mncharity

> American friendliness

I'm hesitant to move further from TFA, but a brief one-shot: Among Boston people who returned after "trying" SV/SF/California, I repeatedly heard of struggling, amidst an unfamiliar ambient "friendliness" culture, to recognize social signals around "actual" friendship - was someone "really" interested in becoming friends, or were they merely performing "friendly". AI tells me this is called peach vs coconut, nice vs kind, guess vs ask cultures.

Similarly, someone (with no family connection to China) learned Chinese, later married into a family from/in China, and would describe "Chinese" culture. Someone else from China glossed their insights with a dismissive 'their friends are criminals [sic], and they're over generalizing from that subculture'.

So amidst country-level cultural themes, and regional variations, and generational differences (and gender, hukou, ethnicity, etc), and subcultural diversity, it's sometimes a challenge to tell what is which.

Yeah, the onboarding process for independent research needs work. Especially with various disruptions coming for existing settings. The Gap by Ira Glass pops to mind.[1] We need a bunch of those. Analogous to the conversation around becoming a writer or other artist. Or to starting a business. With lots of insights into what needs doing, how, and how to sustain. With its fun parts, tedious parts, grueling parts. The universal challenge of crafting a life which sustains, but encompassing the research. Imagining and pulling into existence the rhythms, constraints, network/village, the exoskeleton to usefully produce. As a DIY project, rather than "here's your ID, desk, mentor, advisor, lounge, whiteboard supplies, schedule, the person to see if you need anything done, and the weekly lab social is today at 5". There's also the "I imagined myself a freelancer, but find myself a sole proprietor" twist. Maybe some of this is starting to exist? Perhaps there are blogs out there? I've not looked. A survey of that might make for an interesting blog post or HN discussion.

[1] https://gointothestory.blcklst.com/the-gap-by-ira-glass-9651...

> unless you're running something like a high-refresh-rate gaming monitor that's still 1080p or less.

Or HMDs. Upcoming high-resolution PICO Space Pro is said to be ~45 pixels/degree max. Versus ~50 ppd for a 1080p laptop. Formerly known as Project Swan, its September global release announcement was postponed until... unclear.

"What does this independent research look like?"[1] - timestamp link into Implicit Surfaces & Independent Research (2025) on that research page.

[1] https://www.youtube.com/watch?v=UxGxsGnbyJ4&t=2150s (I started the clip a few seconds early as teaser for rest of talk)

> My favorite project along those lines is Kartik Agaram's Mu[3], which wraps emulation, tracing, and time-travel debugging around a subset of x86 assembly language.

From https://akkartik.name/code : Mu[1], a safe, hackable computing stack that builds up from raw machine code (guided tour[2]; paper '20[3]; talk '20[4]; sample app[5])

[1] https://github.com/akkartik/mu [2] tour https://github.com/akkartik/mu/blob/main/tutorial/index.md [3] paper https://akkartik.name/akkartik-convivial-20200315.pdf [4] talk https://vimeo.com/416363066 [5] sample app https://archive.org/details/akkartik-mu-2021-08-15

Fwiw, top of my own pain-point list (I suppose given the first item, that's a pun) includes:

External/policy-based throttling for temperature control. Unthrottled, my laptop bottom goes skin-burn hot. But fixed compute caps can have non-linearly dreadful performance impacts in particular cases. Plan is a runtime knob, to replace manual limits-kludgery.

I'll use models which barely fit in VRAM+RAM, and are order-1 tok/s slow. So tool call step overhead can be painful - a world where `ls` costs tens of seconds. Plan is blending harness plugins with inference loop, for "no, don't stop - I already have the call result for you - just keep going" (and also some logit games).

Fun: 116 d/s with Earth/etc selected, gives a Sun "orbiting" the Earth/etc effect.

Bug: ISO DEB at 17 min/s looked normal, but at 2.8 h/s it became a solid-ish (lots of crossing lines) red ellipsoid on the opposite side of the Earth. Upon returning to 17 min/s, it snapped back, after some delay. Current-ish Chromium on linux/X11.

Feat: Pressing ESC deselects.

