Comments URL: https://news.ycombinator.com/item?id=48293119

Points: 64

# Comments: 7

I'm not a biologist, but I'm under the impression that your body uses the heuristic of "the more acutely a neurotransmitter is suddenly flooded into our system, the more of a homeostatic counter-response we're going to launch in the form of things like dopamine downregulation (etc, depending upon what neurutransmitter we're talking about)".

I'm not entirely sure this is true, but it seems to be corroborated by other researchers (e.g. Anna Lembke in her book Dopamine Nation, which isn't about caffeine though).

This is why substances like theacrine claim to offer even less tolerance than paraxanthine: it has a super gradual adenosine-blocking curve, with super long half-life (like 12-16 hours, IIRC). So when you take one theacrine, you won't notice it for hours, but its effects will last longer than one day (though I forget what its interaction with sleep is supposed to be?).

1. Software engineering isn't "real" engineering.

2. Nobody cares about bugs when they write software in the first place.

3. Software is hard.

4. Software is early.

Comments URL: https://news.ycombinator.com/item?id=44407421

Points: 3

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=44230477

Points: 3

# Comments: 0

https://reference.wolfram.com/language/ref/ExpressionTree.ht...

Steve was a mischievous person himself, so surely a part of him respected this.

Along similar lines, it would be useful to map a speaker's vowels in vowel-space (and likewise for consonants?) to compare native to non-native speakers.

I can't wait until something like this is available for Japanese.

Comments URL: https://news.ycombinator.com/item?id=43804853

Points: 2

# Comments: 0

> For Claude 3.7 Sonnet and Claude 3.5 Sonnet (new), we use a much simpler approach with minimal scaffolding, where the model decides which commands to run and files to edit in a single session. Our main “no extended thinking” pass@1 result simply equips the model with the two tools described here—a bash tool, and a file editing tool that operates via string replacements—as well as the “planning tool” mentioned above in our TAU-bench results.

Arguably this shouldn't be counted though?

[1] https://www.anthropic.com/_next/image?url=https%3A%2F%2Fwww-...

Incredible how resilient Claude models have been for best-in-coding class.

[1] But by only about 1%, and inclusive of Claude's "custom scaffold" augmentation (which in practice I assume almost no one uses?). The new OpenAI models might still be effectively best in class now (or likely beating Claude with similar augmentation?).

[1] I know we're taking forever D: But we intend for this to be a way to release an intermediate product (which we've been making anyway for our full headsets).

[2] Our next blog update will be about this. Here's a video video preview: https://youtube.com/shorts/Y67D8DkqScU?si=LpdSpjmfGn2k2rxP