I expected LLMs to become the Everything API. I didn’t expect good tooling for that to exist *this year.* Crikey. t.co/aUM3V8CJxS
Tiny little detail from Tsutaya's coworking space: they festooned a lot of the walls and alcoves with artwork on loan from a gallery, with price tags attached, which seems like a really inspired choice assuming it works.
Interesting fact I never knew, per Matt Levine, with relevance to bank rescues: acquiring a bank will result in new bank needing to mark the acquired assets (loans et al) accurately, which semi-artificially realizes the problems which have been a bit swept under the rug.
There is something like a "general factor of infrastructure" and if you get it, you can see around corners of issues like this without knowing the exact shape of the issue in advance. Seems like it would be an interesting subfield for e.g. operations research. t.co/idkEPdqPUT
One fun emerging form of social organization in tech-adjacent spaces: draw attention and talent into underserved spaces by doing patronage on a bounty model. Sometimes the patronage is distributed. t.co/JDIz7Is7cJ
On the one hand I should not be surprised, and on the other hand I am surprised, and on the gripping hand I wonder whether it can do stylometric analysis accurately given cases which are not obvious layups. t.co/3b9LwwSQdm
One of the most common questions I’ve gotten since writing about bond math, which I thought I already answered, is whether it affects loans as well. Yes, it does. A fixed interest loan decreases in value if interest rates rise. It is not worth the balance (and wasn’t on day 1).
Is there a word to describe the phenomenon of a statement carefully worded to be entirely literally accurate, absolutely unreproachable on legal or similar review, inevitably cause people to read entirely the opposite meaning, with this being intentionally designed?
7 months ago
In Kyoto briefly almost twenty years exactly since I first visited here (and Japan) and the nostalgia is almost overpowering. Also, as if scripted by the universe, hearing people who might as well be me from 20 years ago.
7 months ago