Swizec

> As thinking individuals we can and should use our awareness and courage to make biggest "good" choices we can bear to.

That’s the cool thing about incentives, we follow them whether we realize it or not. “bear to” is where the incentives are hiding. We don’t make the best choices, we choose from the choices available to us.

You can’t fight your environment forever.

If the gooder choice is 50km away and the bad choice is down the street, eventually you’ll stop going 50km out of your way every day because it’s just too cumbersome. You might even do it “just this once” because you had a busy day and your kid is crying and you’re just so tired and then before you know it, it’s been months or even years since the last time you did the cumbersome better thing.

Now that you spell it out, I guess this just binary search. You need O(log n) weighings because each weighing splits the search space in half. So I was wrong it’s not an information theory question, it’s just comp sci algorithms. I got my college classes mixed up (it’s been 14years)

> I also got this riddle, in 2015. I couldn't solve it. Tbh, I think it's a terrible question. There isn't really a step-by-step problem solving process, but you just need to have a "leap" to realize you can weight the marbles in groups of 3.

It’s an information theory question similar to the “how far can you drop the egg before it breaks” puzzle. You don’t need a leap of intuition, just remember the right theorems and formulae from college.

I probably would’ve failed the interview. At best I could write an algorithm that empirically arrives at the answer, which may or may not impress the interviewer enough to pass.

Stuff like this is why I dropped out of coding competitions some time in high school. You can always brute force the easy version, then the next tiers drop numbers that cannot be solved unless you know (or can invent) that one weird maths trick.

This matches my experience. Just Talk To It is the best method for working with LLMs if you're an expert.

I've seen this at work (as eng manager/lead/principal/whoevenknowsanymore) – all the big APIs give you stats. We see how much people burn in tokens and we know how much output they produce. There is a pretty strong inverse correlation between token burn and output.

The more tokens people burn, the less likely they are to produce a good outcome.

I find the most practical distinction of scifi vs fantasy is that scifi calls it credits and fantasy calls it coin. This classification has never steered me wrong.

But honestly categories never quite work at the edges. Some of my favorite books are post apocalyptic fantasy where the characters are definitely using scifi technology that I recognize as advanced versions of existing technology but they just think of as magic.

Zach is good people. I still can’t believe I met him at a random meetup in SF once and he walked up to me to say hi and I didn’t connect the dots until many hours later that he’s the Zach whose blog I’ve been reading on and off my whole career and is one of the inspirations that got me to Silicon Valley.

Zach if you’re reading this, thanks

> Tobias Lutke /../ has most recently advocated for a pay-to-play version of democracy where wealth determines your ability to participate

He’s just looking at america post Citizens United and wishing he could ull the same shenans in Canada. It’s been working very well for the folks and orgs who can afford it

> Eventually I never really looked at the folders but it never bothered me as I don't really see them

Most people don't read most newsletters. Some people read some.

This Saturday I sent an email. It has since been opened by 26% of my subscribers, that's 4,600 people. 780 people clicked on a link.

I tweeted the same email as a link and got pretty good engagement! 20 likes, 2 retweets – that's decent for me these days. It was seen 1400 times. 78 people clicked on a link.

I have roughly the same number of followers as I do email subscribers.

> Owning your own home does not align with the density required for a city to function

Much of Europe would like a word. I grew up in a block of flats where everybody owned. 16 apartments, around 60 people, lived on the footprint of a typical American house. All our parents were homeowners, together they owned the building. Two buildings actually and the land they stood on.

So a total of 32 apartments, 120 people, a small parking lot, and a bit of green lawn owned by the people who lived there and self-governed through a sort of HOA. Little communities like this were extremely common in the city.

Nowadays with my generation homeownership is becoming a little more rare. Still people own apartments in these self-governed buildings, but many have become priced out. Too much demand not enough build.

> We can't prove a program will always halt, but we can check it halts in a specific execution.

The computer scientist says “The halting problem cannot be solved, I have proven it” and the software engineer goes “30 second timeout, if the answer’s not ready by then we don’t need it”

I didn't say it was intentional, it's the result of millions of little decisions that all make total sense in isolation.

> how some group would go about preserving a problem like homelessness, even unintentionally

Simplistic version: San Francisco spends roughly $100,000/year on each homeless person. In services, salaries for people working on it, rent for office buildings etc. I am willing to bet many of these people would not be homeless if we just gave them $100,000/year without all the middle bureaucracy layers.

> Is "medium growth" for the next 4 years really the worst people can think of?

At this point anything less than "medium growth" will crash the economy. We'll have bigger problems if that happens (think 2000 or 2008)

> where a preselected tip is selected. Default should be no.

A lot of this push to higher and higher and preselected tip options comes from POS software providers (Square et al) and credit card companies. They make money on transaction volume. Higher transaction -> more fees

> You really shouldn’t make speeding up to make the light a habit

I agree! In this particular case we were dealing with bumper-to-bumper traffic, a light that takes many minutes, and very few cars making every green light. It took us 20 minutes to wait our turn. I really didn't want to wait for another cycle. 2 more cars rushed behind me.

Like I said: We need rush hour mode

> Then there's the collision detection thing. It's basically guaranteed to beep at me whenever I enter my parents' narrow street with cars parked on both sides.

Some of my worst driving experiences have been with collision detection + auto brake.

You try to enter a narrow steep hill driveway and it slams on the brakes with half your car hanging out into [potentially] oncoming traffic. Thanks, car

Or you try to speed up across a wide open intersection because the light is about to turn and it slams on the brakes because there's cars on the other side waiting for the next light down the block. Plenty of room to stop after you've cleared the intersection mind you, but hte car really really doesn't like that you sped up from 25mph to 31mph when it thinks you should be slowly coasting to a stop.

