5 ms·

An Algebraic Language for the Manipulation of Symbolic Expressions (1958) [pdf]

171 points by kbp 9y ago

shakna 9y ago
Page 10. A description of a linked list, and the memory structures required to intermingle them. Incredibly simple, powerful, and to my eyes, beautiful.
john255 9y ago
who cares, useless
- Devagamster 9y ago
  there's that hacker news ray of sunshine I've grown to know and love :) glad all is right in the world
- dang 9y ago
  We detached this subthread from https://news.ycombinator.com/item?id=14886673 https://news.ycombinator.com/item?id=14886673 and marked it off-topic.
johnhenry 9y ago
For anyone wondering about the source: https://en.wikipedia.org/wiki/John_McCarthy_(computer_scientist) https://en.wikipedia.org/wiki/John_McCarthy_(computer_scient...
JBiserkov 9y ago
Reading this I get an urge to reach back in time and tell the author about all the amazing programs and languages this work will lead to, about Emacs, Scheme, Clojure, ...
- sitkack 9y ago
  He found out. Not sure he used any Clojure.
newman8r 9y ago
They've got some interesting content http://www.softwarepreservation.org/projects/ http://www.softwarepreservation.org/projects/ A lot of ideas about what was only theoretically possible at the time - good time to revisit.
cjauvin 9y ago
It [the language] is not so convenient for representing lists of fixed length where one frequently wants the nth element where n is computed rather than obtained by adding 1 to n-1. I find this is an interesting way of saying it, it made me think for a moment..
tpetricek 9y ago
For anyone interested in additional context, the paper "AI and the Origins of the Functional Programming Language Style " by Mark Priestley is a great introduction: https://link.springer.com/article/10.1007/s11023-017-9432-7 https://link.springer.com/article/10.1007/s11023-017-9432-7 (sorry, I don't have a non-paywalled link, but SciHub does the trick).
- shanusmagnus 9y ago
  Here's a PDF from the author's site: http://www.markpriestley.net/pdfs/AIandFunctionalStyle.pdf http://www.markpriestley.net/pdfs/AIandFunctionalStyle.pdf
fernly 9y ago
Reading this my eye was caught by the name Gelernter. Wait, wasn't that the guy who... no, it was his father, Herbert[1], who > implemented, with Nathaniel Rochester, a computer language for list processing within FORTRAN ... the Fortran list processing language (FLPL) His son David, who would have been 3yo at the time McCarthy wrote this paper, has had his own effect on computer science[2]. [1] https://en.wikipedia.org/wiki/Herbert_Gelernter https://en.wikipedia.org/wiki/Herbert_Gelernter [2] https://en.wikipedia.org/wiki/David_Gelernter https://en.wikipedia.org/wiki/David_Gelernter
seesomesense 9y ago
Another classic paper by John McCarthy, the father of Lisp and much of early AI.

171 points by kbp 9y ago

Page 10. A description of a linked list, and the memory structures required to intermingle them. Incredibly simple, powerful, and to my eyes, beautiful.

john255 9y ago

who cares, useless

Devagamster 9y ago
there's that hacker news ray of sunshine I've grown to know and love :) glad all is right in the world
dang 9y ago
We detached this subthread from https://news.ycombinator.com/item?id=14886673 https://news.ycombinator.com/item?id=14886673 and marked it off-topic.

johnhenry 9y ago

For anyone wondering about the source: https://en.wikipedia.org/wiki/John_McCarthy_(computer_scientist) https://en.wikipedia.org/wiki/John_McCarthy_(computer_scient...

JBiserkov 9y ago

Reading this I get an urge to reach back in time and tell the author about all the amazing programs and languages this work will lead to, about Emacs, Scheme, Clojure, ...

sitkack 9y ago
He found out. Not sure he used any Clojure.

newman8r 9y ago

They've got some interesting content http://www.softwarepreservation.org/projects/ http://www.softwarepreservation.org/projects/ A lot of ideas about what was only theoretically possible at the time - good time to revisit.

cjauvin 9y ago

It [the language] is not so convenient for representing lists of fixed length where one frequently wants the nth element where n is computed rather than obtained by adding 1 to n-1. I find this is an interesting way of saying it, it made me think for a moment..

tpetricek 9y ago

For anyone interested in additional context, the paper "AI and the Origins of the Functional Programming Language Style " by Mark Priestley is a great introduction: https://link.springer.com/article/10.1007/s11023-017-9432-7 https://link.springer.com/article/10.1007/s11023-017-9432-7 (sorry, I don't have a non-paywalled link, but SciHub does the trick).

shanusmagnus 9y ago
Here's a PDF from the author's site: http://www.markpriestley.net/pdfs/AIandFunctionalStyle.pdf http://www.markpriestley.net/pdfs/AIandFunctionalStyle.pdf

fernly 9y ago

Reading this my eye was caught by the name Gelernter. Wait, wasn't that the guy who... no, it was his father, Herbert[1], who > implemented, with Nathaniel Rochester, a computer language for list processing within FORTRAN ... the Fortran list processing language (FLPL) His son David, who would have been 3yo at the time McCarthy wrote this paper, has had his own effect on computer science[2]. [1] https://en.wikipedia.org/wiki/Herbert_Gelernter https://en.wikipedia.org/wiki/Herbert_Gelernter [2] https://en.wikipedia.org/wiki/David_Gelernter https://en.wikipedia.org/wiki/David_Gelernter

seesomesense 9y ago

Another classic paper by John McCarthy, the father of Lisp and much of early AI.