dynomight

My popular (?) demand, my fixed NumPy now exists as "real" (ish) package. If you use UV, you can try it by typing this one-liner:

uv run --with dumpy-numpy --with ipython ipython

pypi.org/project/dump...
dynomight.net/dumpy/

Had a dream where I was at some kind of blogger event and people were giving each other feedback. Freddie deBoer was assigned to me and he said, "The main thing about your blog is that it is very very very bad."

There's an interesting debate here about the value of bundling an idea together with a piece of art. Certainly 3-body *felt* more profound than I think reading this synopsis would, even though I think you've basically captured everything. (That's compatible with known physics.)

This is what I used to think, but sum-utilitarians have strong counter-arguments that all other aggregation functions run into problems that are at least as bad.

I'm now pretty convinced that there's no function that doesn't lead to substantial weirdness.

plato.stanford.edu/entries/repu...

seed oil theorists would love this amirite

New colors without shooting lasers into your eyes dynomight.net/colors/

This is amazing

bsky.app/profile/mick...

In all seriousness, I'd probably learn more by doing the opposite experiment and eating exclusively processed food!

My 9-week unprocessed food self-experiment dynomight.net/unprocessed-...

zoomers

They have detailed results in the supplement. There's lots of variability by station, but they include Kendall with much higher numbers than you found, so it's not just location.

I can't find any info about ventilation improvements at any stations other than Back Bay. So... I'm unsure!

There is a lot of research supporting the idea that particulate levels in subways really are quite high: dynomight.net/air/#particl...

Luglio (2020) specifically gives measurements for Boston.

(Of course, I don't doubt your measurements—things vary greatly by line!)

There is a lot of research supporting the idea that particulate levels in subways are in fact quite high: dynomight.net/air/#particl...

Luglio (2020) specifically gives measurements for Boston.

(Of course, I don't doubt your measurements—things vary greatly by line!)

I wonder how much this really suppresses the red cones. It could be that this only does it a little and the lasers would be dramatically more impressive. But it also seems quite possible that the lasers aren't significantly different.

Apparently you can experience something like the same without lasers! www.skytopia.com/project/illu...

(Sadly, I must admit that the correlation you suggest is probably empirically true.)

> I mean yeah an anon account is racist, color me surprised.

☹️

Holy hell, wow.

(Somewhat tempted to try to learn how to do that voluntarily. But it seems like people might find that more disturbing than impressive...)

Links for July dynomight.net/links-3/

eyes, money, nepotism, links, scribbles, futarchy, war and peace, yams

Orexin Pilot Experiment for Reducing Sleep Need Testing intranasal orexin-A administration for sleep need reduction: a 2-3 participant self-blinded randomized controlled trial

You can use magical text-snippet links

manifund.org/projects/ore...

The adverse events are weird... Temperature intolerance? Diabetes? Vitamin D deficiency? Aspartate aminotransferase increased"

I could be convinced these are all minor. But I could also be convinced that you're playing into the hands of people who want IRBs for all human research. :)

I'm just worried about side-effects! The fact that a stage was study has been approved is somewhat relieving. But it looks like an earlier study was halted for safety issues?

clinicaltrials.gov/study/NCT048...

I'm theoretically opposed to this but also very eager to hear the results, to the degree that I feel a bit ethically troubled.

Do blue-blocking glasses improve sleep? dynomight.net/blue-light/

This is a very thoughtful rebuttal. (And extremely well-written!) Recommended. caseymilkweed.substack.com/p/response-t...

Futarchy's fundamental flaw - the market This is a market inspired by this recent blog post. I am trying to capture the scenario from the section "No, order is not preserved" as closely as possible. Suppose there’s a conditional prediction…

Does anyone have some spare Manifold Mana they'd like to donate to increase the liquidity subsidy for this market?

manifold.markets/BoltonBailey...

I do wonder how much of this is technology stagnating versus NVIDIA making ever-larger profits!

Oh I agree. I think I'd go so far as to say that in principle, that's always the best method? (Assuming it's very very carefully done.)

Featuring this tool to make your own scribble-based AI forecasts

Scribble-based forecasting and AI 2027 dynomight.net/scribbles/