Fascinating article where the author grapples with the ephemeral quality of software. I don’t necessarily agree with his approach but it’s worth thinking about.
The problem of “reflection” and permanent value seems to go along with the “is programming literacy” question.
Even more, it feels like answering the question of what separates natural language and programming would be a huge step towards figuring out what intelligence “is”. If every datum a computer got could “stay with it” in the fashion of the words of a great novel, then I think one might have bridged the problem of computer not having information “ready-at-hand” that Winograd and Flores comment on in their classic book.
The problem is that one needs something a bit specific than engaging in reflection to figure what is important about it.