Gregor

@coexact.bsky.social

only made of atoms

let me try to explain. As everyone knows from kindergarten, a massless fermion comes with a definite chirality and a mass term for a fermion is a bilinear coupling between two such of opposite chiralities (a Majorana is the special case where they are each others antiparticle).

Bild

This is a nice discussion with points that I am largely sympathetic to. I have some thoughts on the conclusions drawn at the end about the emergence of time—though these matters are so uncertain that it's impossible for me to have any tightly-held conviction. Nevertheless:

Sean Carroll@seanmcarroll.bsky.social · 5mo ago

Mindscape 355 | Solo: Looking Quantum Mechanics in the Eyeball. I argue that QM is even farther from the everyday world than we generally assume, but connecting them is an ongoing and promising project. #MindscapePodcast www.preposterousuniverse.com/podcast/2026...

Title card for Mindscape episode -- Solo: Looking Quantum Mechanics in the Eyeball.

A new, very serious preprint is out from Veronika and Alberto. No, don't look at today's date it's 100% legit we promise. It's a genuine proof that P=NP... (...if you believe in quantum many-worlds and are willing to destroy almost every observer in the multiverse 👀) arxiv.org/abs/2603.28869

Bild

so I just realized the very obvious fact that if I can apply a noise removal filter to my mic input in calls, I can also apply it to any other audio stream and poorly recorded video lectures just got so much more pleasant

Physicists of Bluesky! I'm looking for a specific type of story from the early days of cyclotron research, for a piece for @physicstoday.bsky.social . Have you heard tales of scientists who aligned beams with the naked eye? Know an older physicists who has? Let me know!

You've heard of Solomon Lefschetz, but did you know he had an evil twin? We just call him perverse weak Lefschetz, he proved the perverse weak Lefschetz theorem

Here's a mathematical fact which I find amazingly counter-intuitive: There exists a polytope (=the convex hull of a finite set) in ℝ⁴ which is not combinatorially equivalent to one with rational vertex coordinates (=the convex hull of a finite subset of ℚ⁴)! 🤯 •1/4

In a strange position where I think there are massive, massive social harms to the proliferation of AI (misinfo, deepfakes, predatory 'companions', surveillance, labour rights) but also feel the mainstream arguments against it range from unconvincing (water) to actively harmful (copyright; "soul")

kinda interesting story here. Can a LLM with some vibecoded framing produce a decent index for a book? I confess I wasn't aware of indexing as a profession prior to this, but even without leaving this site you can find the tail end of a community, with orgs, conferences, meetups..

Ed@ed3d.net · 10mo ago

I want to read this article. Two (meant constructively) things: 1) "demonstrating that LLMs are worthless for indexing" should, like most of these things, have qualifiers like "in our hands and with our LLM" 2) when people talk like this about accuracy, I suspect the workflow as much as the model.

no matter how you slice it, if intelligent life survives all other hurdles, it will eventually end for astrophysical reasons, be that the expansion of the sun or cosmic heat death. It will be like climate change but worse: much slower, much more predictable, not preventable, hopelessly unsurvivable

Isaiah Bishop@isaiahbishop.bsky.social · 10mo ago

so like is life in that galaxy kind of fucked long term? Kind of love hate the idea of being one of the solar systems in the wake thats just flung on its own