I’m disappointed in the maths community but not regretting my decision to avoid taking maths as a major. This is remarkably in line with the mentality cultivated by the typical maths department and must be a frustrating daily experience.

If you ever reopen the topic with your colleagues, an interesting starting point might be their thoughts on theorem 14 from this suite; in my opinion, Lean 4 is clearly untrustable. The lack of awareness of this issue seems to stem from the same blind spot that leads to lack of awareness of non-mathematical cognition: whaddaya mean it doesn’t all reduce to one particular pile of symbols which all fit together perfectly?

Sometimes I feel inadequate. I was clumsy when burning Justin M. Keyes and, regardless of whatever trust was lost, I haven’t felt better even after a week of making fun of fascists. Maybe burning somebody else will help? I’d like to share my notes on Dan “Opcode” Jonsson, a relatively-unknown right-wing influencer, before he gets popular enough to have money. I think he’s spooked on Lobsters recently, but he’s been uploading to Youtube multiple times per week and there’s plenty to explore.

If this is also too clumsy and leads to Jonsson deleting his YT channel: good, actually! It’s the only source of revenue that I can find for his opinions. So I’m not going to worry too much about it.

I’m probably going to increase my presence on DreamHost in California, who I’ve hosted with for over a decade. They historically are First Amendment proponents known for being quietly friendly towards adult content; also, in 2017 DH pushed back on the Department of Justice’s subpoena for 1.3 million IP addresses visiting an anti-Trump protest-organizing website. That said, I’m 100% prepared for yet another Milkshake Duck Moment.

Looking forward to being able to send email, which DigitalOcean still blocks in order to funnel customers towards hosted email-delivery services.

Crosspost, hat-tip to @pnwpixel@programming.dev for finding this. Omarchy announces another patron: DigitalOcean. I will wait for DO to explicitly confirm/deny this, but if this is true then I’m going to be moving my entire small-business footprint.

I guess that I’m the resident compiler engineer today. Let’s go.

So why not write an optimizing compiler in its own language, and then run it on itself?

The process will reach a fixed point after three iterations. In fancier language, Glück 2009 shows that the fourth, fifth, and sixth Futamura projections are equivalent to the third Futamura projection for a fixed choice of (compiler-)compiler and optimizer. This has practical import for cross-compiling; when I used to use Gentoo, I would watch GCC build itself exactly three times, and we still use triples in our targets today.

[S]uppose you built an optimizing compiler that searched over a sufficiently wide range of possible optimizations, that it did not ordinarily have time to do a full search of its own space — so that, when the optimizing compiler ran out of time, it would just implement whatever speedups it had already discovered.

Oh, it’s his lucky day! Yud, you’ve just been Schmidhuber’d! Starting in 2003, Schmidhuber’s lab has published research on Gödel machines, self-improving machines which prove that their self-modifications will always be better than previous iterations. They are named not just after Gödel, but after his First Incompleteness Theorem; Schmidhuber et al proved easily that there will always be at least one speedup theorem which a Gödel machine can never reach (for a given choice of axioms, etc.)

EURISKO used “heuristics” to, for example, design potential space fleets. It also had heuristics for suggesting new heuristics, and metaheuristics could apply to any heuristic, including metaheuristics. … EURISKO could modify even the metaheuristics that modified heuristics. … Still, EURISKO ran out of steam. Its self-improvements did not spark a sufficient number of new self-improvements.

Once again the literature on metaheuristics exists, and it culminates in the discovery of genetic algorithms. As such, we can immediately apply the concept of gene-oriented evolution (“beanbag” or “gene pool” reasoning) and note that, if goals don’t change and new genes don’t enter the pool, then eventually the population stagnates as the possible range of mutated genes is tested and exhausted. It doesn’t matter that some genes are “meta” genes that act on other genes, nor that such actions are indirect. Genes are genes.

I’m gonna close with a sneer from Jay Bellou, who I hope is not a milkshake duck, in the comments:

All “insights” eventually bottom out in the same way that Eurisko bottomed out; the notion of ever-increasing gain by applying some rule or metarule is a fantasy. You make the same sort of mistake about “insight” as do people like Roger Penrose, who believes that humans can “see” things that no computer could, except that you think that a computer can too, whereas in reality neither humans nor computers have access to any such magical “insight” sauce.

I don’t care whether you agree with me. I do think it’s quite interesting that my critique of the topic of your post has you replying with personal insults; I think that you’ve incorrectly assumed that I am one of the capitalists who build these models, rather than an anti-capitalist who encourages destroying corporations.

When this issue is finally settled by law, history will not look back well on you.

Who are you, Lars Ulrich? Copyright is not compatible with information theory, and so far, information theory has won every contest in the court of public opinion.

The point is to alter the economics …

Help defray or establish artists’ funds. Help uncover and prosecute wage theft. Advocate for basic income. If your goal is to remunerate artists, then focus on the efforts which actually help them; don’t support copyright, as it is neither designed nor implemented to help individual artists.

Shame on you …

Which one of us is defending a paper which explicitly offers itself as a way for Disney to protect art which it appropriated from artists?

I can’t endorse Glaze or Nightshade, sorry. If literally nothing else, it’s not Free Software and it’s offered with a nasty license:

You are not permitted to … reverse engineer the Software …

You are not permitted to … permit … any part of the Software … to be combined with or become incorporated in any other software …

So I’m not allowed to have the discussion I’m currently having, nor to include it in any Linux distro. To me, that’s useless at best and malicious at worst. Ironic, considering that their work directly builds upon Stable Diffusion.

Also, Nightshade will be ineffective as an offensive tool. Quoting from their paper:

… the perturbations we optimized on poison images are able to perturb image’s features in text-to-image models, but they have limited impact on the features extracted by alignment models. … We note that it might be possible for model trainers to customize an alignment model to ensure high transferability with poison sample generation, thus making it more effective at detecting poison samples.

This is not only an admission of failure but a roadmap for anybody who wants to work around Nightshade. Identify poisoned images by using an “alignment model,” which correlates images with sets of labels, to test whether an image is poorly labeled; if the image appears well-labeled to a human but not to an alignment model, then it may be poisoned and will need repair/corroboration from alternate sources.

I also ranted about this on Mastodon.