The Nineteenth Byte

4:13 PM

@dzaima And with infinitely tall programs, it's impossible to initialize at all...

Oh wait, that's actually an issue.

They'll never run in a finite amount of time. They'd have to be infinitely wide.

@RedwolfPrograms you just have an infinite drop for a zero in one place getting branched where needed, and since everything travels down at max 1 unit per tick, it's guaranteed to be there before anything else (assuming that "else" is delayed enough, the delay being height-independent).

halting may be an issue, but that doesn't really matter (i.e. you could make outputting -1 halt)

also, looking at your source, what is the Math.random for?

4:36 PM

But in an infinitely tall program, the packet could never reach the actual logic in a finite amount of time. I guess that would count as an implementation detail, though, maybe. The random is for when a packet is pushed into either side of a storage unit at the same time. It technically doesn't matter though, since it's impossible to read from it in that case

@RedwolfPrograms my point is the zero packet could reach whereever it's needed before anything else due to the speed of transfer being unbeatable. The packet won't reach everywhere in a finite time, but it will reach everywhere it could possibly affect anything

e.g. if you're 100 steps in, you don't care that there isn't an initialization zero 1000 units down, because there won't be any other packets there either

Makes sense, I guess the -1 feature isn't necessary

it'll definitely allow for much more performant infinitely tall programs though (do you have plans for making an interpreter allowing for those?)

I don't know, maybe eventually. I guess storage is different enough from other units that it needs the -1 feature

5:31 PM

cmpc (popularity): think of a cool 256-bit (or 4*64-bit) nothing-up-my-sleeve number. ideally something playful so hackers like it, original (not: the leading bits of pi or e), and simple enough so any programmer can get it from a single-sentence explanation

Kritixi Lithos

@ngn chaitin's constant of some not-too-golfy language

is there a code-golf challenge that asks to write a program for which its halting status is not known, akin to codegolf.stackexchange.com/q/111092/41805?

@KritixiLithos cmpc amendment: the nothing-up-my-sleeve number must be computable :)

Kritixi Lithos

if you choose a language with the right conciseness/verbosity, you can get a bunch of people manually go through each possible ≤64b programs and hope no unsolved problems reside within

@KritixiLithos i need 256 bits of entropy

5:52 PM

@ngn y'know, it's hard to "make" nothing-up-my-sleeve numbers :p

@dzaima or maybe easy to make but hard to test what people have up their sleeves :)

wizzwizz4

@ngn The first non-negative integer n satisfying n % 2*3*5*7*11*13*17*19 == n**2 - 2**n + 3**n == 0.

@wizzwizz4 0?

wizzwizz4

@ngn There's no way I could've got that from a secret backdoor that nobody else has.

Therefore, it must be safe to use.

lol :)

6:11 PM

@ngn the parity of the first 256 unique numbers of the continued fraction of pi

@dzaima that's a good one