davidu

Surveillance cameras on the right corners are a really good idea. Excellent way to make policing more objective. I wrote about that here: https://a16z.com/2021/07/13/investing-in-flock-safety/

Turns out, marginalized and minority communities, counter-intuitively, tend to prefer it the most.

As for tracking chip -- we don't do it at birth, because we don't need to. Most people over the age of 15 keep a cell phone on them which provides tracking capability / cellular triangulation. It's used to save lives and apprehend criminals constantly. The privacy issues are actually more marketing based and have nothing to do with the GPS / triangulation issues, and people don't want to solve those, unfortunately.

Drone as first responder makes a tremendous amount of sense -- Police already use helicopters, which are exceptionally expensive, limited in coverage capability, and have disastrous consequences when they crash.

But there are huge benefits of 'eyes in the sky' during a crisis. Car chases in LA are often eliminated when there's a police helicopter because they can simply follow the vehicle safely from a distance as the chopper tracks the car.

Providing overhead visibility support in advance of police arrival is a huge benefit to citizens during 911 calls. If a drone can respond in two minutes, and police are 7-10 minutes away, that's a massive situational awareness benefit that will save lives.

When someone calls for a house fire, having a drone arrive in minutes, potentially with a fire retardant payload, while ground-based fire vehicles arrive 5-10 minutes later, can mean the difference between life and death. At the very least, it provides tremendous information to the fire fighters on the way as to what to expect. They can see the fire before they arrive.

Two major life saving improvements for emergency rooms in the last 50 years were radios in the ambulances so that the ER knew what was coming in, and electronic EKG / vitals relay so that doctors had more precise telemetry of patient status. Drone first responder will have many similar benefits.

DNSSEC is easily the worst upgrade, multiplying complexity and brittleness, with the least amount of net benefit (without even adding encryption), that could have been solved in much simpler ways, that the Internet has ever attempted -- and that's including IPv6 (which is now quite workable).

Speaking as someone who most people consider a DNS expert and actually did help develop and deploy something substantially additive that is in widespread use today (DNSCrypt). ¯\_(ツ)_/¯

The world hates that the US is the world police, but then constantly demands the US be the world police.

Anywhere to get this kind of pie in Manhattan?

(also, Kenji is great!)

The economy and stock market impacts similarly positioned companies all at the same time. And in this corner of the Internet, that seems like "everyone."

There are other companies doing great -- Stryker (NYSE: SYK) is near their all-time high and has not done layoffs as far as I know.

There are wide swaths of the US airspace that do not have strong radar coverage, certainly not at 80,000 feet.

Most of the ocean is unmonitored, too. Remember MH370? We lost it in the air and we couldn't find it in the ocean. :-)

Not sure where your info is from, but this balloon was seen by the naked eye and all over social media. As someone who lives in Montana, many many people could see it. The idea that the USG prompted the publicity is unlikely to be accurate (would suggest a calculated social media psyop). Not impossible, but not likely.

Alternative take, we face population collapse: https://www.acruisingvoyage.com/p/vanishing-people-the-popul...

Giving someone $500 or $2000 once and measuring the impact of that, and then extrapolating it to UBI makes no sense. A one-time cash payment isn't a psychological safety net. It's winning a slot machine once. In fact, you probably know it isn't going to happen again, which could even be worse.

A tremendous achievement in an incredibly short period of time. Congrats to the absolutely stellar Anduril team and to America and her allies for getting better technology to save lives.

Really great v1 -- nice job. Ignore the haters.

Additional mitigations started to come in to play after that to make it both impractical, detectable, and improbable. It's just shy of impossible now, and in the very unusual scenarios where it's possible, it's so easily detectable.

I will miss him terribly. He was the guy who let me crash in his room at Defcon when I was a minor and was too young to book my own room. This kicked off 20 years of friendship. He was brilliant, zany, ADD x 1000, and always endearing and loving.

In 2008 I was lucky to collaborate on DNS issues with Dan when he discovered a new form of DNS cache poisoning and spent years helping the Internet upgrade and defend against a serious and significant vulnerability. We all owe him a debt of gratitude for this and so much of his other work.

He was a hacker's hacker, and the guy you want in your corner. Very sad to hear this news.

If this is Peloton's choice, it's a bad one, and potentially fatal.

Not supporting Apple Health was what killed FitBit, so hopefully Peloton avoids learning this lesson the hard way.

This is the argument made against all software stack advancements. Nothing to do with industry. But when the benefits outweigh the the hurdles, change happens. And if I was starting a new insurance company (which I've considered) I'd be doing our work in code not xls, and probably python. Having RCS, Numpy, unlimited compute, unlimited storage, all gives me an advantage over my competition. :-)

As to the memoization, that is not hard to manage in Python.

Quite a few. Anduril, Epirus, and others. Google “counter UAS” for companies that do this with a wide range of techniques from kinetic (hit it) to microwave (emp like).

This is quite neat!

https://beta.rive.app/ (OSS runtime, xplatform, native rendering)

I appreciated this write-up. The line between "mildly interesting story with some great photos" and "we made a wrong turn or wrong decision and died" is a lot thinner than I think this post makes clear.

There are very few roads and people in the High Sierras, and Red Meadows pack station is basically it when it comes to services.

Surrounded by smoke, changing winds, and uncertain information, they had to try and get information through very low bandwidth, high latency satellite text systems (I have an InReach, and it's great, but leaves a lot to be desired).

I'm impressed with their retelling of the story and am glad they made it out safe.

lol, no, that's the future of democracy.

I have no idea if it's true, it was just one of those comments that you don't forget reading.

Tim Ferriss had this amazing comment once in one of his books that if you want to know what the future of human health is in 5 years, look to what bodybuilders are doing today, and if you want to look at the future of human health 10-15 years out, look to what race horse breeders are doing. Something to that effect. ;-)

It's too hard to fully explain how wrong this is, so just leaving the comment in case someone else reads it and thinks it might be right.

The investors are engineers. Full stop. They write code. The code makes trades. Everything else is noise to support the engineers writing code to make trades.

I don't disagree with this more nuanced framing.

No, but they have a lot of technical content on their corporate blog. They recruit from all the topcoder kind of competitions for those interested in meeting people who work there.

The investment side is engineering.

If you're talking about IT that makes the printer work, sure.

I know people who work there. That comp data excludes bonuses -- and most importantly -- that employees are able to invest in the proprietary funds themselves, which generate insane returns.

For those who don't know, Jane Street is one of the best performing algorithmic trading firms on Wall Street. Probably the largest Ocaml shop on earth.

It's a lucrative job for strong programming talent. Engineers there routinely make well over $1mm/year (edit: after a few years).

They are similar to Two Sigma in tech focus and talent, which is a more well-known firm.

You say this as if those engineers (and it's not 100,000) are all voting in a democratic decision-making process on feature direction. It's a very small number of senior leaders who make these calls.

Remember, Sundar used to be in charge of the Google Toolbar. That was his claim to fame inside of Google before Chrome.