lajarre

Gestell contemplator; program synthesizer; futarch @buttermarkets_

[lajarre @lajarre]

⚫ futarchic Hawking radiation

[lajarre @lajarre]

@EliBenSasson @danrobinson @socrates1024 Godel: ⊨ is not ⊢ (in ℕ, Peano arithmetic is incomplete)
Tarski: ⊢ can't reach ⊨: syntax (⊢) is recursively enumerable, whereas semantics (truths in ℕ, ⊨) is not

https://t.co/wFUqYSMj4o

[lajarre @lajarre]

@kunchenguid there are important odds that Fable doesn't come back soon: https://t.co/mS9on7D9p5
-p appears like a cheap way for A to limit the intensely felt Fable withdrawal pain and limit churn among power users

[lajarre @lajarre]

https://t.co/ehn3TW3vWK

[lajarre @lajarre]

@neogoose_btw obviously you can get finer-grained improvements with Pi; but you do need customization investment
otoh, only a fool would not benefit from the stock harness for off-the-shelf performance

hammer+nail thinking ngmi, in one sense or the other

[lajarre @lajarre]

if we assume at time t, there is going to be a period P from now to before decision where proposal impact (delta conditional prices) is:
1. epistemically constant (no new information)
2. on average, uncorrelated to other factors
3. and where P(Yes)*P(No) is always far enough from 0
allow traders to buy into an instrument that pays the twap of the conditional delta
if P ends up not existing, refund traders

[lajarre @lajarre]

@kunchenguid routinely happened to me with gpt5.5 already
sounds like a proper harness with boundaries is the only way forward

[lajarre @lajarre]

@Scott_eth @Polymarket respectfully, you're overreaching: a Yes combo still doesn't resolve to collateral if one of the legs resolves to True; only No combo unlocks when one leg gets False

compression is needed because poly uses Yes/No positions even in combos
it's a complement to negrisk, in a sense

[lajarre @lajarre]

@sreeramkannan what about:
- embodiment is the source of feeling and taste, thus disembodied ai needs us as oracles
- this might still apply to maths and anything which draws on aesthetics

[lajarre @lajarre]

@danrobinson the unit distance proof is constructive => easy to verify => the model comes up with a local research loop for cheap
let's suspect we'll see more such proofs popping up
on the contrary, results requiring minting new theories are much further away

[lajarre @lajarre]

@paulg Proof Without Words

[lajarre @lajarre]

@seb3point0 thanks for organizing 🙏

[lajarre @lajarre]

@paulg @0xdoug @romlib_ EV[welfare | inevitable] > 0
theorem or axiom?
one has to remember that economics is normative

[lajarre @lajarre]

not just proofs: conjectures, open expositions, even games...
proofs might be less of a differentiating output, conjectures much more so

i don't know we target numbers, but we do want "intellectual societies" that produce some form of scientific progress; perhaps progress has to be redefined?

[lajarre @lajarre]

@FUCORY stuck at lvl 3 for now, gimme the nudge

[lajarre @lajarre]

@jchaskin22 bring Andreas back

[lajarre @lajarre]

@odysseus0z flipsidedly, learning hard stuff is not optional anymore

[lajarre @lajarre]

@shafu0x @deedydas aka the Hard Problem of SFness

[lajarre @lajarre]

@bentlegen @badlogicgames 🙏

[lajarre @lajarre]

@airtightfish godspeed

[lajarre @lajarre]

gotta love stripped-down DSLs: generate bugless unruggable software at inference speed

[leoglisic.eth @Leo_Glisic]

What if we never had to deploy governance smart contracts ever again?

The concept, called 'Governance Nodes', was the topic of my presentation at the Stanford Blockchain Governance Conference last weekend.

Here's a quick run-through.