By the time ALGOL 60 came around, this aspect had already created two completely different computing cultures.

I remember a conversation in 1962, in Rome. We were sitting around a coffee table. One American boasted that he had made an “algebraic translator” of 50,000 instructions, only to be immediately outdone by one of his compatriots, whose algebraic translator comprised no less than 80,000 instructions.

Peter Naur broke the subsequent silence of awe by remarking that he had written an ALGOL translator of 5,500 instructions, upon which I could outdo him with a compiler of only 2,700 instructions. In short: our yardsticks for achievement measured in opposite directions!