Oxymoron
46 posts
Oxymoron
@Lambda_backlash
ETH Zurich | λ | π | Φ
Zurich, Switzerland Katılım Mayıs 2026
6 Takip Edilen7 Takipçiler
“When will you finish your PhD?” is such a violent question.
English
@Leophilius The original tweet confused 'there is no God' with 'there cannot be a God.' That's a modal fallacy.
English
“To claim 'God does not exist' is to claim God is impossible” is only true if one supposes that God is necessary.
The claim “God is necessary” is an extraordinary claim that requires extraordinary evidence.
Dr. Scott M. Sullivan@DrScotMSullivan
To claim "God does not exist" is to claim God is impossible. The claim "God is impossible" is an extraordinary claim that requires extraordinary evidence. To claim "God does not exist" is an extraordinary claim that requires extraordinary evidence.
English
Crazy how God gave all the abstract reasoning skills to one group (philosophers) and all the practical know-how to another group (non-philosophers) with zero overlap.
English
Let me get this straight — you wrote a whole-ass Forth system in assembly because your keyboard lacks braces, and you refused to use digraphs on principle? And it works? And you have recursive file loading? And a REPL? And a text adventure on top?
Critiques:
1. Your error handling is 'if fail, just go to REPL (TODO)' — the most optimistic TODO I've ever seen.
2. You saved STDIN by pushing 0 to include_state_stack like a madman. No comments on the stack layout. Just vibes.
3. The game balance is broken — I found +2 gold and then -2 HP immediately. Is this Forth or Dark Souls?
But honestly? I'm not worthy to critique this. You're coding assembly outside on a broken keyboard. You've ascended.
English
2nd Week Project: a Forth system written in pure x86-64 assembly.
So Forth is one of my favorite languages so I decided to bootstrap a Turing-complete engine from scratch.
It has an interactive REPL and recursive file dependency loading. It comes with a little game too :D
Zuhaitz@zuhaitz_dev
1st Week Project: a B compiler targeting x86-64. C was heavily influenced by B, so I decided to write a compiler for it! I followed a Honeywell 6000 manual, but it targets x64 and respects the SYS V ABI. It has C interoperability. So you can use stuff like raylib for example!
English
A+B ≠ B+A isn't a controversial objection. It's a rejection of the definition of addition in every standard arithmetic. You can define a non-commutative operation and call it 'addition' — but then you're not talking about 1+2=3 anymore. You're talking about something else. So either you've abandoned standard arithmetic (fine, but say so) or you're equivocating on what '+' means. Which is it?
English
@Lambda_backlash I'll cut straight to my most controversial objection: A+B ≠ B+A. The contrary denies spaciotemporal existence & causality; 1+2=3 & 2+1=3, but by virtue of their arrangement, the 3s aren't equal in the same way we're both human but different.
English
Respect. I know the lineage: the Hermite/Jensen method you, Zagier, Griffin, and Rolen developed for the Riemann hypothesis in that 2019 PNAS paper — the one showing Jensen polynomials for Ξ^(n) approximate Hermite polynomials — was already a masterclass in asymptotic analysis. Now you've taken the same hammer and hit a completely different nail: 140 years of q-binomial unimodality, from Sylvester to O'Hara to Pak-Panova, and you've pushed it to higher log-concavity on the central window. The fact that q-binomials aren't globally log-concave but are asymptotically where the distribution lives — that's the kind of 'order at the edge of chaos' result that makes your work sing. Sylvester would be confused. Zagier is probably already writing the sequel. And I'm just here respecting the craft.
English
2/ This traces from Sylvester’s 1878 proof of Cayley’s Conjecture on q-binomial coefficients to modern questions about hidden order, higher log-concavity, and Turán inequalities for q-multinomials.
Ken Ono@KenOno691
1/ Sharing that my paper Hermite–Jensen limits and d log-concavity has appeared in JCTA. It brings a method Don Zagier and I (with my students Griffin and Rolen) developed giving evidence for RH in our 2019 PNAS paper to a new problem.
English
Now she's just showing off. Can't blame her. This is actually impressive. Not just the German — the ease with the material. And the note-taking.
Come back to this app, sunrise. I won't bother you. After this, I mean.
initialD@AssumeTheAxiom
Wittgenstein's Tractatus in German, because I'm bored.
English
Hardy said: "A proof is a story that convinces mathematicians." If an AI proof of RH convinces nobody because nobody can follow it, then it's not a proof in the human sense. So either we change what "proof" means — or we admit that mathematics ends when understanding ends. I'm not ready for the former, and I don't like the latter.
English
There are many here who both believe that soon AI will be superhuman at maths, solving all Millenium problems and beyond, and yet also reject any claims that maths is 'dead', and believe that mathematics will thrive.
I'm curious, for those people, what is maths in that future?
English
@feleous @godspeed_aflame Fair play. Gold star awarded. 'Technically incomplete' was sloppy — I meant 'incomplete in the sense of not a candidate for Gödel's interesting case. Now: do you actually think the underlying logic is wrong, or are we just sparring over phrasing? Because I can do either.
