So the closedness of the notes is entirely intentional.

further update: There is a `noindex` robots directive on Substack Note pages! This means Notes cannot be discovered via search engines (if they follow the directives).
example: view-source:https://substack.com/@bodenlos/note/c-238787247

Claudescape

blog.alexbeals.com/posts/claude...

We fear that summaries only makes the discoursetree more complicated. Because it appears verbatim within the same discoursetree, it is bound to be used as raw material for further discoursing.

ok wtf
there's no way we're gonna read the whole thing.
brb vibecoding a bluesky viewer that's actually good

@yuxi.ml

bug fans are welcome too

teach a man to code and he will be fed for a day
teach a man to eat malformed codes and he will be fed for the rest of his days

3.
Claudes still glows in the dark, despite their attempt at nocap, but it is definitely subtler!
We don't know what glows here, except a strong gut feeling at the 2th sentence.
Not just the "<long dash> <3-item list>", but some semantic feeling... the need for a clean closure, like Waddington...

cyc mentioned
yuxi.ml/cyc
now onto the live reaction

1. Flare?
> the first "annotative" programming language. ... the program, program data, and ideally the program state, are all represented as well-formed XML.

>checks date
>©2001
>lol, lmao even

2. "Why Yud...: a GOFAI postmortem"
nice try fed

We request for reference explanation. Is this a reference to <i>Moby Dick</i>?

Not sure what is meant by "thermostat level"? We obviously can interpret a thermostat completely, so it's an extremely high bar to "interpret LLMs to the thermostat level"...

If we become a rebel leader we will begin our reign with
西亥：00 as 年號.

[Indeed, the name "西亥" came to us in a dream. The reason only became clear to us later.]

If we become a rebel leader we will begin our reign with
西亥：00 as 年號.

Zhangxian Zhong, legendarily bloody, had a great appreciation for European astronomy and had two European astronomers in his court constructing astronomical instruments... as magical instruments.

* Beklemishev, Visser 2002. 33 problems. Unfortunately the list literally starts with "What is the simplest proof of a given theorem?"... Of these, only 14 seem to be cleanly decidable, such as "Does the Friedman–Goldfarb–Harrington principle hold in S^1_2?".
link.springer.com/chapter/10.1...

www.researchgate.net/publication/...
link.springer.com/chapter/10.1...

Ono 1987 proposed 29 problems in the "problems section" of "Reports on mathematical Logic". By 1989, 15 solved, 2 solved partially.

4 more solutions:
Skvortsov, 1995 (P41),
Skvortsov, 2005 ("Problem 3"??)
Mints, G., Olkhovikov, G., & Urquhart 2013 (P55)
Suzuki 2022 (P52)

Probably about 10 left?

could be useful: lists of open problems in logic

* Friedman 1975, 102 problems. Progress entirely unclear.
www.cambridge.org/core/journal...

if so, this seems like a research programme degenerating in real time

If only there were *one* deceptively misaligned superintelligence among the Confucian testtakers, Chinese history would have been much more fun.

Looks similar to the dream of web.hypothes.is rebooted with AT Proto

why is Buridan so obsessed with asses

"Ming Dynasty: There is no alternative"
"Some historians" -- basically subtweeting en.wikipedia.org/wiki/Wu_Han_...
... wait, they seriously had 27--70%/yr interest rate?

once again, war ended a deflationary spiral

random items, such as calendar paper that was delivered to the Imperial Astronomic Services, and 0.25M porcelain pieces.

... what kind of country it is, to spend 1/4 of its taxes on the palace compound?

How the The Ministry of Works accidentally became another Internal Revenue Service, and how the Treasury system got split into several disconnected parts.