Elias Mårtenson @loke@functional.cafe

Is it just me, or is Gradle a really horrible tool?

Maybe it's just the documentation, but so far I've spent an hour or so simply trying to find out how to get one module to depend on another.

I have two Kotlin jvm modules, and all I want is that one of them depends on another.

Another idea is to declare local variables in the beginning of a function. Perhaps something like:

{$x ◊ print x←10}

Where $ would be some symbol or keyword indicating a local declaration.

I think this is what Dyalog uses, but I find it to be too verbose in a language that is supposed to be terse. Dyalog is constrained by backwards compatibility which is not a problem in this language.

One of the ideas I have is to use a special prefix to access global variables (like $foo or whatever), then since most functions are global, I'd need to prefix them as well, and that would be ugly.

I need to decide on a syntax for locally bound variables in my APL-like language.

In regular APL, this would assign 10 to x before printing its value:

{print x←10}

The normal case is that you don't want x to leak outside of the scope of the function. I'm considering that being the default and add some new syntax to access global variables.

But if I do that, what syntax should I use for global variable access?

I'm quite happy with my Internet connection speed right now.

I have been working on a new programming language. We've all been there right?

Mine is an APL at its core, with support for partial APL syntax. But I have some features tat should make it possible to work more easily with very large datasets, and various other changes that strays away from traditional APL in order to to be able to add new features, such as better parallelisation.

Some new information. I suspected some memory related issue, so I wrote a C program that allocated 32 GB of RAM, and wrote one byte to each page of memory.

This program reliably hard hangs the computerised once it reaches around 28 GB.

I'm currently running memtest86 but so far no errors.

I'm wondering if the problem is triggered once it hits swap? That's the next test.

So I got this new computer, but whenever I do something heacy in IntelliJ IDEA the entire machine freezes. I'm not even able to log in to it remotely.

I connected to it using ssh, and rand journalctl -f on it to see if it printed anything before it hung, but the last thing printed was some message regarding packagekit.

Any suggestions? It's a Ryzen 7 3700X CPU, 32 GB RAM and a Radeon 5700 XT. OS is Fedora.

Any debugging tips?

I got 7/10. That's a lot better than I had expected. https://www.androidauthority.com/phone-quiz-home-screen-1074262/

I keep hearing media people talk about the self-driving car revolution. Things like "imagine a city built for people, not cars". Talk about getting rid of traffic, etc.

Explain to me: How will self-driving cars reduce traffic or the need for roads?

As far as I can tell, there will still be the same number of cars on the road, assuming people travel the same distance? And you might get even more traffic because some cars will now be driven by themselves with no people.

Blog post: Agda is difficult to use because it uses non-ASCII symbols in its language and it requires Emacs to use.

Me: Downloading agda

I've been looking at IPFS, and it's a pretty cool piece of technology. But, I'm still looking for a good use-case. Does anyone know of something interesting I can do with it?

It's not like you're sharing it with the cashier

Is there a way to move away from SHA-1 in Git, since Since SHA-1 is now severely compromised?

Look at the networking guide in the documentation for Fedora 25: https://docs.fedoraproject.org/en-US/Fedora/25/html/Networking_Guide/index.html

Compare this to the horrible documentation in the more recent versions. They don't even have a networking guide anymore, and the docs they have is limited to how to install packages and a few other useless topics: https://docs.fedoraproject.org/en-US/fedora/f31/system-administrators-guide/