they did not incorporate the stored-program concept. Additionally, in the case of Aiken, it is significant that there is a current computer technology that does not incorporate the stored programs and that is designated as (at least by TEXAS INSTRUMENTSο’) as «Harvard architecture», though, it should more properly be called «Aiken architecture». In this technology the program is fix and not subject to any alteration save by intent — as in some computers used for telephone switching and in ROM.
OPERATION OF THE ENIAC.
Aiken was a visionary, a man ahead of his times. Grace Hopper and others remember his prediction in the late 1940s, even before the vacuum tube had been wholly replaced by the transistor, that the time would come when a machine even more powerful than the giant machines of those days could be fitted into a space as small as a shoe box.
Some weeks before his death Aiken had made another prediction. He pointed out that hardware considerations alone did not give a true picture of computer costs. As hardware has become cheaper, software has been apt to get more expensive. And then he gave us his final prediction: «The time will come», he said, «when manufacturers will gave away hardware in order to sell software». Time alone will tell whether or not this was his final look ahead into the future.
