zelphirkaltstahl - Network

Oh and what I also tried is whether it syncs messages via bluetooth. It does! Its mesh network character really seems to work.One other...

Oh and what I also tried is whether it syncs messages via bluetooth. It does! Its mesh network character really seems to work.

One other thing I found annoying in #briar is, that introduced contacts are always unverified, and there is no way to make them verified. It is just a couple of machines I am using, but I cannot get them to show up as verified.

Trying stuff with Briar. What I like is the idea of forums and blogs. Quite cool to share that between friends (although I would first need...

Trying stuff with Briar. What I like is the idea of forums and blogs. Quite cool to share that between friends (although I would first need someone to share with). What bugs me is, that I made a forum post, forgetting to tap that little "reply" button and now the post is at the top level, and I cannot at all remove it. The forum is now forever tainted by disorder! Another thing is, that in a forum all things are expanded and shown. Can they be collapsed? Desktop app has huge lags sometimes.

Just got the following error, when trying to install a #guix package (guix on foreign distro):```substitute: looking for substitutes on...

Just got the following error, when trying to install a #guix package (guix on foreign distro):

```
substitute: looking for substitutes on 'https://bordeaux.guix.gnu.org'... 0.0%
substitute: looking for substitutes on 'https://bordeaux.guix.gnu.org'... 0.0%
...
substitute: looking for substitutes on 'https://ci.guix.gnu.org'... 100.0%
guix substitute: warning: bordeaux.guix.gnu.org: connection failed: Connection refused
```

Anyone else experiencing this? `https://ci.guix.gnu.org` seems responsive.

Also you cannot properly delete your account it seems. You can only "deactivate" (probably shadow profile afterwards) it.

Also you cannot properly delete your account it seems. You can only "deactivate" (probably shadow profile afterwards) it.

Also they do not have a desktop app for GNU/Linux. I tried downloading the Windows one, hoping to use it via WINE, but the download button...

Also they do not have a desktop app for GNU/Linux. I tried downloading the Windows one, hoping to use it via WINE, but the download button doesn't even work.
The links in the imprint also do not lead to any e-mail address I could find. They only get you back to the FAQs and the chat bot. So no one at home to talk to in a civilized manner. Abysmal customer service. Seems the money I paid for a book is lost. Will never buy a book on #kobo ever there again.

I cannot recommend buying #ebooks on #kobo .com. When you try to download an ebook that you bought, you only get some URLLink.ascm or...

I cannot recommend buying #ebooks on #kobo .com. When you try to download an ebook that you bought, you only get some URLLink.ascm or something like that, not an ebook file. This file according to their chat bot can only be used with some special Adobe shit tool, because it does #DRM shit. The Android app from them, where you could theoretically read the book is not usable on older Android versions. Returning a book seems impossible, nowhere can you find any customer support e-mail address.

Content warning:contains spoilers for advent of code 2024 day 16This algorithm takes an operations table, that contains all the graph...

Content warning:contains spoilers for advent of code 2024 day 16

This algorithm takes an operations table, that contains all the graph specific things, like a function for getting neighbors, given a node or node name. The node structure itself though is part of the module that defines the algorithm. One could probably also make that an external part, but I am not sure what more one would need in a node, aside from a name and an arbitrarily structured value, that can contain anything. So maybe not worth it.

Content warning:contains spoilers for advent of code 2024 day 16Now about part 2 ... I need to find a way to get all optimal paths, not just...

Content warning:contains spoilers for advent of code 2024 day 16

Now about part 2 ... I need to find a way to get all optimal paths, not just one optimal path from A*. Maybe somehow make it not stop after finding the first path to the goal 🤔 But I wanted to have the functional version of A* first. Now I have capacity to think about the multiple paths problem.

Have to say, I could never have solved this in just one day. It took me soo long to even decide how to represent the graph and how to translate into that.

Content warning:contains spoilers for advent of code 2024 day 16And here is the functional version of the A* algorithm, using functional...

Content warning:contains spoilers for advent of code 2024 day 16

And here is the functional version of the A* algorithm, using functional hash tables based on balanced binary trees from guile-pfds (Bless you Ian Price for your work on the data structures, I hope to one day I have your knowledge.):

https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/1b4c9f9a9279cf33c313da665cec1167b189ac66/utils/aoc2024/algorithms/a-star.scm

Here is the solution for part 1 of day 16 in GNU...

Here is the solution for part 1 of day 16 in GNU #guile:

https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/1b4c9f9a9279cf33c313da665cec1167b189ac66/day-16/part-1-graph-approach.scm

Content warning:contains spoilers for advent of code 2024 day 16So it took me a long time to get everything in place to solve day 16 part 1...

Content warning:contains spoilers for advent of code 2024 day 16

So it took me a long time to get everything in place to solve day 16 part 1 of #adventofcode (2024) in #guile, but! Now I have a purely functional version of A*! At first I solved it using an imperative version, but since I have functional hash tables based on balanced binary trees from the guile-pfds package, I am able to use those in place of the mutable standard ones, where I have to "change" entries in them in A*.

