EWD249 - 51

itself.) The task is then viewed as one of text manipulation; as an aside we may recall that the need to do so has been used as an argument in favour of punched cards as against paper tape as an input medium for program texts. The actual modification of a program text, however, is a clerical matter, which can be dealt with in many different ways; my point is that if we have our grip on the program text primarily as on a linear sequence of symbols, the task to establish and to describe what has to be modified tends to become prohibitively difficult when the texts get longer and longer.

If a program has to exist in two different versions, I would rather not regard (the text of) the one program as a modification of (the text of) the other one. It would be much more attractive if the two different programs could, in some sense or another, be viewed as, say, different children from a common ancestor, where the ancestor represents a more or less abstract program, embodying what the two versions have in common. Hopefully, this common ancestor can be readily recognized in the (prae-)documentation. The intentions are

1. that the two versions share their respective correctness proofs as far as possible;

2. that the two versions share (mechanically) as far as possible the common (or "equal") coding;

3. that the regions affected by the modification are already well-isolated, a condition which is not met when the transition requires "brain-made" modifications scattered all over the text.

Well, this is a lofty goal. It has been inspired by the potential similarity between the task of program modification and program composition: when a program has been built up to an intermediate stage of refinement, what has then been written down is in fact a suitable "common ancestor" for all possible programs produced by further refinements. It is the similarity between "the decision to be changed" and "the decision still left open": in both cases we are left with what remains when we abstract from such a decision.

There is a second source of inspiration to be found in our experience. In the process of step-wise program composition, proceeding from outside inwards, going towards progressive refinements, we have in the earlier stages not only postponed deciding how certain things would be done, but we have also postponed