ema

Looking at yearly solar irradiation actually underplays the disadvantage we have in northern Europe in terms of PV. If all we had to do is built twice the capacity that would be one thing but our winters are quite dark, making PV energy generation almost negible around the winter solstice. Whatever generation capacity we can displace in the summer we still have to keep around to make up the shortfall in December. I love when I see solar panels here but rationally speaking Germany and the Netherlands should be among the late adopters.

It's easy but it ain't simple. Because `f` can do arbitrary things to `x` you have to look at it just to know the general shape of the computation. Reduce gives you a lot of the flexibility of imperative programming but with that also a lot of the problems. Sometimes that's the right trade-off but it is good to be aware that it is a trade-off.

For me it's about the speed of understanding what the code does. Because filter and map are very constrained in what they can do they quickly tell me a lot about the shape of the computation I'm working with. In contrast reduce is much more flexible and so I need to do a much more detailed analysis to just answer a basic question like "is the result a scalar or another collection?".

Please have a look at the Hacker News guidelines https://news.ycombinator.com/newsguidelines.html

To gain more experience with Rust I'm implementing an interpreter for a Lisp-like language in it. As an additional challenge I've committed myself to record the entire development process and put it on YouTube. https://github.com/ema-fox/amber

Read the reply you got more carefully, especially the last sentence is key.

I think Gothic II did this really well. While there are some clear borders a lot of the transitions between the three categories are gradual and organic. You're mostly safe in the secondary settlements but go too far into someone's backyard and you might suddenly find yourself fighting for your life. It also managed to make the place feel expansive without any place feeling like repetitive filler content.

Centralized infrastructure is fragile and to the extent that the internet has become centralized unscheduled Internet shutdowns are bound to happen. The benefit of scheduled Internet shutdown is that people can prepare for it while at the same time gaining experience which helps with dealing with an unscheduled Internet shutdown.

The reason you're not hearing any difference between the words in those pairs is because they are pronounced the same. At least according to Wiktionary and my own subjective judgement as a German native speaker.

There has been open source before Linux and there will be open source after Linux. Yes Linux is a flagship project but the whole culture of open source is much broader than it.

> prevent this scenario from ever happening again.

Every additional nine of not getting hacked takes effort. Getting to 100% takes infinite effort i.e. is impossible. Trying to achieve the impossible will make you spin on the spot chasing ever more obscure solutions.

As soon as you understand a potential solution enough to implement it you also understand that it cannot achieve the impossible. If you keep insisting on achieving the impossible you have to abandon this potential solution and pin your hope on something you don't understand yet. And so the cycle repeats.

It is good to hold people accountable but only demand the impossible from those you want to go crazy.

While I'm always ready to throw away code when I realize that there is a better way to do things I found it quite difficult to write code with the intent to throw it away. However I often do write code with intent of modifying it once I have a better idea of what is needed. It might be because I'm comparatively better at refactoring than at starting from scratch.

The paradox isn't that as a good becomes cheaper we're using more of it. The paradox is that as a good becomes cheaper we're spending more on it.

While STM was a big selling point when Clojure in practice it's actually very rarely used. The persistent data structures are indeed the heart of Clojure.

While for many applications Clojure's performance is good enough it's not anywhere near what you can achieve with Rust. I once did a small game in Clojure trying to be very clever to eke out every last bit of performance and still didn't hit an acceptable frame rate. Made a very naive reimplementation in Rust that involved copying the entire state every frame and it run buttery smooth.

If there is a task for wish persistent data structures are the most performant solution it should be easy enough to implement and use them in rust too. Probably someone already did that.

Clojure is my default programming language but if I want performance (or static types) I reach for Rust.

Maybe the result of NaN === NaN should be neither true nor false but NaB (not a bool).

I've found that when I'm listening to recordings of me my accent really sticks out to me in a way that's completely inaudible when listening to myself live. This happens with both English and my native German.

One thing you can do is listen on a non-standard port.

I suspect that for someone who reads a decent amount reading speed is also bottlenecked by the rate by which we comprehend the ideas being communicated and not the rate at which we recognize the words. I recently did a bunch of reading speed tests in all the languages I understand. I have like two orders of magnitude more experience in reading German and English than in Dutch, French and Spanish. In the former group word recognition is automatic while in the later I have to concentrate and sometimes even sound out words in my mind yet my actual reading speed for my "strong" languages is only about twice that of my "weak" languages.

Translating is a completely different skill that you have to train on top of being fluent in more than one language. The only way I can translate is by sort of forgetting something in one language and then remembering it in another. I suspect if I did this for like a thousand hours hearing something and then repeating it in another language would be as easy as switching the language in which I'm thinking. I think the real difficulty of simultaneous translation comes from having to speak while you're listening. Consider recording your response to an audio message while listening to it, that would also be very difficult but there is only one language involved.

Depends on the length of the blackouts, if it's more than a day then solar panels will allow you to lower the amount of batteries you get.

I think half life already is well on its way to be considered high brow. I played it a few years ago and while I did enjoy it part of my motivation to play it was "this is what the people who have more experience than me think is good".

I mean even when I'm working on my own projects I'm cleaning up whatever code I wrote when I didn't yet know as much about the shape of the problem.

It doesn't only need to make economic sense now but you also need to be fairly certain that all the battery capacity that is likely to be added to the grid in the future will still allow this to be profitable until your expected break even point.

There are many ways to get around this since it is trivial to write code that strips those tags.

It's not quite as effective as a real social connection but the para-social connection you get from say listening to a podcast in the language you're learning does work too in my experience.

I'm making a website for learning languages through comprehensible input: https://input.emanuelrylke.com/

The idea is to aggregate videos by language and then sort them by difficulty.

I always assumed that chess engines are dumbed down by occasionally inserting random moves but now it occurred to me that a way to get more realistic mistakes might be feed them a distorted version of the game state.

There's a big difference between being wildly over-optimistic about your odds of making it back and expecting from the get go you won't make it back.

If you had to choose between everyone dying and everyone except billionaires dying what would you choose?

No, if you start out with 5% probability it means that you have a 95% probability that further information will eventually downgrade it to 0% and 5% probability that it will eventually upgrade to 100%. So further information making a low probability event even less probable is expected.

Speaking is use it or lose it[1] but understanding stays. If you spend on average twenty minutes a day listening to content that is at a level where you understand most of it for twenty years at the end of it you will have no trouble understanding that language in most contexts. You can even have year long gaps as long as the average works out. Learning to understand a language isn't hard (if there's sufficient learner content) it just takes a lot of time. Of course if you have enough free time you can do 4 hours a day and get there in less than two years. Once you are at that level you are in an excellent position to learn to speak the language even if you have to repeat that part of the process if you leave long gaps.

If you want to give Spanish another go I can recommend dreamingspanish as they have just a ton of excellent learner content on various levels.

[1] to the point that I sometimes have trouble expressing certain concepts in my native German even though I live in Germany because so much of my life happens in other languages.