#
✕ nick:MichaelReeves[m]
#asahi-dev · 2026-10-14 12:13
MichaelReeves[m]s/it/in
#asahi-dev · 2026-10-14 12:12
MichaelReeves[m]ehhh im not sure about that it looks like its pulling it a bunch of native deps, would that complicate packaging?
#asahi-dev · 2026-10-14 11:40
MichaelReeves[m]oh that sounds good, thank you! I think web sockets sound good then
#asahi-dev · 2026-10-14 11:38
MichaelReeves[m]how big a dep would a RPC library be? I was hesitant because I thought it would massive but maybe it's manageable
#asahi-dev · 2026-10-14 11:37
MichaelReeves[m]ah, true
#asahi-dev · 2026-10-14 11:36
MichaelReeves[m]REST is just the existing really simple ones in the plan GET /api/info, POST /api/install, GET /api/telemetry, POST /api/shutdown
#asahi-dev · 2026-10-14 11:36
MichaelReeves[m]yeah but then you still need to decide the message protocol and everything I think
#asahi-dev · 2026-10-14 11:34
MichaelReeves[m]ok after doing my research I think keeping everything REST and then using SSE for log/progress streaming would be best. WebSockets look cool but very complex, and I think we should KISS
#asahi-dev · 2026-10-14 11:24
MichaelReeves[m]cool, thanks!
#asahi-dev · 2026-10-14 11:20
MichaelReeves[m]its getting off topic but the layers of approval needed for chromium is actually pretty high... to add 1 CSS alias, I needed to get approval from Privacy, Security, Enterprise, Debuggability, Testing, Adoption and API Owners (all seperate teams at google LOL)
#asahi-dev · 2026-10-14 11:18
MichaelReeves[m]Alex0: well not really, cause the other implementors need to support it too. like Privacy Sandbox died despite shipping in Chromium
#asahi-dev · 2026-10-14 11:08
MichaelReeves[m]im totally clueless at backend dev so thank you for everyone for helping! never really been my thing (until now I guess)
#asahi-dev · 2026-10-14 11:08
MichaelReeves[m]ah yeah true, sounds good
#asahi-dev · 2026-10-14 11:05
MichaelReeves[m]oh googling that it actually doesnt seem too bad? whats the dragons?
#asahi-dev · 2026-10-14 11:01
MichaelReeves[m]for progress updates, seems like a good idea, I'll have to look into WS now
#asahi-dev · 2026-10-14 11:01
MichaelReeves[m]just for progress updates or the entire thing?
#asahi-dev · 2026-10-14 10:38
MichaelReeves[m]to avoid confusion
#asahi-dev · 2026-10-14 10:37
MichaelReeves[m]so people can retry if there's a failure
#asahi-dev · 2026-10-14 10:37
MichaelReeves[m]wait actually I remembered why I put UUIDs
#asahi-dev · 2026-10-14 10:35
MichaelReeves[m]yeah I was a bit worried that random websites could (in theory) send stuff to localhost and mess with computer so put that in
#asahi-dev · 2026-10-14 10:35
MichaelReeves[m]oh good idea I'll take that out
#asahi-dev · 2026-10-14 10:31
MichaelReeves[m]I don't do HTTP/APIs much so I have missed something easy to make it simpler
#asahi-dev · 2026-10-14 10:30
MichaelReeves[m]I tried to keep it as simple as I could think, what were you picturing?
#asahi-dev · 2026-10-14 10:23
MichaelReeves[m]ack
#asahi-dev · 2026-10-14 10:22
MichaelReeves[m]thoughts / feedback ?
#asahi-dev · 2026-10-14 10:22
MichaelReeves[m]the web app currently follows that, but it can be easily changed
#asahi-dev · 2026-10-14 10:22
MichaelReeves[m]draft API spec: https://gist.github.com/IntegralPilot/47ca5f9ae4a877bfb2a36d530d4e90de
#asahi-dev · 2026-10-14 09:06
MichaelReeves[m]yay
#asahi-dev · 2026-10-14 08:39
MichaelReeves[m]Once I’ve seen the reverse, 500 ISE { “error_code”: null, “message”: “” } that means it works
#asahi-dev · 2026-10-14 08:37
MichaelReeves[m]ack. I’ll write up the API I currently made it call so we can see if design is good or need improvement
#asahi-dev · 2026-10-14 08:33
MichaelReeves[m]also I don’t think we need to worry abt semantic codes, just do 200 or 500 for everything
#asahi-dev · 2026-10-14 08:18
MichaelReeves[m]Like if the i18n string is “[volume] is not an APFS volume” and then volume is subst in as a var
#asahi-dev · 2026-10-14 08:17
MichaelReeves[m]And then front end substs the vars into the sentence for each localisation
#asahi-dev · 2026-10-14 08:16
MichaelReeves[m]Could be smth like { “msg”: “whatever”, “vars”: [“name”, 5, …] }
#asahi-dev · 2026-10-14 08:16
MichaelReeves[m]ah
#asahi-dev · 2026-10-14 08:14
MichaelReeves[m]Will?
#asahi-dev · 2026-10-14 08:13
MichaelReeves[m]And then front end can look it up in its localisations
#asahi-dev · 2026-10-14 08:13
MichaelReeves[m]return string name then?
#asahi-dev · 2026-10-14 08:12
MichaelReeves[m]I think the idea is that the front end is served by the python process, so if the python throws an exception then the front end will die too
#asahi-dev · 2026-10-14 08:11
MichaelReeves[m]Return 500 with error in body on API and I’ll make it show in front end
#asahi-dev · 2026-10-14 07:27
MichaelReeves[m]uploaded an image: (307KiB) < https://matrix.org/oftc/media/v1/media/download/AT6E8whhJaJUeQv5eeBIa9KtGuKpDBwVBhmV-QNqzD9_JmvCfjseQywphH1JCeh7qEdTeZTMUzkPsmKFKiU6rnxCehgWhCiwAG1hdHJpeC5vcmcvR0pWeGt6TGxGVmN0THdibllqbVJRYWZw >
#asahi-dev · 2026-10-14 07:27
MichaelReeves[m]Because to non dyslexic people it’s a rly ugly font (imo)
#asahi-dev · 2026-10-14 07:26
MichaelReeves[m]it’s already offered I was joking
#asahi-dev · 2026-10-14 07:24
MichaelReeves[m]no, we should definitely use OpenDyslexic as the default font with no option to turn off
#asahi-dev · 2026-10-14 07:21
MichaelReeves[m]yeah we decided that before
#asahi-dev · 2026-10-14 05:59
MichaelReeves[m]that's the current default
#asahi-dev · 2026-10-14 05:59
MichaelReeves[m]which becomes SF Pro on Macs
#asahi-dev · 2026-10-14 05:59
MichaelReeves[m]well we dont explicitly bundle SF pro, its just --font: system-ui, sans-serif;
#asahi-dev · 2026-10-14 05:58
MichaelReeves[m]or I mean SF Pro actually isn't that bad and makes it feel more natural for macOS so idk. I'm really bad at subjective stuff like this cause I keep going back and forth
#asahi-dev · 2026-10-14 05:57
MichaelReeves[m]maybe for the default font we should find a good-looking seperate semi-serif font?

plus ›