ทวีตที่ปักหมุด
*in my most woo voice*
yeah, i study the harmonics of crystals' natural energy resonances. this one is titanium:sapphire
English
Peter Johnsen, PhD
2.4K posts
Peter Johnsen, PhD
@rocketjohnsen
Man of science, metal bender, EUV producer, and human-powered traveler
Cambridge, MA เข้าร่วม Ağustos 2021
281 กำลังติดตาม2.4K ผู้ติดตาม
one of the great virtues of American culture since its founding, and something that truly sets the US apart from most of the world, is Americans' pursuit of material prosperity as opposed to negative sum tribal conflict
wanye@xwanyex
I just completely reject that we are wealthier than we were 20 or 30 years ago in any way that matters. Whatever data seems to show that we are is very clearly not measuring what you want to measure. I don’t think burrito taxis matter. I think the phones have made us miserable. I don’t think the advancements in communication are important. I don’t care that the cars have more features. I don’t think any of this stuff matters like even a little bit. And I certainly wouldn’t trade my country for more of it. I think we long ago reached a level of comfort that’s sufficient. We aren’t hungry. We aren’t cold. We aren’t worked to death in fields. I wouldn’t trade my family for more stuff. I wouldn’t trade my country for more stuff. The people who see all this so-called material wealth and think that the opportunity costs are just too large to have children or that we need to give away the country to foreigners to keep the train rolling are in my view demented, deranged, lost. And their addiction to antidepressants and other mood-altering pills seems to confirm it.
English
@juliet_turner6 @DrAllyLouks I heard a story of someone who accidentally ingested moth pheromones (idk how), and for many months he had moths that would congregate by his feet
English
@DrAllyLouks I don’t know about Androsterone but I do know that you can buy moth pheromones online if your goal is to attract lots of beautiful moths
English
If you want to learn more about AI + hardware integrations and the effectiveness of prompt-level safety constraints, check this out: peterjohnsen.substack.com/p/i-gave-claud…
English
I did exactly that, and Claude is better at experimental physics than I expected! But it still doesn't know how to stay safe in the lab
English
Everyone wants to integrate AI with hardware these days, so...
What if we gave Claude a laser!?
English
@sierras_account I heard that "robot" means "machine that doesn't work" and that once it works, you stop calling it a robot and start calling it a machine
ex: ATMs were originally called robot cashiers
English
The impressed distinction between the two seems to be the lack of mobility of the frame underneath.
Machine can't move in granular ways, robot can.
"Actuator" designs suck and are seldom the lowest complexity path to doing what you want. Speaking from personal experience
English
This is how wrong most people are about framing problems to be solved robotics.
The robot doesn't need tactile feedback and to use hand tools ($$$$ to use $)
It needs a $$ torque sensor on its deburring tool hand and then it can determine how done the job is.
PhotonCommander☄️@photoncmndr
@mattdykema Gonna need tactile feedback sensors for detecting how sharp edges are
English
@rocketjohnsen Nice list =D but we need more neighborhood specifics
English
Places I've lived alone as an adult:
P*** A***, California
Indian Reservation outside Santa Fe, NM
West Hollywood, LA
Pico Union, LA
Noe Valley, SF
North End, Boston
Undisclosed, Pittsburgh PA
I'm 24 but I think it benefits a man to place himself in many locales early on
English
@notadampaul @jeremyphoward Surely at least things like discovering/inventing the transformer architecture should count?
English
@jeremyphoward great -- now can you explain why humans haven't done any independent breakthrough scientific research in decades?
nature.com/articles/d4158…
English
If you're wondering why LLMs haven't done any independent breakthrough scientific research yet, I explained *18 months ago* why that's not gonna happen (unless there's a major change to how LLMs work):
Jeremy Howard@jeremyphoward
For those that hope (or worry) that LLMs will do breakthrough scientific research, I've got good (or bad) news: LLMs are particularly, exceedingly, marvellously ill-suited to this task. (if you're a researcher, you'll have noticed this already) Here's why🧵
English
@jeremyphoward My take is that we should start teaching LLMs to run real world experiments
English
@charlesxjyang I can't wait for it to do a Stern-Gerlach level experiment
English
"AI will be generating papers that are as good as their papers"
Thats exactly right, you've perfectly captured how science works. It's not brilliance, its just papers
Natalie Wolchover@nattyover
Anthropic co-founder Jared Kaplan had this to say about the future of physics research. A bold claim that I might not take as seriously if Kaplan wasn’t such a brilliant physicist before he left the field for AI:
English
@eshear @micsolana Point 6 is basically land value tax funding UBI in a way that doesn't immediately get eaten by rent
English
Key insights on the housing crisis:
1) we are obviously building too little housing in ~every major metro, all sensible ppl can see this
2) building more housing is positive sum for people overall esp bc it enables mobility, increases efficiency via density, and also temporarily reduces prices
3) but increasing density+prosperity of a metro over time also greatly increases prices there, squeezing the bottom 30-50%
4) in theory Just Build More Housing except that if only a few cities are building housing they receive disproportionate population inflow and prices never equilibrate
5) therefore ideally we want agreement between cities to all enable construction simultaneously, bc otherwise cities are actually just following voter interests by blocking it
6) either that or we should do an Alaska Permanent Fund in every city but based on property taxes with a 20 year vesting schedule
English
our insane cost of housing is the problem of problems. almost every social ill is exacerbated by a general sense of anxiety rooted in housing insecurity. every politician should have a plan to address it, and if they don't they should to be fired.
English
Read about the best laser-based neutron sources here: nature.com/articles/s4146…
English
Neutrons can see light atoms inside heavier elements, like hydrogen in steel, or lithium in batteries. Reactors still win on raw flux, but this laser makes 100-picosecond neutron flashes, so you could watch lithium ions move during a battery short-circuit
English
Who needs nuclear reactors when you can make neutrons with ✨lasers✨
English
Read more about the race to discover cool new critters: science.org/doi/10.1126/sc…
English
Maybe you're thinking - oh, but that's just because they're discovering lots of new fungi or something. True, but we're also discovering more animals, and even vertebrates
English
Things I did not know: we're discovering new species faster than ever! We found more species in 2020 than any other year. As someone who's wanted to discover a new species since I was a kid, but went into physics instead, looks like there's still hope
English