Not Even Woit

Forays into the unseen spooky comment-at-a-distance youtube comments sections

This post is not even going on the T4G github. There was an interview with Peter Woit, who has things to say about the sick state of fundamental theoretical physics, and is an infamous critic of String Theory. But even he has some weak takes, or poor language framing. I realize I am being a bit pedantic, but I think the language is important.

At the end of this article, just be aware I am 100% confident in MMT, far less so in T4G. But the Clifford algebra is without doubt supreme, it really is I think the “MMT of theoretical physics” or near enough.

Hyping the Ashtekar Variables

Hype not good.

@1:50:00 listen up (or read this!) young physicists: the Ashtekar variables and spinors and twistors; (a) have a far more natural description in the spacetime (Clifford) algebra. No ℂ-numbers needed, and in fact you get several real geometric "complex" (aka. bivector) structures. They generate all the rotations (+boosts) with sandwich products of spinors, translations too if you go to the conformal algebra. If interpreted as local frame fields they describe the unitary transformations (you can think of these as localized spacetime topology changes, i.e., it's particles, not fields). We need fields as accounting tools to describe Hamiltonian time evolution, but they are fictions, as Feynman told everyone long ago, no one seemed to listen.

(b) The spinors are instructions to dilate and rotate. They are not "the particle". Instructions to transform the physical observables (which are composed form the spacetime frame elements) are fields because you can apply them anywhere in spacetime. (c) Twistors are a bi-spinor algebra (4D spinors) in conformal spacetime. Because they are extended in time t the twistors can partially encode entanglement (nonlocal correlators for the conventional Dirac spinors,etc.) though I think this is not the full story of entanglement (it is too general). You do not need uninterpreted ℂ numbers. arxiv.org/pdf/math-ph/0603037

(d) You can work on these ideas. Lots of easy low hanging fruit to pluck. Probably no one will pay you for it, but the intrinsic rewards are so damn good I do it for less than breadcrumbs, and give it all away for free... with a lot of geeky jokes: https://t4gu.gitlab.io/t4gu/ These ideas are a bit crazy, so I say you are welcome to steal them.

That was what I posted on the youtube comments section, and had to add this ribbing:

Is this 👆🏽 my little propaganda post? Sure. Please do not pay it any heed. Go and do your loops and strings homies.

My little attempt at anti-advertisement.

“No new experiments” 🤣

Woit repeated the pablum about the “terrible state” of fundamental physics, since “nothing new” showed up at the LHC. This really irks me. It is so wrong. Hence my comment for anyone reading:

@1:55:00 that's a terrible false framing. There have been petabytes of new experimental results, and that's even ignoring everything not at the LHC. They showed no new particles other than the Higgs. But those are massive results. If you go hunting over millions of hectares and find no new beasts not even sandworms, that might tell you something about the desert you are hunting in. Makes you really go back to the drawing board. A la Turok & Boyle — who have used these huge null results to motivate a whole new breathe of life into the LR symmetric Standard Model.

“All the money is in AI” 🥳😤

Fair to say a science geek commenting on macroeconomics is sometimes going to sound stupid, but should not be laughed at just because of ignorance. Anyhoooo… another chance to pretend youtube comments are social discourse:

@2:00:10 I've loved this conversation, not half because again I can strongly object. "The money" does not "come from AI"‡. The money (state tax credits) can only come from the government. Otherwise it is counterfeit. Chartered banks can only legally issue credit, they cannot create net currency (aka. tax credits, aka. government scorepoints).

Any government can fully fund any research provided the people to do the research exist and want to earn tax credits. Governments are not financially constrained, they are morally constrained by available real resources and wise use of them (or immorally unwise use).

What happens is “the money” channels through AI research and junk products. Who ends up with the tax credits are not the AI researchers. The researchers tend to get to spend all their income, especially if they live in San Fransisco or Mumbai. Capisce? (I actually have no idea about the tech worker wages, I’m just guessing.)

‡ It is even worse, AI junk clogs the Internet and (like crypto, or worse) wastefully consumes vast amounts of electricity for little public good. What comes from AI is pollution and junk, and a few statistically fairly accurate strings in response to prompts, plus FUD about the end of human civilization. It is gross and disgusting. (Yet still I use github Copilot on occasion, it cuts my typos down by half. There is no intelligence there, except for all the humans who generated the code the statistical inference software exploits.)

I’m not debating any AI shills. I am just waiting for the year they all get to upload their minds into silicon. For fair energy resource use they’ll need to terminate their biology. This is going to be a hyper-darwin day. A lot of idiots forever lost from our gene pool. The sad thing is that evolution depends upon our tolerance for weaklings and miscreants — for diversity, for anti-fragility. (The meme “survival of the fittest” is false, gross, eugenicist propaganda.) So their loss will still be sad.

Critique of the AdS-CFT Program

Woit also critiqued the Gauge-Gravity duality research. Sure, sure. But so what if our cosmos is not Anti-de Sitter? What you’ve got are some interesting algebraic correspondences. Although I have never bothered to dig into it all, there is probably theoretical gold there to be found. But not in just the way the former Stringy Theorists think. Hence this comment:

@1:10:00 yeah, but what the (4+1)-AdS stringy theorists ignore is that you can consider this a 3-space Conformal spacetime (Clifford) algebra. The conformal GA (Clifford geometric algebra) extends a euclidean GA with two basis vectors one of + signature and one −. So it is pretty similar to the AdS/CFT bulk spacetime, right? In projective and conformal geometric algebra the line and sphere "at infinity" become concrete, and this permits rotations, translations and dilatons to all be described by rotor sandwich products. This is much nicer because now you are not restricted to AdS, but can do all the analysis for not just de Sitter but also in messy old FRWL. Conformal GR is a proper theory, an extension of Einstein-Cartan gravity, and is even perhaps the superior GR, since with Weyl symmetry we can "explain" the Big Bang, dark matter and dark energy, all without employing the (fictional) 'inflaton' by postulating CPT-Symmetry (q.v. Turok & Boyle). This conformal GA + wormhole topology is thus practically "QM=GR" or even "GR⇒QM" by any other koan. (The particle sector is isomorphic, or at least functorially related to the Hilbert space formalism, up to gauge redundancy of the latter.) Here though, you do not need the extra 5 Calabi-Yau, you do not need the superstrings. Non-locality can be handled by wormhole topology around the Planck scale (non-traversable except to qubits.) The particle phenomenology can also be handled by 4D topology. You do not need to promote fictional fibre bundles to concrete reality (though they are a useful math tools for some purposes.)

The more time marches on, the more I get the feeling the Clifford geometric algebra is in physics darn closely analogous to MMT in macroeconomics. Everyone just ignores it. But without it you do not know your full policy spacetime.