dimes

The private equity company that scooped up her music rights, most likely.

I could be wrong, but I don't think the initial Go implementation was a C transpiler. It was written in C, but it did its own compilation.

I wrote Dihedral, a compile-time dependency injection framework for Go [0]. It was inspired by the Java framework Dagger. It worked pretty well, but was a little clunky with Go's syntax. Ultimately, decided it wasn't worth it given the simplicity of manually constructing Go objects in a service setting.

0: https://github.com/dimes/dihedral

Yes, totally agree. Any code that has complicated logic with few / no dependencies benefits from unit testing.

I rewrote the backend on a team I used to work on. The service had a ton of unit tests. Given that this was a full rewrite, those unit test were useless. I spent the first few days writing a comprehensive suite of integration tests I could run against the existing service. These tests directly mimicked client calls, so the same tests should be just as valid for the rewritten service. Using these tests, I was able to catch 90%+ of potential issues before cutting over to the new service.

Personally, I find unit tests to be mostly useless. Every time I touch code with a unit test, I also need to change the unit test. Rather than testing, it feels like writing the same code twice.

You probably have resist fingerprinting turned on

Will one be able to embed a Hetchr ATOM into Hetchr in the future?

Modern bridges made from concrete are not designed to last more than a century. I believe the expected usefulness of something like a cable stayed bridge is around 100 years. Compare that to a suspension bridge which can be used almost indefinitely.

It would be much better to store the blog content directly on the blockchain. This is very expensive to do on Ethereum, but should hopefully get cheaper over time.

Is IPFS really resistant to censorship? It seems like any state-based actor could easily block access to IPFS nodes if they were serving a specific CID.

Yes, fees were included for selling the house, but I did not include capitol gains taxes in the investment calculation.

I made a spread sheet comparing renting to buying a few years ago. The premise was that I wanted to compare my investment in my home vs. paying less to rent and investing the difference in the stock market. The assumptions were something like a 3.5% mortgage rate (30y) and 4% appreciation on my home, vs. a 7% appreciation in the market.

What I found is that on a time horizon of ~7 years, it absolutely makes sense to buy a home. The reason is that when you have a mortgage, you're making a highly leveraged investment. If you buy a house for 100k and put 20k down, then you're 5x leveraged. If you're able to sell the house for 110k. The price has increased 10%, but your ROI is 50%.

There was a definite inflection point after 7 years, however, where the amount of leverage decreases to the point that the higher gains in the market begin to dominate the modest increase in home value.

The alternative methods of transport will be more expensive, which will lead to a decrease in demand.

Why is it statistically impossible? Where one chooses to live is likely highly correlated with the job market for their skillset in that region.

Genuinely curious about alternatives. I have a simple node application that all I really need to do is run npm install and node index.js. Even with such a simple setup, provisioning hosts was a huge pain. Each one needed node installed on it, and I had to deploy the app to each new host. With k8s, I just change the number of hosts I want in a config file and everything just works.

Almost free apartment... I think you mean almost free room for a family of four, inside a three bedroom apartment with two other families.

It’s easy to look back now and say: “They had free health care, everything was great!” But that doesn’t explain why they had to build a wall and use exit visas to keep people in.

That’s amazing! Congrats. One question: why is var only supported for top level declarations? I’m not a compiler or programming language expert by any means, but I thought the general approach was to define a grammar and write a parser for it. I’m curious how global and local declarations diverged so significantly.

Simply paying for a service doesn’t remove the economic incentive for the service provider to add tracking. It will always be more profitable to track users, except in cases like DDG or Brave that stake their reputation on privacy. For instance, I pay for groceries, yet my grocery store tracks my purchases and sells that information. We can’t rely on the market to protect our privacy. Government regulation is needed.

Is it training, or is it convenience? Personally, I don't like soaking beans 24 hours before cooking. In fact, any additional time is too much for me, as bad as that may be. A article in the NYT called The Tyranny of Convenience really crystalized this for me: https://www.nytimes.com/2018/02/16/opinion/sunday/tyranny-co...

As a reference, I think Google Chrome sends a key frame every 90 seconds by default

Usually you’ll have an SFU between the users and the streamer that can limit the number of requests to one every X seconds.

Yes, this is a good point. Intra refresh does reduce variability of the bitrate, but the bitrate is still higher than it would need to be if rtcp was supported.

I was not aware of this at the time of writing, but it solves a large problem we've been having. Thank you so much for pointing that out.

Edit: I've just tried using intra refresh, and it works pretty well, but the key frame interval is still required.

Thanks for the feedback. I've removed it.

In Europe, 0-based indexing is used for describing the floor number, e.g the first floor is one floor above ground level. Americans may find this unnatural, but it’s perfectly normal to Europeans because they’ve grown used to the convention. It’s the same for programming languages. The vast majority of programmers have gotten used to a specific convention and it will probably stay that way until there is a compelling reason to change.

While true that price controls create dead weight loss, the failure of neo-liberal economic policy shows that dead weight loss isn’t always a bad thing and that economic efficiency at the expense of people can have terrible consequences.

Supporting domestic industry, when cheaper global alternatives exist is a form of dead weight loss, but it prevents your middle class from being hollowed out and keeps your society healthy. The same can be said for living wages.

Having tried writing a game engine in Rust, I can't imagine that Rust would become the future of game development. Lack of safety is a feature in game development, because the optimizations required are typically unorthodox and a super strict language slows development down. Additionally, object ownership can be unclear in a game development setting, which typically makes use of global variables for state. The benefits of using Rust over C++ for game development are unclear in these regards.

When a company does an IPO, they hire an underwriter that guarantees that they'll sell a specific amount of stock at a specific price. So the company knows how much money they'll raise by the time the first day of trading arrives.

Java and Kotlin are both great languages. I do wish it was simple to build / run like Go though. Dealing with gradle is a huge pain. The only reason I use Go over Kotlin for personal projects is that setting up and maintaining a gradle project is quite a chore.

I think you can derive enough real analysis to define the Cantor function using only ZFC. You can read the axioms in [1], but they are very simple and “obvious” so to speak. What’s more interesting is where ZFC is known to be incomplete. Cantor’s continuum hypothesis is known to be unprovable in ZFC, which is fun the think about.

[1] https://en.m.wikipedia.org/wiki/Zermelo–Fraenkel_set_theory

The proof is in the pudding.