I put my macOS-native Gemini client on GitHub with GPL3. It's not usable as-is, but it does at least load pages and do very basic rendering. Could be the seed of something much nicer.

In Swift, if you want to hack on it.


Anyone have a clipboard history utility for Linux that they would recommend?

Added rudimentary formatting and handling of redirects. The links don't actually work because I haven't gotten it to register itself as a handler for the gemini: scheme.

I've hit that phase of all my personal projects where the initial mania has worn off, and the scope of the things left to do is daunting. Maybe that means I should just GPL what I have and put the source up somewhere, to revisit when the mood strikes.

It also doesn't run properly outside Xcode and I have no idea why.

Show thread

Trying to simultaneously learn Swift and a bit about Gemini by implementing a barebones Mac client. Right now just fetches and displays the raw result. Doesn't look at status codes or format the page. Error handling is minimal. Binary only 302 KB so far!

And all it took was: USB-to-3.5mm adapter -> line-to-mic attenuator -> gender changer -> stereo-to-double-mono adapter

Show thread

At last succeeded in getting some synthwave onto a cassette for my sweet 80s boombox.

I said I was going to leave it alone, but I can't stop picking at AoC day 10 part 2.

Finally got my brute-force recursive solution to solve both of the example data sets correctly (which is a first for me). But after running all night, it still doesn't correctly solve the real data set. I am perplexed by this.

Just got a sweet 80s boombox off eBay. It has 3.5mm left/right mic inputs. Would love to record some stuff from computer -> cassette. If I get a cable to split the stereo channels from the headphone jack on my computer and turn the computer volume down really low, am I ok, or do I need an actual line -> mic attenuator?

I can't stop thinking about AoC day 10 part 2. It's driving me crazy because I feel like I _should_ be able to solve it. But I can't think of any solution that isn't a brute-force recursive walk of the entire graph of possible combinations.

But deeper than that, what I really want to understand is what piece of knowledge others have that I don't that enabled them to finish it quickly. I want to put a name to it so I can study it. Is it a particular math or CS concept?

Still stumped on Day 10, Part 2 at the moment! I'm making a graph of all the possible connections, but I've no idea how I'd arrive at the answer without brute force traversal of every path through the graph.

Not only that, but my brute force traversal only generates the right answer for the first, smaller set of sample data. Not the second. πŸ€”

I'm gonna have to come back to part 2 tomorrow when I'm less sleepy. πŸ˜…

My solutions for day 9:



Thinking about redoing some of the previous puzzles in Python, since I don't know Python, and these seem like good exercises to learn it!

My Day 7 Part 1 solution was so gross I had to do it over!


Code's now half as long and runs in 0.06s instead of 2.44s!

That was a bit of a mind-bender! My solutions for day 7:



My part 1 is so inefficient. πŸ˜… I was really trying to rush through it.

Why I decided to parse out the key-value pairs one char at a time I may never know.

Show thread

Part 1 has a bug where it can read past the end of its buffer, which happened to not get triggered by the rule set.

It did in part 2 though, so a fix was added.

Show thread
Show older

Merveilles is a community project aimed at the establishment of new ways of speaking, seeing and organizing information β€” A culture that seeks augmentation through the arts of engineering and design. A warm welcome to any like-minded people who feel these ideals resonate with them.