@brandon oh wow yeah this is awesome, ty
@ellyxir Last time I did something like this I just read their source code in devtools and injected a script in the console to simulate clicks with JavaScript :P
Its not just some purity thing. It's much more about reducing cognitive load everywhere I can on the web. When I see underlined text (ideally some shade of blue) I know I can click that to show a different thing. All the behavior is something expected.
This goes for buttons & forms. Being clever and doing things differently than every other website just makes people have to think about what they're interacting with, even if only for a split second. That's not innovation, it's unnecessary friction.
Great article spelling out the disadvantages to heavy React based websites. https://infrequently.org/2025/06/conferences-clarity-and-smokescreens/
@trey @cwebber That makes sense, I usually set my display scale factor to 2.0 to compensate but some UIs just can't handle it sadly. The aliasing sounds rough. SimulaXR did custom stuff to wayland rendering to make text more clear but that hasb't made it out of there and sadly the xreals have no chance of doing the dame at the X1 chip level.
@lvk I think llama.cpp does this with this save/load demo actually: https://github.com/ggml-org/llama.cpp/blob/master/examples/save-load-state/save-load-state.cpp
@lvk ooo yeah. I wonder how hard it'd be. Was most of the state just the context and kv cache? Sounds like not a lot. For some reason I though Ollama had some sort of "continue from the last call" feature already but I might be hallucinatinf
@jfred I wrote a bespoke parser and AST transformer thing for SAS (statistical analysis lang used in pharma). Sadly all proprietary. 😅
I wanna get into it for more langs though. I find a lot of parser ecosystems don't put enough effort into making it easy to serialize back to text after parsing + transforming 😅
@jfred I treat all whitespace as nodes in my ASTs and just gloss it over when I stitch stuff into the tree but I use them to serialize back into text. With whatever formatting is there
Anyone know of tools kind of like #languageServerProtocol but instead of keeping the AST in memory they do streaming parse / search on the fly? I'm not a huge fan of masssive memory use and it feels like we're leaving performance on the table by parsing entire files/folders instead of just enough to get to what you want.
@reflex @jfred @cwebber Yeah! They actually have an extra level of dimming rather than just on/off.
I mostly use my vitures for coding when on the go and for watching stuff via the neckband while I'm doing chores.
The xreal beam pro seems neat as a companion device but I'll probably just use my phone instead if I end up getting their glasses.
Occult Enby that's making local-first software with peer to peer protocols, mesh networks, and the web.
Exploring what a local-first cyberspace might look like in my spare time.