wryun

Yes, we're trying to upstream this (and iw support) currently. This was blocked on having S1G and a driver in the kernel.

I'm not the expert in this area, but the current state of play AFAIK in terms of working chips is:

- Morse Micro - Newracom - Taixin are mostly not compliant (but cheap) - Amazon have a proprietary not quite compatible HaLow in their Blink cameras

There's varying levels of interest in WFA compliance/interop. Morse and Newracom are the most interoperable, but even then there are gotchas.

The Morse ones, are, of course, the best ;)

Aside from upstreaming, one of the other things we're chasing is a better regulatory environment. In particular, the EU S1G bands are very tightly constrained.

I'm excited because of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/lin..., which is the first kernel driver to use the new Wi-Fi HaLow (s1g/900MHz Wi-Fi) support.

(Disclosure: I work for Morse)

OpenWrt devs had this epiphany many years ago:

https://openwrt.org/docs/techref/ubus https://openwrt.org/docs/techref/ubus#what_s_the_difference_...

(no endorsement implied; in particular, ubus has not much of a security model, though OpenWrt has an excuse for that)

Which doesn't mean that it's easy to use an ARM device in the way I'd want to (i.e. as a trouble-free laptop or desktop with complete upstream kernel support).

I guess you must be doing trickier things than I ever have. I've found docker's emulation via qemu pretty reliable, and I'd be pretty surprised if there was a corner case that wouldn't show on it but would show on a native system.

Surprising. Cross compilation too annoying to set up? No CI pipelines for things you're actually deploying?

(I'm keen about ARM and RISC-V systems, but I can never actually justify them given the spotty Linux situation and no actual use case)

The tree mesh thing you're thinking of is actually just EasyMesh (it's a standard!). We're using prplmesh.

But yes, 11s Mesh also works. Let us know on the forum (https://community.morsemicro.com/) or via github (https://github.com/MorseMicro/) if you're having issues. Err, I work for Morse in case that wasn't clear.

But what if you're using mesh or running an AP? Won't the software out of the box allow you to set the regdomain to AU?

If this is like the RPI4 with a single global model I'm wondering how it passes FCC regs, in that presumably the Wi-Fi is flexible enough to transmit on illegal US channels (i.e. Ch 12-14 on 2.4).

Am I missing something?

They do have: https://github.com/jolibrain/colette/blob/main/pyproject.tom...

I agree it's better to have the full licence at top level, but is there a legal reason why this would be inadequate?

Search for WiFi HaLow.

(I work for Morse Micro; we can do 30mbps udp)

On most openwrt devices you can compare /rom/etc/config to /etc/config (though this will get you a bunch of automatic changes as well).

Just use a Surface Go? Tablet linux has been working well for years...

https://en.m.wikipedia.org/wiki/Masters_of_Rome is an enjoyable fictionalised account.

But also:

https://www.theguardian.com/business/2023/sep/06/deezer-and-...

My vote is for x32.

(you didn't know? https://wiki.debian.org/X32Port)

I like it, but it feels funny to have this discussion without acknowledging this is really similar to pjax/turbolinks. Ok, now I'm feeling better; carry on!

Or nftables...

The sweet spot in my opinion is to use an ordinary Linux box as the router and just wire up whatever wireless router(s) you have lying around for the APs (give them a static address in the appropriate subnet and you're done). No need to run OpenWRT on on the AP, since it's just mindlessly pushing packets around on the internal network.

For bonus points, block it from accessing the internet itself.

Figure out why docker and snap are broken?

---

You could use https://wiki.debian.org/nspawn unless it's really old.

I believe you can somewhat mitigate these concerns by reading and removing the secrets on startup, ideally whitelisting the entire env.

Yeah, good thought, though I worry that would add another layer of planning and make it seem more like work. Probably should do Mik.

"If he had been a great and wise philosopher, like the writer of this book, he would now have comprehended that Work consists of whatever a body is obliged to do, and that Play consists of whatever a body is not obliged to do. And this would help him to understand why constructing artificial flowers or performing on a tread-mill is work, while rolling ten-pins or climbing Mont Blanc is only amusement." (Mark Twain)

I have a perceptual problem where my mind quite quickly labels things as work as soon as I start doing a bunch of planning. Once something is 'work' that I'm not being paid for, I then don't feel like starting it. I rarely play computer games any more largely because I have too many _plans_ about what computer games I should play.

This doesn't mean I wouldn't have more fun doing the programming than sitting around browsing the internet.

Thanks!

JOSS is definitely one of those 'heh, this sounds fun' things, despite being pretty useless. I too am pretty obsessed with the colour changing; I'd love to implement it so it seemed like a typewriter terminal as well (e.g. slow speed, optional sound effects).

I started implementing it in Pony originally to make it truly shared, but then decided it made more sense to just run the code in the frontend and the shared bit would just be common storage (i.e. so you could easily load other people's programs, as you would on the original systems). Of course, haven't actually done any of this yet...

Unfortunately, I seem to mostly value lying on the couch and browsing the internet.

Conversely, if someone pays me to sit at a desk, I'm quite motivated to do whatever they tell me!

I imagine that from their perspective, they're trying to decrease overall fraud, not cater to individual preferences. They probably see (or imagine they see) more fraud attempts coming via email than SMS due to insecure email accounts; as you say, "I appreciate not all Ally's customers are technically capable of setting up their email in this way...".