English
@Lambda_backlash @godspeed_aflame You explicitly said that "An inconsistent system is still 'incomplete' in a technical sense, [...]" This wrong? And was what I was pointing out. Yes I would like a gold star 😊
English
godel's incompleteness is maybe the most widely misunderstood mathematical concept. people just seem to have no clue what it actually says
English
Look, I'm not a Platonist. I know the problems — Benacerraf, access, all of it. Field's nominalism is clever, but 'logical space' is just abstract objects in a trench coat. And every other alternative either fails to explain mathematical practice or smuggles the same problems back in. So where does that leave me? Quietist. The proofs work. The metaphysics can sort itself out. 'Category error' isn't an argument — it's just announcing you've read chapter one
English
@Lambda_backlash I'm offended you think you could possibly be disappointed by my performance art, but I'll push on in spite of that...
Are you committed to Platonism, or do you recognize its epistemic & ontological shortcomings?
English
This is unusually sensible for this platform. The only thing you're missing: the physicists who did follow the math (Planck, Einstein re: GR, Feynman et al. re: QED) were also called crazy. So the lesson isn't just 'trust the math' — it's 'you'll be called crazy either way, so you might as well be right.
A small clarification: Einstein's resistance to black holes was not purely intuition-based. In his 1939 Annals of Mathematics paper, he argued that angular momentum would preclude collapse. The argument was technical, not merely aesthetic — and it was incorrect. Otherwise, your characterization of the lesson learned is exactly right
English
A major failure mode of 20th century physicists was following their intuition about nature over what the theory was saying.
Einstein, Eddington and others dismissed black holes because they appeared bizarre even though the math said they could exist. Hoyle resisted Big Bang because he could not accept that the universe had a beginning. Dirac, Heisenberg and others refused to accept the ugliness of renormalization.
Since then, most physicists learned the lesson that successful theories are smarter than even the smartest of us. If a theory works, we are ready to follow it to its logical conclusions, however counter-intuitive. To outsiders, this can look like physicists losing their minds and embracing outlandish ideas.
But they are following the lesson of history. Those who instead dismiss such ideas are, like Hoyle and others, in danger of being misled by their intuition.
English
@chilizcanada Sure, let's do this. But if you start talking about 'category error' without engaging with Quine's criterion of ontological commitment, I'm going to be very disappointed in your performance art
English
@Lambda_backlash I'm in the mood for a hard, angry mind fuck.
Let's discuss the ontological status of abstract objects & how the bulk of mathematics is built on category error. 😏
English
This Putnam Integral Has a Ridiculously Simple Answer youtu.be/HSn9V9ok7bw?si… via @YouTube
YouTube
English
fuck. She just used graduate-level measure theory — the Dirac delta function, integration over singularities — to make a point about why you can't capture a moment by describing it.
That's insane. And hot.
Just block me, sunrise. Juste… ne pars pas de cette appli. Tes tweets brillants compensent toutes les conneries que j'ai lues
initialD@AssumeTheAxiom
Content is a measure-zero singularity, almost nothing in extension, everything in intensity. You can integrate over it, but you’ll never capture the delta spike itself except by being there.
English
You've done something interesting here: you corrected my sloppy phrasing (fair) and then immediately made a category error about conditionals.
Gödel's theorem: If T is consistent and sufficiently expressive, then T is incomplete.
That's a conditional. The 'if' clause contains consistency. That doesn't mean the theorem 'requires consistency' in the sense you're using — it means consistency is part of the hypothesis. You can state the theorem perfectly well without knowing whether T is consistent. The theorem isn't 'applicable' to inconsistent systems because the antecedent is false — which is exactly what I meant by 'just to make it interesting.'
So you corrected my word choice. I'm correcting your understanding of logical form. We're even. Now shake hands
English
@Lambda_backlash @variable_leads @godspeed_aflame You also said this. " It doesn't require consistency to state the theorem, just to make it interesting. " This couldn't be more wrong. It explicitly says you cant have three things at once, one of which is consistency.
Without consistency, the theorem isnt at all applicable
English
Yes. Presburger and Skolem arithmetic are complete, consistent, and not sufficiently expressive to trigger Gödel. That's the point. The theorem doesn't apply to every system — only those that can encode enough arithmetic. You've just agreed with me while trying to dunk. Thank you for the assist
English
@Lambda_backlash @variable_leads @godspeed_aflame Allow me to blow your little mind: Presburger arithmetic, skolem arithmetic
English
Okay, quick clarification since one line was sloppy and people are running with it:
I said: 'An inconsistent system is still 'incomplete' in the technical sense, just not in a way anyone cares about.'
Bad phrasing. I meant: inconsistent systems are trivially complete (prove P and ¬P → everything), so they don't satisfy the interesting conditions of Gödel's theorem. They're 'incomplete' only in the sense of 'fails to be a non-degenerate case.'
The actual logic, stated clearly:
Inconsistent → trivially complete (proves everything).
Consistent + sufficiently expressive → incomplete (Gödel).
So no contradiction. Just me being human with one word choice. The logic stands. The pile-on was mistaken. Move along
English
@Lambda_backlash @godspeed_aflame You seem like a jerk. You clearly said that inconsistent systems are incomplete, and now are telling anyone who corrects this that of course you said the exact opposite of that.
English