IRC channel logs

2024-09-27.log

back to list of logs

<mala>jfred: apply for a travel scholarship! you never know!
<cwebber>TZAG!
<haugh>Warning: useless speculation. Something about Radul that really jumped out at me was that he chose Scheme because it was convenient, but his odd language sounds almost disappointed that there was no simple physical substrate on which propagators could be constructed.
<haugh>I can't stop thinking about what would be possible if we had propagators at a foundational level, rather than layers of simulations ultimately based on turing machines
<haugh>Completely useless thought but I had to get it out and this seemed like the place
<jfred>mala: Nah, money isn't the issue for me, those should go to people who need them :)
<jfred>haugh: Gerry Sussman touched on that a bit in "We Really Don't Know How to Compute"
<jfred>I got the impression that part of the motivating factor behind the propagator research was indeed to find a way of programming systems with much more parallel architectures
<haugh>jfred that's right, and Sussman was hinting about genome data as well. What changes at an architectural constraint level if you sacrifice the ability to actively delete? This is so far out of my area that I wish I could just stop thinking about it.