Breza

I agree with you. Subscription revenue has explicitly been part of Tesla's strategy for years. I'm not sure when this statement first appeared in their SEC findings, but this is from their 2021 10-K:

"As our vehicles are capable of being updated remotely over-the-air, our customers may purchase additional paid options and features through the Tesla app or through the in-vehicle user interface. We expect that this functionality will also allow us to offer certain options and features on a subscription basis in the future."

Here in Washington DC we've set up many different types of traffic cameras. One side effect is a reduction in the number of negative interactions the average person has with the police. There are communities where the cops treat people like piggy banks and spend tons of time aggressively enforcing minor traffic violations. That has negative effects on trust, not to mention the number of fatal interactions that have stemmed from traffic stops. Other jobs that can be done by the police in other places are handled by Parking Enforcement, the Department of Buildings, etc.

The recent presence of federal agents and soldiers has reversed some of the hard-fought gains in trust, but my broader point still stands: more automated enforcement of traffic laws has positive effects in how people interact with the police. This effect needs to be balanced against the harms of increased surveillance.

[1][https://www.nbcwashington.com/news/local/transportation/poli...]

This is a weird suggestion, but the American Girl Doll book Samantha Learns a Lesson has the most ELI5 explanation of the pitfalls of the Industrial Revolution I've ever seen. I mean that literally, my five year old loves the book and the movie based on it. This era is my favorite in all of human history, I've read thick tomes about it, and this old children's novel is seriously fantastic.

R! If you're a data person and you've never used R, give it a shot. It's a lovely language for cleaning and analyzing data, and the core development team keeps making improvements.

Those moderate Trump voters are a big deal, regardless of their numbers. If two out of a hundred Trump voters had gone the other way, we'd have a different president right now.

This sounds like the kinds of healthcare claims data that I often analyze for work. Just in case you have a team of seven doctors working on you at the same time while you're in hospice, they keep a column for just about everything.

I've found it's easier just to keep things NULL in many cases. Sure, if you have a student database with the column favorite_sport, maybe it makes sense to use the value "Not a sports fan." But think about an email list with the column unsubscribe_date. For active recipients, you'll want that field to be NULL.

I agree with you. I also find that adding DISTINCT can sometimes make it easier for my colleagues to understand code, especially when I'm using multiple CTEs and it might be easy to miss a one-to-many join.

This assumption got me in trouble as a junior analyst years ago. I was asked to analyze our customer base and wrote something like the below. Management congratulated me on finding thousands more customers than we'd ever had before.

SELECT zipcode.rural_urban_code, COUNT(*) AS n_customer FROM customer INNER JOIN zipcode USING(zipcode) GROUP BY 1;

Why not host in house? If you have an application with stable resource needs, it can often be the cheaper and more stable option. At a certain scale, you can buy the servers, hire a sysadmin, and still spend less money than relying on AWS.

If you have an app that experiences 1000x demand spikes at unpredictable times then sure, go with the cloud. But there are a lot of companies that would be better off if they seriously considered their options before choosing the cloud for everything.

I totally agree with you. Where I work, we self-host almost everything. Exceptions are we use a CDN for one area where we want lower latency, and we use BigQuery when we need to parse a few billion datapoints into something usable.

It's amazing how few problems we have. Honestly, I don't think we have to worry about configuration issues as often as people who rely on the cloud.

Not GP, but my company also self-hosts. We rent rackspace in a colo. We used to keep my team's research server in the back closet before we went full-remote.

Whoa, did not expect to see a Sony Clie reference today. I had one in college! I loved taking notes with a folding keyboard, listening to music while playing games on the shuttle bus, and taking pictures with my friends. Oh man, memories.

> One tool alone doesn't usually solve any problems.

I completely agree. The world of developmental economics has had so many great "One tool to fix everything!" ideas, but at the end of the day, they generally don't add up to much without a functioning government that's focused on serving its citizens.

So in 2025, what self-hosted options does a 15 year old have to manage his finances using Bitcoin? Assume he gets a monthly paycheck of 0.00125 BTC. How would his employer pay him? How much would be eaten up by fees?

Back in 2013, I loved the idea of Bitcoin. Then I actually tried using it. Such a pain. I switched to Coinbase until I gave up entirely on crypto around 2017 and became highly skeptical it was going to change the world as promised. I would love to hear that the world of self-custodied Bitcoin has become less onerous.

Why not just use cash at that point? Crypto doesn't make it any easier to create your own social institutions. It just adds volatility, complexity, and risk.

What are your thoughts on using JAX? I've used TensorFlow and Pytorch and I feel like I'm missing out by not having experience with JAX. But at the same time, I'm not sure what the advantages are.

Yeah, and there are areas where it's really hard to tell if something is accurate, like astronomy.

To be fair, I got into distance because I wanted to run without pushing myself to the highest heart rate zones, which you naturally do in shorter distances. But with an ultra, I can make a zone 2 run sound impressive. So even an ultra can be a form of moderation. At least I'm telling myself that.

Motivation has to come from you, of course, but you might find it once you see your future baby and realize you have a shot at really seeing how he or she turns out as an adult. That was the impetus for me to start taking my health more seriously.

I feel the same way. I remember thinking back in the day that every office would have a cluster of old servers to fire up once a month for a big data task. Instead, you can get massively more power for less than $100 of credits on BigQuery.

The progression in difficulty is great, really hooks you then ramps up

I learned to type in the nineties and I used two hyphens. I also learned to put two spaces between sentences but dropped that in the oughts.

There are a few days a year where it would be a major challenge. I've long wondered if there will be a time when AVs are good enough that it doesn't make sense for humans to make a living driving but not good enough to handle edge cases. Imagine leaving the Garden after a Rangers game, realizing it's snowing, and seeing the streets devoid of cabs.

Unpopular opinion, but I actually like traffic enforcement cameras. They don't know what race you are, and they never end up escalating to using lethal force.

I completely agree. Here in DC, we have sporadic enforcement of things like fare evasion, reckless moped driving, unlicensed food trucks, and speeding on the shoulder of the highway. It definitely helps somewhat.

Here in Washington DC, we've installed a variety of different kinds of traffic cameras. And the local police engage in targeted enforcement actions against things like driving on the highway shoulder or recklessly operating a moped. It makes a difference. An unexpected upside is that the regular police spend a lot less time catching speeders and stop sign runners, which frees them up for more serious situations.

What Python libraries do you prefer? Even after doing this for years, I have trouble making anything remotely complicated in matplotlib without at least one look at the documentation.

For data viz, I'm absolutely smitten with R and ggplot. It works the same way as my brain, "OK I want to use the students dataset, specifically the age variable, I want to make a histogram, and I'd like to label the axes." You build the viz in that order, with one function call for each thought.

I guess I'm lucky to have learned it in 2012 when base, ggplot, and lattice were all being taught at the same time. I love ggplot, but I dip into base from time to time when hist(students$age) is faster then ggplot(students, aes(x = age)) + geom_histogram()

I totally agree with you. Some of the most intense thinking I've put into writing code has been when I go out for a run in the middle of the day and come back 30 minutes later with solutions to the thing I spent all morning trying to fix.