i³c @smalljaguar.bsky.social · 08/10/2026oh also if you're super paranoid you assume NIST only approves pqc algorithms if NSA have backdoors for them 000
i³c @smalljaguar.bsky.social · 08/10/2026still think odds that new pqc algorithms have significant problems is low due to the selection process rigour, but if they do exist then it makes sense to assume they will be found by LLMs 100
i³c @smalljaguar.bsky.social · 08/10/2026my strongest idea for this would be: all pre-quantum algorithms broken by combo of improvement in hardware and algorithms (seems plausible looking at trendlines accelerating) + new pqc algorithms having weaknesses, not implausible given many other proposed algorithms broken + LLMs doing analysis 110
i³c @smalljaguar.bsky.social · 08/10/2026uses github.com/layoutit/pol..., apparently CSS has 3d support (ish) built in with matrix3d developer.mozilla.org/en-US/docs/W... 050
Reposted by i³csamhain ⎔ @personhood.removal.surgery · 07/10/2026yes i've been reading gender theory again 1221
i³c @smalljaguar.bsky.social · 06/10/202611 square packing proven optimal, proof: github.com/Queuingtheor... linked catalogue of square packings: jlevy.github.io/squares/case... 001
Reposted by i³cStella Biderman @stellaathena.bsky.social · 03/10/2026I’m starting a blog! My first post is on how 3rd party embedded evaluators seem totally unsuited to addressing the problems we are currently facing, and what the real problem is. stellabiderman.ai/blog/embedde...stellabiderman.aiEmbedded Evaluators Can’t Fix Companies That Choose to Be Bad — Stella BidermanEmbedded evaluators can report violations, but they cannot fix AI companies that knowingly disregard basic cybersecurity and safety practices. 37817
i³c @smalljaguar.bsky.social · 03/10/2026en.wikipedia.org/wiki/Toyota_...en.wikipedia.orgToyota War - Wikipedia 0110
i³c @smalljaguar.bsky.social · 01/10/2026from wikipedia: "Rationalists define rationality to include epistemic rationality (coming to true beliefs about the world) and instrumental rationality (acting in a way to achieve one's objectives)" this is obviously entirely consistent with empiricism 000
i³c @smalljaguar.bsky.social · 01/10/2026> Part of their deal is that you can figure stuff out using pure reason instead of experimentation this just isn't true by any of the rationalists' definition of rationalism. saying this requires you to assume they are lying or don't understand their own beliefs. 200
i³c @smalljaguar.bsky.social · 29/09/2026www.ft.com/content/e9b1...ft.comSuspect in RAF Fairford incident called 999 shortly before arrestMotive for the emergency call is unclear, as police continue to investigate incident at air base used by US bombers 142
i³c @smalljaguar.bsky.social · 28/09/2026en.wiktionary.org/wiki/discuss...en.wiktionary.orgdiscuss Uganda - Wiktionary, the free dictionary 001
i³c @smalljaguar.bsky.social · 28/09/2026m.xkcd.com/611/ alt-text:Hurricane forums are full of excited comments about central pressure and wind speed and comparisons to Camille and 1931 and 1938, with hastily-tacked-on notes about how it will be tragic if anyone dies and they hope it's a dud.m.xkcd.comxkcd: Disaster Voyeurism 020
i³c @smalljaguar.bsky.social · 24/09/2026 just checking, you do understand this is a common turn of phrase dating back decades, e.g. news.bbc.co.uk/2/hi/america...news.bbc.co.uk BBC News | AMERICAS | Pinochet defies court order 000
Reposted by i³cFinn @thegoblinpoet.bsky.social · 24/09/2026LITTLE ALEX HORNE: nice simple task, get the queen's first born without her guessing your name GREG: nice and easy, I can't imagine anyone would do something stupid like just stand in a clearing singing their name? LAH: let's see Rumpelstiltskin's attempt RUMPELSTILTSKIN: *head in his hands* 07315
i³c @smalljaguar.bsky.social · 22/09/2026according to this blog, most car designs "overfit" to the tests and score badly when new tests are added danluu.com/car-safety/danluu.comHow do cars do in out-of-sample crash testing? 020
i³c @smalljaguar.bsky.social · 18/09/2026How would this be achieved? Would you support military intervention? Seems unlikely for anything even remotely similar to this to occur otherwise 020
i³c @smalljaguar.bsky.social · 18/09/2026b) but with caveat that they have ~unlimited budget & hire plenty of very smart people so should both be aware of security issues and have capacity to fix them. Comparison I would draw is quant firms who have very tight security because of how much money rests on it 130
i³c @smalljaguar.bsky.social · 17/09/2026clearly we need the anon qt feature to subtly rewrite the text while maintaining the core concept 020
i³c @smalljaguar.bsky.social · 17/09/2026i've read echopraxia (good but a bit weird) but not blindside, really need to get round to it and possibly also re-read echopraxia 110
i³c @smalljaguar.bsky.social · 16/09/2026yeah, this definitely has been a thing cats do since the instant they were domesticated lol 120
