[Avamander]

The 802.11ah offerings right now are a mess though. Mostly proprietary and just generally very buggy. I don't know of a single chip that can actually be used with up-to-date Linux. Do you? Be it Morse Micro, Newracom, Taixin or any other, they all suck in some aspect.

Hostapd people also do not seem interested in bringing in any 802.11ah support. So it's crap in that aspect as well. Drivers all fake 802.11n or the chipset offers some garbage AT-command interface and does all of the networking.

On the other hand MeshCore and Meshtastic have similarly terrible codebases as far as I've seen. At least they're somewhat usable though.

Honestly no clue why these software stacks are all this dangerously written, unstable and haven't improved in years.

[KurSix]

Mesh networking is still mostly a playground for hobbyists and hacky, built-on-the-knee implementations. People love shipping a cool PoC, but as soon as the boring stuff starts - stabilization, drivers, edge cases - everyone bails to chase the next hype protocol. We’re left with mountains of half-baked C++ legacy that nobody dares to refactor because the whole house of cards would collapse iirc

[jauntywundrkind]

I see it much more as a chicken & egg problem. There's few places to even start to engage because so little hardware supports interesting modes well, particularly when coexisting with our normal operating modes.

Marvell shipped some 802.11s in an already quite obsolete wifi driver for xopc a long time ago, but generally theres just very very very little availability of drivers. The friefunk and Batman and other wireless meshes had to fight fight fight, pick very very selectively what hardware to run on, and they are far less than the ad-hoc ambient connectivity we see.

I'd hoped wifi-nan (neighbor aware networking) was going to open things up a lot. And hostapd allegedly has some support now! But last I looked it's very unclear how to use it. And it's much more a low bandwidth back channel than an actual rendezvous system for making good wiif-direct or other faster connections, I believe.

There's been so little hardware that offers anything interesting. So that makes me feel like the concern here, that folks keep chasing higher level protocols that look shiny, isn't quite apt. There's such a broad unavailability of starting places, for l2 link layer options.