Feat: Perhaps spacebar could pause/unpause time? To avoid fiddly "note current time ratio by counting arrows or hovering, then click pause. Afterward, click the right place to restore time ratio".

UX: Would you believe I wrote a Feat for 'add a "Now" button to date popup', not "seeing" the current button? Hmm, perhaps move it between date and time ratio?

UX: Just now for me, upon page load, nothing is moving and the Earth seems a black ball (night over Africa, with night lights obscured by satellite dots). So I could spend several minutes exploring, without being both close enough and oriented to notice there's a texture map. Perhaps start positioned over the terminator (interestingly pretty) and with scaled time (interestingly moving; demos motion; draws attention to time zoom. Maybe 17 min/s?).

Issue: The planets maps are true-ish color, modulo intensity and shadow. The Sun is... not. Given that "the Sun is some color other than white" is so pervasive a misconception, reinforcing it seems... sigh. Can I interest you in a white field with a few pretty sunspots? Perhaps with limb darkening and tint?

Show HN: Reladraw – A diagram language where you decide where to place things[1], which adds relative placement constraints, and provides agent skills. Nearby on the front page just now, fwiw.

[1] https://news.ycombinator.com/item?id=49858513

> map!

Maps.[1][2]

Too far south for an ice bridge - Newfoundland is south like modern Florida - tropical. And it's the Jurassic, hot with high sea levels, so even though its continental shelf extends out towards Spain, there are no ice ages to bare it (as was the Eastern US shelf a mere 12k years ago). But Pangea is ripping apart, it's a mess with crust blocks variously uplifted and tilted, massive weathering and sediment fluxes, so...

From the TFA paper[3]:

During the early Mesozoic, North America and Europe were connected to form the western and central part of the supercontinent Laurasia. With the opening of the North Atlantic during the Late Jurassic and earliest Cretaceous, the region fragmented into smaller tectonic units55–58 that might have periodically formed contiguous land masses57,58 during sea level low-stands, characterized by exposure of continental margins and draining of continental interior seaways55–58. The presence of a land bridge between North America and Europe during the late Kimmeridgian-earliest Tithonian13 might explain similarities in some dinosaurs such as Stegosaurus59 and Allosaurus60. However, the absence of turiasaurs in the Morrison Formation suggests that this clade dispersed into western North America after the mid-Tithonian. Thus, unless we wish to argue that turiasaurs were actually present in the Morrison Formation but have not been found yet, a second later land connection between Europe and North America is required. Specifically, during the mid-Valanginian (c.137Ma), a dramatic drop in sea level7,55,56 potentially reconnected Europe and North America and so permitted a fi al exchange of vertebrate clades across the Atlantic. Th s Valanginian event could also explain apparent faunal exchanges between these continents proposed for the late Hauterivian-early Barremian13. Taxa that seem to disperse in the Hauterivian/ Barremian might actually have done so during the Valanginian lowstand, but gaps in the fossil record mean that the results of these dispersals might not be detected until we access better sampled post-Valanginian deposits.

58 is available with a plate.[4] 57 is old, with only low-res global plates, ggl books preview has 138 Ma on p42. 56 I skipped. 55 is available[5] but is temporally low res.

There's a 2016 paper Late Mesozoic North Atlantic land bridges[6], Elsevier paywalled with excerpts, and on sci-hub.

And that's my time budget.

[1] https://portal.gplates.org/cesium/?view=topo_recon [2] rocks: https://portal.gplates.org/cesium/?view=GeologyR key: https://portal.gplates.org/portal/geology/ [3] quote is on page 9: https://dinodata.de/bibliothek/pdf_d/2017/001_s41598-017-146... [4] paper ref 58, 150/136 Ma plates on page 31, key on page 29, 3 MB but downloads slowly: https://archimer.ifremer.fr/doc/00117/22805/20913.pdf [5] https://www.researchgate.net/profile/Golonka-Jan/publication... [6] Late Mesozoic North Atlantic land bridges (2016). Available on sci-hub. https://www.sciencedirect.com/science/article/abs/pii/S00128...

ERRATA: Those google book links now do have hits for "bound" and "round". Also "range" and "interval". The searches really didn't before, thus the reference links. Perhaps the absence should have failed sanity check, but given that one hit... partial failure of Google Books search wasn't something I recall noting previously. Oops.

