njoubert

Can you say more?

I had the pleasure of meeting Andrew during the height of the 3D Robotics days, and spending time pair-programming with him on a GPS Driver for ArduPilot. It was incredible to see his confidence in moving around the codebase, making edits. It left an impression on me, it was watching someone who really knew what he was doing. I was inspired. He was also very jovial and fun.

I also remember how concerned he was about the safety of flying vehicles. We were all on the rooftop of a building in Berkeley showing off hackathon projects on drones. He was pointedly staying under the one roof overhang, telling people that this is definitely unsafe. Minutes later someone crashed a drone at full speed into a utility pole, sending shrapnel and lithium battery chunks everywhere. How it didn't catch fire I'll never know. I think 3DR was a little cowboy for his taste at times, he.

Tridge, if you read this, it was a real pleasure working with you!

Don't bring the finished product to the site, bring the manufacturing capabilities and raw materials to the site.

You're painting an unfairly one-sided picture here, as so many pro-psychedelic people do. These are highly potent chemicals that also have negative physical and psychological effects. It's not as simple as "these substances cure a condition and the fear of psychedelics keep them repressed". You are saying the equivalent of the college stoner friend saying "you can't get addicted!" about weed. Even this article points to fatal heartbeat irregularities as a dangerous side effect that has limited ibogaine's study.

Great comment! Indeed there is a recent Nature paper that's causing a lot of discussion that is investigating that missing link between the physics of matter and the emergence of the natural selection and evolution processes that give rise to life. this is a really fascinating area! https://www.nature.com/articles/s41586-023-06600-9

this is a pretty silly fluff piece.

Thank you for this, this is awesome. You shared a long, challenging, and deeply personal experience with candor and empathy. Well done!

I grew up in an agricultural area and everyone had respiratory issues from the massive amounts of pesticide and pollen in the area. I was also allergic to house dust mites. It was bad! I remember being unable to breathe through my nose at all for years, having surgery, and completely changing my experience of life. I've recently noticed that i'm having quite a bit of trouble breathing at night, and those old demons are coming back. Your writeup is inspiring me to do a closer investigation as I have been experiencing very similar issues - brainfog, memory loss, all the above.

Your comments is also on point around the shortcomings of OneMedical-style rotating cast of practitioners rather than the longer term relationship with a caregiver.

Ah, tailsitters. Everyone loves the theory, then the practical realities of the takeoff position exposing the entire wing surface to the prevailing wind quickly kills real-world applications. I remember back during the 3D Robotics heyday watching Chris Anderson repeatedly run down to set his tailsitter upright, take a few steps back, only to watch it fall over again.

Lots of denial and sarcasm in the comments here. A person close to me had many of the experiences described in this, including the psychotic episodes that I had to help them through. Try to imagine how frightening and difficult it is to see someone you care about experience that, from a substance that's portrayed as safe and harmless and fun. they are still working to get back to a healthy place after years of cannibis abuse.

Thanks Mike. Big hugs.

Peter, I'm lucky to have called you a friend. This happened to suddenly and quickly, I'm reeling. You were magic.

He exuded love and charm. He would be overjoyed to see me and give the best hugs whenever we ran into each other. He is this super accomplished person but that was never the conversation. I've known him for years and it's only now that I discover his LetsEncrypt involvement. It speaks volumes to him, he was so focused on everyone around him and filled with love for them, never self-promoting, just loving and being amazing. He would give the best hugs, and few seconds longer than most, and you could hear him smiling while he does so. Thank you Peter

I grew up in an area like this and it was pretty great!

Wow! Which direction from SF??

I find it hard to believe any article that describes taxi services as "efficient but undercapitalizes". Does the author remember the experience of trying to find a cab if you weren't in a downtown area already?

That's not quite right. There are learned models inside the planning systems of most of the sophisticated self-driving car companies. Here's a public video that gives a few high-level examples: https://vimeo.com/618478174

Full disclosure, I work for Aurora in the Planning group.

$1500/hr to $2000/hr was the going rate for Stanford professors to be expert witnesses on patent law cases when i was consulting at DLA Piper.

There’s a pandemic that forced everyone inside and remote. Computer companies does great. End of story.

Do you have any experience with the stuff you’re making such bold statements about, because I have; and my god did you oversimplify things.

The article title is still misleading. The sensor fusion was the root cause of the malfunction, it was not the advanced control technique that was deployed to survive the anomaly.

I see no mention of "advanced control techniques"? Sounds like there is just a limit on roll/pitch angles and a limit on distance applied. Saying "advanced control techniques [for multirotors]" sets an expectation for something along the lines of Tedrake's Underactuated Control approaches: http://underactuated.mit.edu/

The single benefit of Uber having a global presence likely already outweighs the value of these local coops.

You can get off an airplane just about anywhere in the world and get an Uber. That’s huge value.

These are different efforts from ATG, the self-driving car division.

There are two ways to compensate with stock: purchasing stock on the open market with cash, or, hand out stock from a pool of withheld shares. The pool of shares might be periodically increased through creating new shares (and diluting) but this must be reported to the SEC and is only done infrequently.

GAAP reporting requires that stock compensation is written off as an expense. It does not reduce the company’s cash, but it does reduce the company’s earnings. Hence it changes the price to earnings figure.

This is out of date and not representative of current Uber business. Keep in mind Slack did not exist for several years after Uber was founded. Also, today all of Uber uses Slack, which gave them a nice stock bump. https://markets.businessinsider.com/news/stocks/slack-stock-...

Only 3800 are core engineers (people reporting to Thuan) https://techcrunch.com/2020/04/28/thuan-pham-who-fled-vietna...

What the fuck? “Should not get more than market rate”?

This is very strange. Why is WaPo spending so much energy attempting to discredit self-driving cars?

Not an annoying question whatsoever.

Dense urban environments are challenging. The error distribution is more complex there, its hard to definitively say. The numbers I'm quoting is for open sky environments where you have plenty of satellites and little multipath.

For the average end-user, these are $500 GPS units (this is the one I work on: https://www.swiftnav.com/piksi-multi). The technology is already moving down the supply chain to ~$10 GNSS units from Broadcom and ST (we're doing this: https://www.swiftnav.com/news/swift-%E2%80%8B%E2%80%8Bnaviga...)

The issue with phones is mainly an antenna issue. Your phone antenna is tiny and squeezed in alongside half a dozen other antennas and radios. This degrades the GNSS signal quality a lot. A stopgap here is external GNSS antennas for your phone: you might plug your phone into your car and get lane-accuracy turn-by-turn directions, but it's unclear when we'll have high accuracy GNSS inside phones, but most major phonemakers are indeed working on this.

Modern GNSS can reliably measure altitude at <50cm 2-sigma accuracy using techniques such as RTK and PPP-AR. Modern GNSS is a better source of altitude data than barometers.

Likely SpaceX is using a combination of radar altimeters and GNSS.

Source: I do engineering on modern GNSS

This is a common misconception. It is not a small amount of noise that causes misclassification of images. It is a carefully designed and quite unique pattern that causes misclassification. It only looks like noise to the human eye, but it really isn't.

Yes, neural networks are susceptible to adversarial attacks. No, just adding noise to an image doesn't break neural networks.