@ftacv.bsky.social
As always the problems begin when *other people* use python, a group that includes your past self from two weeks ago
December 26, 2025 at 12:44 PM
As always the problems begin when *other people* use python, a group that includes your past self from two weeks ago
👍 Neat way of providing user-alterable defaults, convenient way to report config errors
👎👎👎 Spelunking a massive kwargs object through a series of nested function calls
👎👎👎 Spelunking a massive kwargs object through a series of nested function calls
December 26, 2025 at 12:36 PM
👍 Neat way of providing user-alterable defaults, convenient way to report config errors
👎👎👎 Spelunking a massive kwargs object through a series of nested function calls
👎👎👎 Spelunking a massive kwargs object through a series of nested function calls
Yes!! Piketty talks about the "machinery of justification" which allows us to permit a particular inequality regime. Ours is the fetishisation of intelligence, and almost every argument about intelligence, artificial or otherwise, buys into this machinery
November 2, 2025 at 8:07 PM
Yes!! Piketty talks about the "machinery of justification" which allows us to permit a particular inequality regime. Ours is the fetishisation of intelligence, and almost every argument about intelligence, artificial or otherwise, buys into this machinery
This does make me wonder though, if you took the connectome of everyone in Tokyo, would you be able to do something similar? Like a MDL for the set of connections that refer to living in tokyo
October 13, 2025 at 7:09 AM
This does make me wonder though, if you took the connectome of everyone in Tokyo, would you be able to do something similar? Like a MDL for the set of connections that refer to living in tokyo
Right, so this is coming in as how many bits to specify a person so log2(Hpop) (gwern.net/death-note-a... ), and I guess people are going "but quantum!" because they're thinking of it as a "how much information to reproduce a person"
<em>Death Note</em>: L, Anonymity & Eluding Entropy
Applied Computer Science: On Murder Considered As STEM Field—using information theory to quantify the magnitude of Light Yagami’s mistakes in <em>Death Note</em> and considering fixes
gwern.net
October 13, 2025 at 7:06 AM
Right, so this is coming in as how many bits to specify a person so log2(Hpop) (gwern.net/death-note-a... ), and I guess people are going "but quantum!" because they're thinking of it as a "how much information to reproduce a person"
Is that because it'd be in the form of a posterior over the parameters and then each instantiation is just a draw from that distribution? Otherwise I don't see how you guarantee how the path from one node to another can be specified by a few bits
October 13, 2025 at 6:49 AM
Is that because it'd be in the form of a posterior over the parameters and then each instantiation is just a draw from that distribution? Otherwise I don't see how you guarantee how the path from one node to another can be specified by a few bits
I suppose you'd store this as a graph or a series of embeddings, so then who's the node with the highest degree? Or the person that's the closest to the primary eigenvector. Presumably a Chinese guy that likes beer
October 13, 2025 at 6:40 AM
I suppose you'd store this as a graph or a series of embeddings, so then who's the node with the highest degree? Or the person that's the closest to the primary eigenvector. Presumably a Chinese guy that likes beer
An average of under 65?!?! I thought the SMBC guy was better than this tbh
October 8, 2025 at 3:32 PM
An average of under 65?!?! I thought the SMBC guy was better than this tbh
If we were productive members of society we wouldn't be here! But thanks, if you do formalise it, my priority for any biological modelling attempt is "what is the actual measurement going to be"
October 3, 2025 at 10:54 PM
If we were productive members of society we wouldn't be here! But thanks, if you do formalise it, my priority for any biological modelling attempt is "what is the actual measurement going to be"
Would you mind writing downa and sharing the actual maths at some point?
October 3, 2025 at 10:15 PM
Would you mind writing downa and sharing the actual maths at some point?
Incidentally, although the "intuition through natural language -> repeated observation" approach pisses everyone off, biology has eaten many modellers who believed that the problem was a lack of mathematical sophistication amongst biologists.
October 3, 2025 at 10:09 PM
Incidentally, although the "intuition through natural language -> repeated observation" approach pisses everyone off, biology has eaten many modellers who believed that the problem was a lack of mathematical sophistication amongst biologists.
Ok, thanks! So, as I understand it your point is that unless you can observe a direct mechanistic improvement, you can't map a property to a causal mechanism because there are too many competing paths?
October 3, 2025 at 10:05 PM
Ok, thanks! So, as I understand it your point is that unless you can observe a direct mechanistic improvement, you can't map a property to a causal mechanism because there are too many competing paths?
Apologies, I can't follow this, would you mind restating your point here?
October 3, 2025 at 8:58 AM
Apologies, I can't follow this, would you mind restating your point here?
(In fairness, maybe insects?) But it's just very hard to do any comparative experimental evolutionary biology with much rigour - you are either limited to intuition from repeated observation (ethnographies almost), experiments on limited model systems, or inferential statistics on sequence data.