Does European public dialog do much causal comparisons among states? Does somewhere else?

Or do you mean professionally... Instead of merely searching for "ways to do X", one might translate the query into a couple of languages, and translate the results, to get a broader culture sampling. Hmm... That seems something an AI-based search could integrate smoothly - cross cultural insights as a normal part of search results?

While there's a lot of diversity in policy and culture among US regions and states, contrasting them isn't leveraged much. Perhaps failing a "that's analysis, not news" threshold for being part of public dialog? So maybe it's not "non-English cultures" neglected as much as comparative discussion in general?

One "criminally underutilized" which really struck me, was Japan having a far more sensible and powerful COVID safety one-liner than the US managed - "Three C's" (Closed spaces, Crowded places, and Close-range contact - poorly-ventilated / crowd / conversation-contact - avoid overlapping all three) vs "6-foot rule".

Oh, good point! Something to check. I've been thinking of Chinese as a strength of Chinese models, but other languages might be sufficient here.

> What does "estimation" mean in the US?!

"Estimation" as a skill in US education, primary school on, is pervasively about point estimates - single numbers. Bounding, a range of safe/reasonable/likely/possible numbers, is very rare. In Fermi questions/problems also.

Even in such goodness as Sanjoy Mahajan's Art of Insight and Street-Fighting[1]... the only only "bound" mentioned is "Printed and bound in [USA]", and there is no "round".[2]

In contrast, IIUC, China is pervasively about "Da Gu" and "Xiao Gu", "round up" and "round down". Point estimates are secondary. Lower and upper bounds are apparently even a cultural concept: an individual's degree vs capability/fate; policy equity vs excellence; economic targets.

In education, my understanding is point estimates have much poorer conversational/collaborative dynamics than range estimates. Bounding lends itself to incremental collaboration. "Can anyone suggest another low or high bound?". Discussions around point estimates seem usually merely about the collection of individual estimates.

> NGSS

Ah, NGSS isn't bad particularly. I meant that discussion/mention of NGSS appears so very much in English training data, that AI chat on topics related to NGSS seem to draw regurgitations of NGSS. AI "slop" in the sense that "ignore NGSS, just reason from a superset of its underlying concepts"... isn't an LLM strength.

> hard to understand

Sorry - Thanks for asking!

[1] open access: https://direct.mit.edu/books/oa-monograph/5345/The-Art-of-In... https://direct.mit.edu/books/oa-monograph/5339/Street-Fighti... [2] search "bound": https://www.google.com/books/edition/The_Art_of_Insight_in_S... https://www.google.com/books/edition/Street_Fighting_Mathema...

Fwiw, it's been very easy to nudge Qwen 27B/35B to think in Chinese. No need for <think> prefills like "思考：" ("Think:") or such.

Relatedly, when chatting with frontier models about science education content design, I've found it very helpful to mix in Chinese education terms. In English, for example, NGSS is such a massive attractor, discussing nearby topics often yields NGSS "slop". And "estimation" (educational) in the US means one (dysfunctional:) thing, which similarly distracts. Perhaps if AIs become increasingly multilingual, but remain weak at deep conceptual reasoning, it may be fruitful to have multilingual thesauruses, to use language-associated cultural conceptual differences as a way to convey conceptual nuances with which LLMs otherwise struggle?

Brief blog post, with picture of UI: https://plasticlabs.ai/blog/research/Plastic-Labs-at-ICLR-20... (site has an scroller bug with anchor seeking - relevant bit is half-way down the short page).

Github: https://github.com/ibm-granite/granite.debug-tools/tree/main...

Paper: LogitScope: A Framework for Analyzing LLM Uncertainty Through Information Metrics https://arxiv.org/abs/2603.24929

> it uses very clear legal language. I was wondering how I could get it to speak like that all the time

At least with smaller models, reframing a task can alter code style. As in, we're not creating an X app, we're creating an exemplar of ..., which just happens to use an X app as the illustrative example. Which shifts style away from generic app cruft, towards exemplar of whatever.

