3 ms·
This site reckons there are 10: http://hyperpolyglot.wikidot.com/lisp#ten-primitives http://hyperpolyglot.wikidot.com/lisp#ten-primitives Steve Yegge said ther
by juliangamble 5y ago
This site reckons there are 10:
http://hyperpolyglot.wikidot.com/lisp#ten-primitives http://hyperpolyglot.wikidot.com/lisp#ten-primitives
Steve Yegge said there were seven (or five):
McCarthy had five.
On an x86 architecture - you can do it with one.
https://stackoverflow.com/questions/3482389/how-many-primitives-does-it-take-to-build-a-lisp-machine-ten-seven-or-five/ https://stackoverflow.com/questions/3482389/how-many-primiti...
- lioeters 5y agoI had imagined this was merely theoretical, but learned that there are computers implemented with a single instruction. https://en.wikipedia.org/wiki/One-instruction_set_computer#Machine_architecture https://en.wikipedia.org/wiki/One-instruction_set_computer#M... https://esolangs.org/wiki/OISC https://esolangs.org/wiki/OISC
- eatonphil 5y agoI don't think movfuscator proves you only need one lisp primitive. Just that any language can be evaluated using mov's many variant forms.
- tromp 5y agoIn the functional realm which LISP inhabits, two combinators suffice: S and K. https://en.wikipedia.org/wiki/SKI_combinator_calculus https://en.wikipedia.org/wiki/SKI_combinator_calculus
- afiori 5y agoin the context of S and K, lisp is not functional