| Tweetovi |
|
James Ough
@james_ough
|
15. srp |
|
Yes & the astonishing thing is how localized it can be: often a 10X difference in neighboring streets.
e.g. here's a high-resolution map of NO2 concentration in SF pic.twitter.com/Z01lstOVkA
|
||
|
|
||
|
James Ough
@james_ough
|
11. tra |
|
A young Henry Ford pitches the gasoline engine to Thomas Edison.
Edison replies: 'Young man. You have the thing. Keep at it'
Ford writes, much later: 'No man up to then had given me any encouragement'
quod.lib.umich.edu/g/genpub/39366… pic.twitter.com/IOwcnVcw8D
|
||
|
|
||
|
James Ough
@james_ough
|
6. tra |
|
What's today's version of the homebrew club/crypto-counterculture?
Small group of tight-knit people obsessed with a piece of future tech which other people are ignoring
|
||
|
|
||
|
James Ough
@james_ough
|
30. ožu |
|
it's just an amazing combination of unfettered enthusiasm and extremely detailed technical treatment that you rarely see together today.
bring back 1980s NASA pic.twitter.com/s7ij1li3Ln
|
||
|
|
||
|
James Ough
@james_ough
|
30. ožu |
|
the most interesting bit IMO is the categorization of every important industrial process (i.e. what do you ACTUALLY need to run a civilization), and the analysis of 'closure'; the degree to which the facility is self-sustaining.
|
||
|
|
||
|
James Ough
@james_ough
|
30. ožu |
|
In 1980 there was a NASA conference on building an automated, self-replicating manufacturing facility in space.
You can read the conference summary. Wild and compelling.
ntrs.nasa.gov/archive/nasa/c… pic.twitter.com/MHyvLQNM1s
|
||
|
|
||
|
James Ough
@james_ough
|
13. velj |
|
The game has begun! Which path do you take?
Press A to try to build AGI to solve the problem for you
Press B to work in Biology and push longevity research
Press C to sign up for Cryonics and hope someone else wins instead
Press D to Die after an honest threescore and ten
|
||
|
|
||
|
James Ough
@james_ough
|
13. velj |
|
You might not have to die.
It's probably just a technical problem.
You have 50yrs to solve this problem.
good luck
|
||
|
|
||
|
James Ough
@james_ough
|
11. velj |
|
Meanwhile in China:
Massive geoengineering project (translated as "Sky River") underway to solve China's water-shortage problem by cloud-seeding an area twice the size of Texas with silver iodide.
scmp.com/news/china/soc… pic.twitter.com/NDCavYvdsI
|
||
|
|
||
|
James Ough
@james_ough
|
17. sij 2019. |
|
Announcing the formation of the 'Accelerating Science Society'
Now accepting donations + your wacky ideas
Interestingness guaranteed
|
||
|
|
||
|
James Ough
@james_ough
|
13. sij 2019. |
|
oh no
|
||
|
|
||
|
James Ough
@james_ough
|
13. sij 2019. |
|
|
||
|
James Ough
@james_ough
|
13. sij 2019. |
|
Live-stream the first prototype station spinning up, as the astronauts gently fall to new ground.
Could have similar emotional impact to FH boosters landing.
Paradigm shift. pic.twitter.com/zs7mxqPwSm
|
||
|
|
||
|
James Ough
@james_ough
|
12. sij 2019. |
|
longevity realpolitik
|
||
|
|
||
|
James Ough
@james_ough
|
9. sij 2019. |
|
From 'A Pioneer Remembered -- Biographical Notes about Arthur Constant Lunn': pic.twitter.com/CK7XgmrBvm
|
||
|
|
||
| James Ough proslijedio/la je tweet | ||
|
Adam Marblestone
@AdamMarblestone
|
1. sij 2019. |
|
I started a blog in order to answer your questions: longitudinal.science.blog/2019/01/01/on-…
|
||
|
|
||
|
James Ough
@james_ough
|
31. pro 2018. |
|
Genes, Culture, and Personality: An Empirical Approach:
dropbox.com/s/i3ko708slj18…
|
||
|
|
||
| James Ough proslijedio/la je tweet | ||
|
Emma Salinas
@emmalsalinas
|
11. pro 2018. |
|
Great dinner with great people :) pic.twitter.com/hXyjLwG6Ln
|
||
|
|
||
|
James Ough
@james_ough
|
27. stu 2018. |
|
On the hardware side:
Stanford 235 is OK for mechanical design: youtube.com/user/StanfordC…
The ODrive forums forum.odrive.com and discord server are OK for motor control.
But you'll really have to just buy some motors + play around, there's not much good info online.
|
||
|
|
||
|
James Ough
@james_ough
|
27. stu 2018. |
|
On the software side:
@svlevine and @chelseabfinn's Deep RL course is the best single resource I've found: rll.berkeley.edu/deeprlcoursesp…
Not sure it's a good use of time to learn traditional robotics algos... but SLAM (orbslam, lsdslam) and Kalman filters probably useful.
|
||
|
|
||