So perhaps try to establish a legal context? Maybe "Compliance and Legal will be reviewing our conversation today. So it is important to communicate in a style they will find comfortable/familiar." or some such? "This conversation will become part of a legal deposition ...".

UX report. I wished to examine attention state step by step, but I found the animation moved along too fast for that. So I tried pausing...

On Chromium/linux, pressing pause doesn't pause, instead resetting the animation to it's pre-play state - the current attention highlighting disappears. Pressing play again, restarts at the beginning. Having a commonplace "pause pauses, and play resumes" UI, could allow more time to look over state. A youtube-like slow playback 0.25? option might similarly help. Or perhaps even better, buttons for single stepping. Tnx for your work.

I was wondering a few days ago, if for teaching geography, one might collect compelling images/stories, which combined, might weave a familiarity with the layout of the planet?

Commonplace, is Earth having a water hemisphere, and landed one. Or the now-distant but coupled shapes of South America and Africa. The rumpled bedsheets of the Indian plate collision. Then there's a green crescent of green China, SE Asia, and India, curved around a contrasting desert China and the Stans. Maybe China/Mongolia as a wedge inserted between Russian and SE Asia plus India, poking the Stans, injecting Silk Roads towards the Seas. The Blue Banana of Europe. Ring of Fire. Great Plains topographic funneling and tornados. Africa now desert and green, was khaki/green and less green, and Last Glacial Maximum was desert and dry.

TFA map is another.

Thoughts?

TFA is background for Kickstarter[1] CHC5: Open Camera System - World's First Commercially available Modular, Customizable, Programmable Machine Vision Camera with Open Source Software.

[1] https://www.kickstarter.com/projects/circuitvalley/chc5-open...

Part 1 - Sensor Specification Comparison (TFA) - https://www.circuitvalley.com/2026/08/chc5-open-camera-senso...

Part 2 - Camera System - https://www.circuitvalley.com/2026/08/chc5-open-camera-syste...

Part 3 - Sensor and Lens Mount - https://www.circuitvalley.com/2026/08/chc5-open-camera-lens-...

Part 4 - Software and Architecture - https://www.circuitvalley.com/2026/08/chc5-open-camera-gige-...

Part 5 - FPGA Architecture - https://www.circuitvalley.com/2026/08/chc5-open-camera-isp-f...

Too compressed? Something better unpacked?

> 1500 t/s [...] all you can eat tokens as fast as you can eat them

Pfft, you can eat tokens far far faster than that. Many orders of magnitude more. Just switch from frontier-style bespoke artisanal pets to cloud burst-parallel, latent subspace exploring/exploiting/searching, mass ensembles of cow herds.

There's N-Version Programming. Work the problem in English, in Chinese, in Haskell, Lisp, Rust, etc. Then work ports to the target lang.

There's design space sampling. Work the problem emphasizing performance, or security, or monitoring, readability, etc. Then work a synthesis.

There's non-determinism sampling. Work the problem order 10 or 100 times. Then work to combine the best bits from each.

There's sample synthesis. NP-hard aggregation of insights.

There's genetic exploration. Work populations of trees of work variants under selective pressure.

There's repo quantum superpositions of implementation space. The unspecified remains indeterminate - state space collapse occurs not upon each edit/commit, but as JIT-synthesized fuzzing/search upon each execution.

There's maintaining a pretty dev UI, but that >>10k tok/s is trivial, because like symbiotic adversary cocreation, fine-grain agent swarms, scenario analysis/forecasting, etc, etc, it is unlike the preceding items... which scale combinatorially.

"All you need is 1500 t/s"? "All you need is 640k RAM" is only 5 orders of magnitude off from 64 GB. It takes "All you need is a single Intel 3101's 64 bits", to get 9 orders of magnitude from 64 GB. Then datacenters...

Or base 2. 2^1.5 (2.83) is the low bound on two orders of magnitude base 2. So tripled fits.

All this concern, when he merely assumed everyone would realize he meant oom base 2. :P

> has grown 100x (minimum number of "orders")

Two orders of magnitude is 31.6 to 316 (10^1.5 to 10^2.5), so 32x would be sufficient.