i³c @smalljaguar.bsky.social · 16/09/2026tbh I don't think it really applies to scandi countries much either 120
i³c @smalljaguar.bsky.social · 16/09/2026openai.com/research/ver... proves AI generated. pretty easy to tell from the text on the page being super stilted and obviously ai-ish 130
i³c @smalljaguar.bsky.social · 16/09/2026can you expand on why veganism is irrational? The argument that animal suffering is widespread, bad, and that changing diet lowers the total amount in the world substantially for the amount of effort one has to make seems pretty straightforward. Do you deny the existence/badness of animal suffering? 050
i³c @smalljaguar.bsky.social · 15/09/2026eritrea maybe? just going off obscure areas with horrible violence in them 100
i³c @smalljaguar.bsky.social · 15/09/2026less ugly version of same post www.jefftk.com/p/teleoperat...jefftk.comTeleoperated HumansWhen I look at why I expect the world to change a lot in the next few years, and why other people expect slower changes, I think a big component is disagreement on the extent to which AI will affect n... 000
i³c @smalljaguar.bsky.social · 12/09/2026> If anything the opposite seems to be true. interested by this claim, can you expand? 000
i³c @smalljaguar.bsky.social · 11/09/2026www.globalprotection.com/blog/newsroo... fuck Trump? no; no fucking coz of Trump!globalprotection.comGlobal Protection Corp.® CEO Davin Wedel Addresses Condom Supply Chain Disruptions and Rising Costs Amid Middle East Conflict 001
i³c @smalljaguar.bsky.social · 10/09/2026the FT merely assumes their readers aren't moronic and so don't need to tell them what to think in the headline 000
i³c @smalljaguar.bsky.social · 10/09/2026Half the article is just talking about how unpopular trump is! 100
i³c @smalljaguar.bsky.social · 10/09/2026Are you serious? "Capital" hates Trump's tariffs. first line of the article is "President puts himself at centre of congressional election campaign despite his plummeting approval ratings" FT has endorsed democrats in last 3 elections, runs many investigations+criticisms of trump/us gov 100
i³c @smalljaguar.bsky.social · 10/09/2026their answer would presumably be just that AI would be able to do everything knowledge workers can do for much cheaper (in this scenario, which is not a prediction of what is most likely to happen) 100
i³c @smalljaguar.bsky.social · 10/09/2026question was > Zooming out to the broader picture, if we’re looking at 15 per cent annual growth in 2030, what does the economy look like? 100
i³c @smalljaguar.bsky.social · 10/09/2026> So the implication is, the world has new stuff that we don’t currently have today, and this model does not predict the shape of the policy environment people are in. 200
i³c @smalljaguar.bsky.social · 10/09/2026> The implications of a very high growth world are that AI has helped invent or bring about entirely new industries, which suggest entirely new things that people are buying, but also things that are changing the way the world works. 100
i³c @smalljaguar.bsky.social · 10/09/2026> We’re not making predictions here. We built a model, but Ben Bernanke, who’s on the [Anthropic] Long-Term Benefit Trust, had a good take on this, where he said, “If you want to be above 5 per cent GDP growth, you need to be creating entirely new industries”. I think that rings true. 100
Reposted by i³cGrant @syncingsinking.bsky.social · 10/09/2026This is just to say I have solved the open questions that were in your paper and which you were probably saving for your phd students Forgive me they were delicious so incentive-laden and followed so readily once your postdoc's drafts mysteriously found their way into my training data 4357115
i³c @smalljaguar.bsky.social · 08/09/2026ah he said something similar quoteinvestigator.com/2012/04/28/s... > You’ll have to excuse my lengthiness—the reason I dread writing letters is because I am so apt to get to slinging wisdom & forget to let up. Thus much precious time is lost.quoteinvestigator.com 030
i³c @smalljaguar.bsky.social · 08/09/2026thought this was originally a pascal quote? tbh it would be surprising if a mark twain quote was actually said by him at this point 120
i³c @smalljaguar.bsky.social · 08/09/2026I remembered that they used to sell some merch, but when I went to look for it couldn't find it, apparently there was a legal dispute? Annoying. www.maproomblog.com/2025/08/ordn...maproomblog.comOrdnance Survey Settles with Family Business Over Map Blanket Design Dispute – The Map Room 010
Reposted by i³cDarius Baktash @sol-cantus.myatproto.social · 08/12/2025There's cardinal sins, ordinal sins, and ortolan sins. No matter which you choose, sin-eating will be controversial. 021
Reposted by i³c🐺Cold🪦Sam @pdfmerchant.bsky.social · 23/09/2023forget cardinal sins i want to commit ordinal sins. sins only comprehensible by considering the relative postion of all other possible sins 0295
