The difference probably is that GCC extensions have been stable for decades. Meanwhile Rust experimental features have breaking changes between versions. So a Rust version 6 months from now likely won't be able to compile the kernel we have today, but a GCC version in a decade will still work.
Not that you'd usually need this if you have IPv6 but might still be useful to bypass firewalls or forward access for IPv4 clients from your newer IPv6-only resources.
How do you feel about something like CBOR? In which stage would you say it's stuck in evolution compared to ASN.1 (since you said Protobuf is still TLV)?
CBOR and JSON are just encodings, not schema, though there are schemas for them. I've not looked at their schema languages but I doubt they support typed hole formalisms (though they could be added as it's just schema). And since CBOR and JSON are just encodings, they are stuck being what they are -- new encodings will have compatibility problems. For example, CBOR is mostly just like JSON but with a few new types, but then things like jq have to evolve too or else those new types are not really usable. Whereas ASN.1 has much more freedom to introduce new types and new encoding rules because ASN.1 is schema and just because you introduce a new type doesn't mean that existing code has to accept it since you will evolve _protocols_. But to be fair JSON is incredibly useful sans schema, while ASN.1 is really not useful at all if you want to avoid defining modules (schemas).
I was considering CBOR+CDDL heavily for a project a while so they're a tad intertwined in my head. I very much liked CBOR's capability of being able to define wholly new types and describe them neatly in CDDL. You could even add some basic value constraints (less than, greater equal, etc.). That seemed really powerful and lacking ASN.1 experience it sounds like a very lite JSON-like subset of that.
I recently wanted to do point-to-point Wi-Fi for transferring some data but apparently support for the ad-hoc IBSS mode wasn't available on my MT7925. Wi-Fi Aware is completely new to me and didn't come up while searching on the topic at all. I can't find anything about using it on Linux now either. Anybody have any references on its support?
There's a single kernel commit referencing Wi-Fi Aware from 2023 [0].
iw supposedly supports a few commands pertaining to it [1].
The WiFi Alliance has a habit of always have a marketing name and a different name in the spec, you'll a lot more references to it in places like WPA supplicant if you search for Neighbor Awareness Networking (NAN). Also here is the link to the spec https://www.wi-fi.org/system/files/Wi-Fi%20Aware%20Specifica...
Any WiFi operation besides STA is in general a crapshoot, especially if the card is not meant for use in an AP. WiFi hardware vendors can't be bothered to provide fully usable stacks for anything else (if even that).
For example Intel's broken Location Aware Regulatory completely breaks any use-cases where your device is not the STA (on anything besides 2.4GHz). Most cards also have no DFS support, meaning you'll be left with a microscopic usable segment. Then there's also the problem with incorrect regulatory information.
All of which in the end makes reliable high-speed point-to-point operation very annoying to achieve. Even if it'd be totally legal. Leaving you with a terribly slow link.
Adhoc was the coolest thing, I still miss it. One day in 2002-ish, I was showing a friend some photos on my laptop and noticed a crowd had gathered over my shoulder, and there simply wasn't enough room for everyone to get a good view.
"Fire up adhoc, set it to this ssid, vnc to this address"
Two minutes later, my photos are on five screens around the coffee shop and everyone can see.
Adhoc just worked, and that's more than I can say for a great many things before or since.
My windows laptop supports creating a wifi hotspot. It even allows sharing my upstream wifi internet connection over the hotspot, which I wasn't aware was a thing until recently (my Pixel 7 also supports this). I'm sure you could do the same thing with Linux with the right incantation. Not as cool as adhoc but it's also a paradigm people are very familiar with these days.
The explicitly includes Cloudflare as one of the big services they currently used and needed to excise from their life as part of this move. Promoting consolidation from many providers to one while also switching from a generic solution to a vendor locked-in one would probably be a downgrade in their book.
Side note: the site has an IPv4-mapped IPv6 address as its AAAA record. Surprisingly this actually works. Unfortunately there's no way to contact them over email to let them know that it's wrong/useless.
reply