tynet-lichat/shirakumo

11:44:25 |3b| or maybe it is OK but has some odd slow path? not sure what its doing

11:45:26 |3b| ah, it modifies the blocks, so ub128 fails horribly there for consing

11:46:40 |3b| so yeah, nibbles and a macro to fix that "load N X-bit contiguous values" pattern

11:52:16 |3b| https://gist.github.com/3b/eab5e745a315d5bad0509f5da3ce9ae0 for example

11:53:12 |3b| 32bit path potentially needs to read 2 words and assemble them, but who cares about 32 bits :p

11:53:50 shinmera my android boards are crying, |3b|

11:53:57 shinmera the android boards that definitely will run Weiss are crying

12:03:29 |3b| probably could get better code gen if num-ebits / num-wbits were known at compile time for the cases where it is expected to be fixed

12:04:10 |3b| for example if they are fixed for a mode, and the function only does 1 mode

12:07:04 |3b| ah, num-wbits is a boolean, so just num-ebits?

12:11:26 glenneth So my intent for today is to get these changes correct and merged in to master. after that I think someone using it in anger could be helpful? Thoughts?

12:14:43 shinmera I can work on the benchmarking and integration testing

12:14:55 shinmera using it in anger idk tho, no clue when I'll be well enough again to get back to gamedev

12:16:14 glenneth or even a more realistic test env. This is not my area of expertise, but it is becoming one 😆

12:16:21 shinmera sure

12:16:31 shinmera I already offered to do that anyhow :)

12:17:43 glenneth Yeah. I need to learn to let go lol

12:23:52 shinmera meantime my poor brain is being overexerted trying to think about how to make stream communication work in the presence of multithreading without accidentally locking up one thread indefinitely while another thread is waiting for a specific query response

12:25:24 glenneth for ex-lichat?

12:25:31 shinmera no

12:25:33 shinmera for Commander

12:25:59 shinmera a thing like Swank but using a bunch of other libraries to do the portability bits rather than rolling everything into its own ad-hoc structure

12:26:18 shinmera a prerequisite for the Trial-IDE

12:27:08 |3b| more untested code: https://gist.github.com/3b/eab5e745a315d5bad0509f5da3ce9ae0

12:28:36 |3b| *|3b| forgot to replace that MEMBER call with a bitmask or something though*

12:29:06 |3b| even building the mask there might be an improvement

12:29:34 |3b| *|3b| has been sitting too long though, so afk indefinitely*

14:42:47 Colleen Good Morning

14:42:47 Fade Good Morning

15:05:42 glenneth Hey Fade o/

16:09:10 Colleen heya, glenneth

16:09:10 Fade heya, glenneth

16:16:53 glenneth Sorry, was afk. What's happening?

16:22:10 Hugo good afternoon

16:22:28 Hugo or like, evening I guess

16:26:34 shinmera in other things, I added a human.json according to https://codeberg.org/robida/human.json to shinmera.com

16:33:06 glenneth Nice. I like that.

16:33:09 glenneth Hey hugo

16:33:23 shinmera it's a neat idea, though like anything small web idk that it'll take off enough to be useful

16:39:46 Hugo I can't wrap my head around it but I love the idea

16:40:08 Hugo hey glenneth:

16:40:11 shinmera you publish a file on your webzone saying "these sites are sites I know and are written by a human"

16:40:38 shinmera and then the extension, when you go on a site with such a file, will see that. by linking together the references to other sites, you can get a web of trust

16:41:18 shinmera you yourself have to start the chain by saying "these are sites I trust" in the browser, so there's no need for a central authority or anything

16:41:58 Hugo I want to participate but I don't have a webzone yet

16:47:38 Hugo when I read through I felt like it's the kind of idea that could change the world

16:49:45 Hugo at the least, it's targeting something that the world is in dire need of right now

17:01:06 fungal i like the idea

17:01:40 fungal and i'm considering adding it on my webbed site (just need to finish my current work that is postponed for months)