6,949 karma · joined May 5, 2018
He co-authored this article about the development of the emulator: http://pt.withy.org/publications/VLM.html
The fact that he still has hope is significant.
It was being developed in parallel with Smalltalk-80 (and earlier versions of Smalltalk): https://interlisp.org/medley/history/timeline/
An interview with Ed Fredkin during the Triple-I Reunion October 14th, 2014: https://www.youtube.com/playlist?list=PLZKOq5_Vcp4GGdZHCzKzr...
Relevant video: https://youtu.be/5foJ-cwj4Mk
https://dspace.mit.edu/handle/1721.1/5731
https://dspace.mit.edu/handle/1721.1/6334
Pretty much anything else from Edi Weitz is also great.
Mezzano is quite elegant in my opinion, especially for an operating system. For example, this is the USB mass storage driver: https://github.com/froggey/Mezzano/blob/master/drivers/usb/m...
I agree, it's very concerning that the opacity of ML models is now leaking into large codebases and making them potentially even less comprehensible.
I'm pretty sure computer algebra systems and automatic theorem provers work quite well. The CPU you wrote this comment on was developed using the other kind of AI that supposedly doesn't work: https://en.wikipedia.org/wiki/ACL2
So around $50-100k.
No, it's a fully custom CPU architecture and implementation, though they could use a 68020 as the "LX" co-processor. Also their primary competitor Symbolics used two 68000s in addition to their custom CPU: one as an I/O and boot controller, another as the heart of their console i.e. a graphical terminal.
"Back before the Computing Dark Ages of the Nineties began, I hacked on Lisp Machines. If you've still got an Explorer or Symbolics, you might find some of the stuff I wrote in the good old days useful. It's archived in the CMU AI Repository."