Or head tracking.[1] Aphysical rotation exaggeration avoids trading eye for neck stress.
> I do feel a little weird wearing these in public, but not that weird.
Non-weird can be an expensive constraint, fruitful to relax if going beyond a minimalism setup. A baseball hat can barnacle quite a bit before people find it remarkable... at least around Boston. For instance, for head tracking, an Intel RealSense, or a hat fisheye camera and tennis ball on the table, or an optical marker on a hat chopstick, can be simpler, easier, lower power, and less expensive, than invoking "and it has to look non-weird". With current tech, that's almost as challenging as "and it has to be a product".
> as these AR glasses continue to improve and Linux continues to be flexible and awesome.
I suggest Nreal (now Xreal) made a bad call here. They developed internally on Ubuntu, but chose to shut out linux. (Caveat: I've not followed in a few years, maybe that's changed.) Unicorn dreams and race to mass market - maybe the right call if everyone started watching media on phones with glasses. But it could have been an inexpensive risk mitigation, and a worthwhile investment once market fit was clearly a long haul. Is there some doc which lays out alternatives for a company who thinks they have a crown-jewel binary blob, to allow the community to wrap it for linux consumption, with minimal "we just throw a blob over the wall - we don't support linuxes" cost? It's been a lot of years that something like TFA has been possible, during which a lot of developers could have been exploring for viable market niches. Instead of... not.