Most Uber 5-year growth numbers seem less than 3x. Steelmanning, some programs seem new in 2021 (Uber One, autonomous driving), and thus might have grown more dramatically as they spun up.

But yes, eep. Maybe misuse will, by getting more exposure for the term, eventually get us more people who know and use it properly? Maybe?

An contrasting addendum: Most causal chains^H roots are deep and hairy, and most roles are supporting roles, rather than a principal with clear story impact. That friend who forwarded Ansary's email onward.

So maybe your question prompts some response here, which the author or someone then improves upon when this topic rolls around again, which someday affects a list Claude generates when someone asks it to aggregate "What can I do?" ideas from the web, which ... who knows? But better the question was asked than not. And better the responses are novel and insightful and compelling, than not. And better our discussion here refines those ideas, than not. No?

"Better", in some hopeful but poorly-instrumented fuzzy-metric sense. Which is a difficulty. Whether it's worth the opportunity cost of however else you might be spending your time this evening... perhaps better this topic/thread, than at least some others?

Just a thought, which at T+36 min hasn't been mentioned yet.

Many systems are chaotic. At particular points in state space, a very small delta can result in very different subsequent trajectories.

Viral influence being one example. I'm reminded of Tamim Ansary's email[1] to 20 odd friends after 9-11, that went massively viral ("We come now to the question of bombing Afghanistan back to the Stone Age. Trouble is, that's been done. The Soviets took care of it already. Make the Afghans suffer? They're already suffering. Level their houses? Done. [...]").

Also, some moments of teaching can change lives. And in everyday life, sometimes recognized and sometimes not, things done and not done can have big impacts on people. Ripples.

It is said of serendipity and opportunity, that you can't control them, but you can shut yourself off from them. And thus advice to live in a way that leaves yourself open to them, and to cultivate the awareness and skills to recognize and execute when they go by.

One can't know when one is a potential part of a state space fork. But might one make throwing spaghetti at the wall part of one's life? An occasional hobby? Support spaghetti throwing? Or potluck spaghetti throwing communities? Or pasta R&D or manufacture? Trebuchets?

Perhaps a category of "Things worth doing all the time, which are far from being worth doing every time, but are still worth doing each time." Or Chesterton’s "If a thing is worth doing, it is worth doing badly."

[1] https://cmsw.mit.edu/reconstructions/expressions/tamim.html

> The valley where the floods occurred in Nepal has a V-shape

https://earth.google.com/web/@28.34107217,85.43477356,2834.6...

> it's a fascinating story worth reading

Nod. Thus my comment, mostly aimed at the poster, that the (then) absence of votes/comments/engagement shouldn't necessarily be taken as lack of HN interest, but might reflect paywall friction. I could have been more clear, but was thinking more of supporting a dead post's postmortem, than of creating distraction on a subsequently active post. Tnx.

Interesting. At the time, the vote count was notably low given the post age, and there were no comments IIRC. So my thought was, to avoid the poster thinking "well that didn't work - I guess HN didn't like this - doesn't want future similar posts from me", by noting an engagement funnel bounce step they may not have been aware of. Given that paywalls are often probabalistic UX. Sort of using an empty comment thread as TFA poster DM channel?

Hmm. So, maybe, if my comment had noted the age/vote pair, and was clearer about the motivation for mentioning the paywall, that would have been less likely to create a meta discussion? And perhaps been consistent with the "don't post complaints about paywalls" policy?

In case anyone wonders over the low vote count, for me just now, the link only briefly shows the article before swapping it for a paywall.

Fwiw, just to share old work, here's[1] an archived page (slowly loading; wasn't intended to be public; variously broken) exploring sort of a "teach scale down to atoms, and then nucleons up to materials" approach, based on hands-on scale modeling, physically realistic representations, and real-world connections. Vision was one would start early with piles of tiny sticky atom fuzzballs, containing massive blurred sometimes-oddly-shaped (eg, ground-state Neon's wine-bottle) nuclei. Nuclear dynamics are traditionally not much covered, but have analogs for atomic phenomenon which are, hypothetically providing an opportunity for "see it again, but different".

[1] https://web.archive.org/web/20221007220513/http://www.clarif...