* Formerly: NASA JPL, Amazon Robotics, Outrider.ai, shield.ai
## What
Enjoys a little exercise, wishes for more time to build and read. Obsessive about embedded systems, planning algorithms, and applying optimization to far too many things.
## Web/Contact
* https://josh.vanderhook.info
* https://jodavaho.io
* https://mastodon.social/@jodavaho
* https://x.com/jodavaho
## Keybase Verifier
[ my public key: https://keybase.io/jodavaho; my proof: https://keybase.io/jodavaho/sigs/LyGOcHPgNACB2NqAN9IU0Ps0Upm4nLekjyF12kgKNDo ]
One counterpoint is that the "labor as TAM" argument is far larger than it needs to be. Only a fraction of it needs to be captured to justify all the capex and make 5 new companies displace FAANG, and this does not have to translate to unemployment to succeed.
The difference in the unemployment vs efficient employment model is mostly user driven adoption vs company mandated adoption, or centaurs vs reverse centaurs.
GGP didn't break the law (in fact he conformed wtih it), he just didn't disobey his orders (risking his salary) in order to go the extra mile in order to make a negligible positive impact in the world. Seems rational.
Ever speed on the way to/from work? That's a worse offense than a strictly-compliant website - you're increasing danger for yourself and others and breaking a law, precisely for your own benefit.
It is very concise, and reads precisely as you suggest: to exploit properties already discovered and therefore combined in a novel way.
I'm just delighted by the prose. It reads like an old paper. The ones that were just straightforward theorems with proofs that do exactly what they say.
Viewed independently, I think VZ, Ukraine, and Iran look like a hodgepodge of strange actions. There is a consistent view that joins them, namely as we approach the end of the davidson window, putting pressure on China and Russia via oil restrictions might help dissuade a larger war in the Pacific.
This is just a personal opinion, but lines up with Ukraine (burning refineries), EU (seizing sanctioned tankers FINALLY), VZ (cutting exports to China and moving to USA), and Iran (close the strait every so often). The unifying thread seems to be applying consistent pressure around China.
The multi-sensor comments are confusing. This issue is a command->semantic understanding problem, not a sensor fusion problem or trajectory planning problem per se.
It's not like the true depth of field is important for the robot to plan when it's moving at turtle speed and can stop quickly.
I should have stated something weaker: There are legitimate arguements, even on this forum, against geofencing, so entertaining arguments against camera-based always-on tracking shouldn't be automatically out of scope.
> You went from recording license plates on public roads to logging people in public places.
No, man, the argument is in the linked article, the one from TFA, the one I quoted. It is about public spaces and recording "what anyone standing on the same street could already observe".
That is insufficiently restrictive of a criterion because it's overly broad, and therefore can lead to absurd situations we'd never expect. Like kid tracking mafias. Any device that records "what anyone on the street observe" becomes awful creepy real quick, and we shouldn't accept that kind of argument ever.
TFA goes on to say that it breaks down at scale, and I'm trying to call out that, no, it is creepy at local scales too, because recording public activity is a bad precedent to set. I don't know how to make it more clear.
> stolen car
Having a use for an overbroad surveillance tech is only a defense for you, not for me and probably not for the courts either. It's not like they only activate it when there's a car stolen. It records all the time.
We've stated our opinions clearly now, I think the back and forth can end.
I'm calling out, specifically, that
"A camera capturing an image of a license plate that is openly displayed on a vehicle is not searching for someone's private life. It is recording what anyone standing on the same street could already observe."
... implies that a very absurd and objectionable thing like folks standing around each playground recording children and comparing notes is actually also supported by that defense and that we should consider if that defense is objectionable or not based on what it enables as much as what it is defending.
On this very forum, you can find backlash against geofencing, and here, support for flock cameras? The contradiction is bananas. Automated logging of people in public places is dystopian. You can object with that claim, fine.
It's commercialized stalking. A group of people who physically stood in each park every day photographing children playing and sharing those images between each other would be run out of town so quick. Laws would be invented in a heartbeat to stop this.
Agree, but more specifically Math is clearly about a human understanding structure of things. Math is basically for humans. It's one of the main reasons understandable proof is so important.
Love the analogy.
I a lot of people had their mind blown by vector space representation of words. The idea that Female + King averaged out around "Queen". Words might foundational to concepts or just really-well-designed ways to transmit them.
* Planning Scientist at Flight Science.ai
* Formerly: NASA JPL, Amazon Robotics, Outrider.ai, shield.ai
## What
Enjoys a little exercise, wishes for more time to build and read. Obsessive about embedded systems, planning algorithms, and applying optimization to far too many things.
## Web/Contact
* https://josh.vanderhook.info
* https://jodavaho.io
* https://mastodon.social/@jodavaho
* https://x.com/jodavaho
## Keybase Verifier
[ my public key: https://keybase.io/jodavaho; my proof: https://keybase.io/jodavaho/sigs/LyGOcHPgNACB2NqAN9IU0Ps0Upm4nLekjyF12kgKNDo ]
jvanderbot.at.hn