Re:What Made Lisp Different (2002)
by fiddlerwoaroof on Sunday February 18, 2018 @08:49PM
(#16410181)
What should be possible, though, would be to distinguish a surface-syntax from the underlying code: e.g. a language that can transparently convert to/from c-style curly brace blocks, Pascal-style begin/end and s-expressions. possibly implemented as a bunch of git hooks that transforms to a normalized representation (as, e.g. prettier does for JS) on push and transforms to the programmer's preferred representation on pull.
[ Parent
| on HN ]
I suspect you'd find that what look like surface differences actually go deeper into the meanings of programs, like a plant whose roots go deeper than one would expect.
There's a special level of Hell that involves streaming xml processing and cobol. Since cobol lacks dynamic memory allocation (speaking of cobol 85 here) and you don't generally build things like linked lists or trees, there's a limited number of things you can conveniently do with streaming xml data. Many cobol programmers are left with a feeling of despair, I wager.
The Sapir-Whorf hypothesis is certainly real in programming languages[1].