Now thinking about it, I wonder though if there's a broad category of problems where it's useful to represent ranges with A..B containing X where (B < X) == (X < A), which is to say, double-inclusive if A<B and double-exclusive if B<A.

On an additional tangent from my tangent, I wonder if it's feasible to conduct niche elimination via algebraic invariants this way, i.e. we know that a standard range has low<high, so you can tag a union of a range and anything that fits into two fewer bits than it by setting low.msb=1 and high.msb=0 and then populating the rest of the range with the alternatve value.

hmm, wonder if Youtube or its owners might have an incentive to disenfranchise the dead internet theory /s

Teams has a record button in every call, you just need to click it. That button has saved my bacon multiple times.

"okay that sounds good, can I have that in writing for future reference?"

@whimsical I'm sorry, that was completely unfair. It works fine and it does a bunch of important things very well such as mobile support. I don't like to dump low tier requests on people unless they actively solicit it like @blindxfish did on https://devplace.net because I know how demoralizing it is to receive a laundry list of polishing tasks about a side project when I'm not interested in it, so what I should have said is that it's not as polished as the classic interface on https://dr.molodetz.nl , because rantii isn't lame.

Top-exclusive ranges are the standard because they make more sense than bottom-exclusive ranges, and because making exactly one side of the range inclusive makes all sorts of math a lot easier. If both ends were included, an empty slice would have to be either unrepresentable or it would be encoded as end + 1 = start where neither end nor start are actually included.

Also, rantii.molodetz.nl is a lamer client but it works on mobile.

dr.molodetz.nl for alternate client

devplace.net for a possible successor site though most of the community have not embraced it yet

I did it again and accidentally destroyed a bunch of testing data that's nontrivial to replicate. On the bright side, it wasn't prod.

The overlap between the use-cases of nails and screws ends at fixing pictures to the wall. If you understand the things the LLM creates and you're supposed to review as well as you understand handywork, I'm not surprised that you think its output is good.

Israel and the US are betting heavily on Iran's inability to complete the project, or rather their own ability to sabotage it. It's a dangerous game, but given Mossad's track record, I think the confidence is not misplaced.

I made a kickass chicken curry for 2 friends. I've been procrastinating for a year over getting into the habit of cooking but I was nearly fired in December so I had to confront the fact that I would have a very comfortable rainy day fund by now if I didn't eat out so much, and I'm trying to make the most of this newfound motivation.

hank you!

@regeneratoor nooo

I mainly get those errors when I'm trying to run a game on a video card that doesn't support some features it uses, which is difficult to mitigate. My impression is that devs add these aborts into error handlers for hard errors that can only really trigger outside their platform support bracket. Which is fair, that is why you have a platform support bracket.

@Lensflare I also would like to know why submodules are a bad idea for development, I'm considering using them

I rewrote the request-notify duplex comm primitive on top of it, it works well, and testing is a bliss

And that's a stepping stone laid.

https://crates.io/crates/...

I'm excited because this should pave the road towards testing some important and complex subcomponents, not even considering how it will enable dylib plugins.

@retoor no deadlock I just messed up the test because I thought an i32 was 8 bytes. It works fine as best I can tell.

@retoor I extracted it and wrote a nice test scaffold with random sized reads and writes:

https://git.lbfalvy.com/Orchid/...

There's a deadlock in it somewhere but I cannot for the life of me figure out where, this has me wondering if there's a rule about futures that I didn't learn right that's causing all these weird deadlocks in Orchid as well.

This is pretty solid but not nearly sufficient testing, I want to try some weird scenarios with flushing and closing as well once this one passes, and probably a benchmark for the use-case of a sequence of many small writes of varying sizes followed by a sequence of matching sized reads, because this is how Orchid and I imagine most real-world binary protocols use their channels.

to begin with, a non-thread-safe ringbuffer pretty much only makes sense if your architectural decisions are as weird as mine.

@retoor I'm not sure what kind of projects @BlindXfish is expecting us to upload, and this is purely a vibes-based assertion, but it seems very visual with all the cards and stuff so I'm guessing that a non-thread-safe async binary-safe ringbuffer library for Rust probably doesn't fit very well.

@whimsical It's actually moved, https://devplace.net now works and it's probably better if we link to that one (although I've no idea whether pagerank is still worth anything)

@Lensflare i mean you have to be a _little bit_ smart about it, but just blasting through all 1M combinations of 1000 coordinate triads with a loop is more than feasible. If you avoid expensive operations in the loop, it's indeed a couple of seconds in a compiled language.

@Lensflare I avoid bruteforce solutions in AoC even when they're possible because I think they're boring.

@Lensflare I realized day of that this would be bruteforceable in Rust and probably manageable with just 5x5x5 chunks in Python even, but I think this solution is fun so I'm sticking to it.

@TrayKnots you gotta submit part 1 to get part 2. The idea is that part 2 is usually a more complex form of the problem that you shouldn't plan for when you come up with the simpler strategy for part 1.

Still building an octree for day 8. If I accomplish nothing else for the rest of AoC but finish this, it will have been worthwhile.

AI generating a rant is weird and sad

but whatever really, it's not like most of us here are normal or happy

General purpose large scale competitions are stupid. Who's the best singer in Europe is a dumb question. Who can do the best aria of the night queen or the best death growl may be interesting questions, but a general purpose judge is ill equipped to judge either.

I mean, if people want to watch this crap then go right ahead, but let's all recognize that from the very principle it's purely a competition of tastes and not a competition of talent because talent doesn't compare across genres. No wonder then that it attracts party politics and flag-waving rather than meaningful conversations about social issues.