October 3, 2025 at 8:54 AM
(In fairness, maybe insects?) But it's just very hard to do any comparative experimental evolutionary biology with much rigour - you are either limited to intuition from repeated observation (ethnographies almost), experiments on limited model systems, or inferential statistics on sequence data.
Ok, but no population is in equilibrium, you can only imperfectly sample it, there is no constant mutation rate even within an individual, sampled observation of mating practice. This is a bar I think no study of animal trait could hope to clear
October 3, 2025 at 8:50 AM
Ok, but no population is in equilibrium, you can only imperfectly sample it, there is no constant mutation rate even within an individual, sampled observation of mating practice. This is a bar I think no study of animal trait could hope to clear
What does your objection mean analytically? From the perspective of the optimiser, there is only fitness and fitness-correlation (on the unit of selection, YMMV). Humans loosely categorise phenotypic traits into a schema. Is your objection to the categorisation, or the categories?
October 3, 2025 at 8:18 AM
What does your objection mean analytically? From the perspective of the optimiser, there is only fitness and fitness-correlation (on the unit of selection, YMMV). Humans loosely categorise phenotypic traits into a schema. Is your objection to the categorisation, or the categories?
It really changed my mind over the need for a strong demarcation between our rights and responsibilities as citizens and scientists. As with so much else, the dysfunction of the American right has allowed many other lesser evils to fester
September 9, 2025 at 8:42 PM
It really changed my mind over the need for a strong demarcation between our rights and responsibilities as citizens and scientists. As with so much else, the dysfunction of the American right has allowed many other lesser evils to fester
Deeply frustrating, then and now, how much of it became shadowboxing over the BLM protests, but the sci-com/PH Nexus really did not cover itself in glory during COVID. A horribly toxic cycle between (implemented!) policy, shoddy work and groupthink.
September 9, 2025 at 8:38 PM
Deeply frustrating, then and now, how much of it became shadowboxing over the BLM protests, but the sci-com/PH Nexus really did not cover itself in glory during COVID. A horribly toxic cycle between (implemented!) policy, shoddy work and groupthink.
Reminds me of periodic nomadic invasions into China -> initially an incredibly brutal and purely extractive regime gets kinda swallowed by the culture - mixture of strong existing norms and a need to compromise to actually administer something so much more complex than a nomadic warband
September 5, 2025 at 1:36 PM
Reminds me of periodic nomadic invasions into China -> initially an incredibly brutal and purely extractive regime gets kinda swallowed by the culture - mixture of strong existing norms and a need to compromise to actually administer something so much more complex than a nomadic warband
September 5, 2025 at 6:51 AM
Numbers fuckenstein-esque character trying to convince some violent death cultists that the long-run murder productivity can be increased with moderate welfare spending is funny but doesn't seem like a terribly stable equilibrium
September 5, 2025 at 6:48 AM
Numbers fuckenstein-esque character trying to convince some violent death cultists that the long-run murder productivity can be increased with moderate welfare spending is funny but doesn't seem like a terribly stable equilibrium
Enjoyed reading the SU worldbuilding doc. I got the sense that it's an iterated game where elites determine they can maximise violence with some degree of societal co-ordination. Does that consensus periodically breakdown and give you a few futile cycles of bloody civil war before being reasserted?
September 5, 2025 at 6:22 AM
Enjoyed reading the SU worldbuilding doc. I got the sense that it's an iterated game where elites determine they can maximise violence with some degree of societal co-ordination. Does that consensus periodically breakdown and give you a few futile cycles of bloody civil war before being reasserted?
Yeah you start to get into trouble with the concepts of "it" and "approximates", but in general yes. When I was in undergrad there were suggestions of effectively step size control via mutation frequency and heat shock proteins, but no idea if anything ever came of it
August 23, 2025 at 6:12 PM
Yeah you start to get into trouble with the concepts of "it" and "approximates", but in general yes. When I was in undergrad there were suggestions of effectively step size control via mutation frequency and heat shock proteins, but no idea if anything ever came of it
Also in terms differential equations, everything that changes in time in proportion to itself (i.e most things) is exponential in time. I'm sure if I was a mathematician that would be an obvious fact but it's mad to me that this constant appears all over reality
August 23, 2025 at 6:08 PM
Also in terms differential equations, everything that changes in time in proportion to itself (i.e most things) is exponential in time. I'm sure if I was a mathematician that would be an obvious fact but it's mad to me that this constant appears all over reality
Evolution is an interesting one, because it's an optimiser that has no way of evaluating dSurvival/dGene.
August 23, 2025 at 6:05 PM
Evolution is an interesting one, because it's an optimiser that has no way of evaluating dSurvival/dGene.