I have read it in an old "hacker test" fun text, about in the middle 90s. It was about 4.
Today googling refers to the idea that maybe it had been about writing into a read-only intended memory area.
But I am not so sure. For old languages and OSes, a concept of lvalue was maybe not very clear yet. I think, for language developers of the 60s, 70s, the main concern was, how to compile the least amount of asm operands per source code line. Probably at the time, "lvalue" was not yet very well known, I see it more probable that it was considered a semantic expectation of the language, on the left side of the assignments can only exist actual memory addresses.
What might have been the reality?
Where and how was it possible to change the value of the constant "4"?