And finally, after lots of modification of code from part 1, I solved day 15 part 2 of #adventofcode in #guile [1]. The good thing about...

And finally, after lots of modification of code from part 1, I solved day 15 part 2 of #adventofcode in #guile [1].

The good thing about that is, that now I have quite a few convenience functions for working with integers as bitboards/bitmaps and using them, I can use integers as "arrays" in a functional way, without mutation. Even got a nice way to display multiple such bitmaps in one output, as one grid. Check it out [2].

[1]: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/86ec400f60fb7acd158562870d1053703c9ead41/day-15/part-02.scm
[2]: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/86ec400f60fb7acd158562870d1053703c9ead41/utils/aoc2024/bit-integers.scm

Solved day 15 part 1 of #adventofcode using #guile:...

Solved day 15 part 1 of #adventofcode using #guile: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/0947fe35087f0a86dae9d8e55108ea2500c7588f/day-15/part-01.scm

Since I wanted to solve it without mutation, I could not go for arrays, unless I recreate a whole array at each move. I implemented it using integers as bits, a bitboard approach, that I got to know once when looking at some chess engine code. (Was it Hiarcs?) It is quite performant.

Not sure whether I will do part 2. It seems to complicate things quite a bit. But perhaps less than I imagine.

Proud though, that my first approach is able to solve the puzzle, given good parameters.

Proud though, that my first approach is able to solve the puzzle, given good parameters.

Picked up where I left it, and continued solving #adventofcode puzzles using #guile. Day 14.Part 1:...

Picked up where I left it, and continued solving #adventofcode puzzles using #guile. Day 14.

Part 1: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/3849b79d42f30d4485f1453cc0b2852b61c9636a/day-14/part-01.scm
Part 2: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/3849b79d42f30d4485f1453cc0b2852b61c9636a/day-14/part-02.scm

I checked on aoc subreddit though, whether I am just stupid and don't understand, what I am supposed to look for, or I really need to guess around what the puzzle author means. I came up with a criterion, that actually works, but only if you know or guess how to parameterize it and give it a bit of time to run. To me this was unpleasant.

Is it normal, that #guix needs more than a _day_ to `guix install --no-substitutes wine` on quite powerful hardware (8 cores, and 32 GB...

Is it normal, that #guix needs more than a _day_ to `guix install --no-substitutes wine` on quite powerful hardware (8 cores, and 32 GB RAM)? I see lots of dependencies of #wine being built. Apparently it depends on Python stuff and also Rust stuff, which in turn depends on LLVM stuff ... Not sure whether I am doing something wrong. Maybe no one would consider installing wine through guix?

Content warning:contains spoilers about advent of code 2024 day 12Solved day 12 of #adventofcode in #guile . On one hand I am happy to have...

Content warning:contains spoilers about advent of code 2024 day 12

Solved day 12 of #adventofcode in #guile . On one hand I am happy to have implemented a kind of line scan or what it is called, since I did never do that before, but on the other hand I am a bit disgusted by my own code, having written basically the same function 4 times for scanning from each of the 4 directions and not having found a good generalization for it yet.

Part 1: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/921c6968778479c94446f49242ff9dc0e9682533/day-12/part-01.scm
Part 2: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/921c6968778479c94446f49242ff9dc0e9682533/day-12/part-02.scm

Content warning:politicsI feel ashamed to be citizen of a country, #germany which in line with USA, UK, and others, supports genocide of the...

Content warning:politics

I feel ashamed to be citizen of a country, #germany which in line with USA, UK, and others, supports genocide of the #israeli military in #gaza . Their actions are despicable. It is truly shameful, how little we Germans, especially our "politicians" have learned from our history.

I hope that one day the perpetrators will be brought into a court and judged for their actions. The probability is small, but I hope the international society wakes up and gets them one day.

Content warning:politicsAfD openly disrespecting the victims of the recent attack in Magdeburg. Weidel trying to retcon the motives of the...

Content warning:politics

AfD openly disrespecting the victims of the recent attack in Magdeburg. Weidel trying to retcon the motives of the attacker.

How despicable of her, to abuse such a situation for personal gain. And lets not pretend it was not for her personal gain, haha, that would be utterly naive. She does as right-wingers and populists do all around the world.

Content warning:contains AdventOfCode spoilers for day 12Day 12 of #adventofcode in #guile made me implement functions to find labels in...

Content warning:contains AdventOfCode spoilers for day 12

Day 12 of #adventofcode in #guile made me implement functions to find labels in arrays[1]. I got the term "label" from SciPy, where I believe it is called that. I am not sure why exactly, but I also don't have a better word anyway.

[1]: https://codeberg.org/ZelphirKaltstahl/advent-of-code-2024/src/commit/ef439847a320eb43e09c98ab04ed57c575cc182e/day-12/part-01.scm#L117

If anyone knows a better term for it or has an explanation why it is named "label" feel free to share!