On the other hand, driving a motorcycle, I love other people's auto brake. Makes lane splitting (at reasonable speed deltas) easier because every Tesla will tap the brakes when you cut into its lane.

Anyway, I wish driving assists had rush our mode. They're pretty decent in average conditions but ho boy tightly packed aggressive rush hour traffic is hell in a modern car. So much beeping and constantly fighting with the assists.

> Just this week I had a package that was supposed to be delivered by Monday that lost tracking and didn't show up until Wednesday.

This is the sort of problem you solve with more funding not privatization.

Here’s how these privatizations work:

1. Cut funding for public service

2. Service becomes bad

3. Cut more funding because service is bad and unused

4. Service becomes worse

5. Privatize

6. Strip the service for parts, a bunch of people get rich, classic PE stuff but worse

7. Start extracting rents, you have a nice monopoly

8. Public has no or worse service for higher cost

Mid 30s

> More and more studies have been indicating that even just a few minutes of intense exercise can outperform long/slow LISS type cardios.

For best results run fast and far. During my personal best marathon (3h 15min) my heart rate averaged in the 170 range

> This must have had an effect, beyond health. Surely?

Watch old music videos and TV or movies, everyone looks so old! You’ll have 30 year olds running around looking 50.

Then you look at modern media and think “Wow look at that 30 year old baby” and the person is pushing 50. Everyone looks so young these days! Cigarettes age your skin super fast.

> And if they want to keep their jobs they have to comply.

Worse still, many mariners are effectively prisoners and can become trapped if corporate decides to cut them loose without also providing passage off the vessel

Stranded sailor allowed to leave abandoned ship after four years https://www.bbc.com/news/world-middle-east-56842506

Stuck at sea for years, a sailor’s plight highlights a surge in shipowner abandonment https://apnews.com/article/abandoned-seafarers-labor-unpaid-...

> the point from GP was that both cars brake/ decelerate at the same rate

Point is that’s not always true. If they are the same type of car, and the car happens to be the kind with downforce, then their rate of deceleration greatly depends on air speed. A downforce car decelerates faster at higher speeds.

This is why you often see race cars lock their wheels towards the end of the braking zone, never at the beginning. The driver has to release the brakes as the car decelerates because there’s less friction available. You go from pulling 4G at the beginning of the braking zone to pulling the usual 1G once your speed drops enough for downforce to become negligible.

Alos! Many non-race cars actualy produce lift. Meaning the faster car decelerates at a slower rate than the slower car (0.8G vs 1G), making the effect from OP even more pronounced.

> the precision inherent to communicating scientific concept /./ abandoned when it comes to prose.

Honestly that was a typo and I noticed too late to edit. Thanks for catching

> The blue car, using ½mv², shed (~70²=) 4900 units of energy (we'll hand wave away the constants). So the red car, which had (100²=) 10000 units of kinetic energy to start, also shed 4900 units, which means it had 5100 units of energy when it collided, and so was going (√5100~) 71

But if the cars produce downforce this is no longer true because you brake harder (more friction available) at higher speeds!

This is how F1 cars pull 4G when breaking. Some custom cars (like one of Ken Block’s last monsters or the Valkyre) use active aero braking to even greater effect.

> While vehicle sizes are indeed trending upward in Australia and Europe, giant SUV and especially giant truck penetration in both countries is still far behind US levels

Many big SUVs are flat out illegal in Europe, you’d need a small truck license. It’s hard to comprehend how large American cars are unless you’ve lived in both places.

For example:

we took an Uber to the airport for our annual pilgrimage back home. We had 4 suitcases this time because new babies need gifts. In USA we got picked up by a sedan and everything fit in the trunk. In Europe we rented an SUV and had to break down the rear seats to fit everything.

> big companies need to self police and if a child can reach their service they have to pay the child like lets say GBP 10k per instance?

HIPAA has been super effective this way. As we all know, American companies don’t give two shits about user privacy or even security. But wave the HIPAA flag and everyone starts caring real hard and taking extremely cumbersome steps to comply with patient privacy.

Very simple: Each HIPAA violation comes with a financial penalty for the business and personal penalty for every person involved in the leak. Very effective.

> (we’re celebrating 250 years - some folks in Europe live in houses older than that!)

We used to smoke weed on the roman wall behind my friend’s high school. Very popular hangout spot. Lots of people using it for rock climbing practice (you’re not far off the ground and can climb laterally for hundreds of meters).

The local castle, about 1000 years old, is a popular makeout spot for teens.

> I lean very heavily towards outcome based prompting. Say exactly what do you want achieved and then maybe give some constraints, ie. what definitely not to do.

I do this when writing stories/projects/issues/epics for humans. Works great.

If you read any management book published in the last ~70 or so years, you’ll find that “Make sure people understand the goal” is the ultimate hack. They even use this in militaries!

“go take that hill” works a lot better than “walk 50ft to the right and shoot at those bushes”. You always get what you ask for :)

> Yet! It is the next frontier

While researching my book I read papers from the 80’s saying this. If you get a good enough spec and define the contracts and architecture, you then just hand off implementation to juniors/offshore/etc

So far has not worked. Maybe this time!

> We’re still missing a good way to express and measure architectural quality

Architectural complexity[1]! There’s several really good papers on this.

Unfortunately it never caught on and we don’t have great automated tools to spit out a number. Also the majority of people just don’t care enough. Research in this field kinda died out when we invented microservices and started treating those as a silver bullet to The Architecture Problem (it’s not [2])

[1] https://swizec.com/blog/why-taming-architectural-complexity-...

[2] https://youtu.be/y8OnoxKotPQ