Subvocal interfaces have been a staple of SF since probably the 1960s, but what this really reminds me of is the "ensemble coding" used by kids in the book Rainbows End.

Fun. Vaguely reminiscent of the Amiga HAM video mode.

The article is riddled with AI errors (for example, the company's name is Viosys, not Biosys).

Looks like the innovation here is the multi-wavelength LED, some kind of multiple-quantum-well thing maybe?

Yikes. We've changed the URL from https://www.ajupress.com/view/20261028171670308 (and banned that site). I replaced it with the article that Google returns for the title, which looks like it might be the original source, or at least closer to it.

Might want to consider not doing that.

See also the other comments here: https://news.ycombinator.com/item?id=50328023 V/F doesn't really exist in Korean and B equivalent in latin character/ English sound gets used instead.

And many Korean legitimate news sites use AI to translate their articles to English (and other languages).

Never heard of Ajupress, but they seem legitimate based on short research. More than this semi conductor today site I'd say. The news site being Korean, would think they're way more likely to be the original source, which then got picked up by the semi conductor site.

This is their original article, in Korean, if you prefer original sources, in their respective language: https://www.ajunews.com/view/20261028171303299

Ok we'll unban the domain. Thanks!

Fwiw korean doesnt have a ‘V’ sound/character at all, and the V/B switch is quite common on Korean words that come from English. For example, “video” in Korean gets written with ㅂ at the start, and sounds more like “bideo”. And the same character is used in the Korean/Hangul version of their written company name.

Yeah, pronounced in korean it definitely is more similar to Biosys than Viosys. But since they have written it as Viosys in all english documentation and websites it is very suspect that the article writer would misspell it like that. So probably AI translating a korean source or something, but AI also does these mistakes much less often these days.

They have probably chosen "V" for the English transliteration to avoid being confused with a bio-engineering company.

A web search for "exploit chain uaf" found me a few real-world and a few tutorial examples, though I don't know which are good.

At a high level, though: once you've gotten a new object allocated where the old one was, you now have two different views of the same bit of memory. You use valid operations on one view to put the other one into a corrupted state. For example: in a unixlike kernel, maybe one view has a field containing a function pointer (for method dispatch, essentially). You read that field using the other view, use it to deduce something about the target's memory layout, write a new value to the pointer, then use the field via the first view to cause execution to jump to a location you control.

Why even learn to speak? Why not give up sentience entirely? (There's an SF novel from the last few decades which had this as part of the setting: that intelligence is only useful at some stages of evolution, and most spacefaring species eventually lose it?)

What? How are you gonna communicate with anything without speaking? The LLM will keep showing you videos and you will draw it pictures of what you want?

That matches my experience as well. Some of the most disastrous bugs that got shipped were from 2-line "obviously correct and safe" changes.

Looks like Gitlab ships its own forked copy of the Ruby Onigmo regex engine.

Why? Is/was Ruby vulnerable?

It's not just obvious, it's evident.

Why it's important: It's the honest take.

Last summer. The CVEs are from 2025.

Its manufacturer's competitor isn't, though, and they're probably really interested in that marketing info.