I learned about an interesting corner case last week: devices that use USB-C physical connectors but have USB 2.0 PHY. You can get away with a much longer cable at that lower Baud rate, but that cable likely won't work for most USB-C devices.

Some manufacturers have even introduced charging-only cables which don't even have the USB 2.0 wires. This isn't allowed by the specs, but that doesn't seem to be stopping anyone.

I think it is allowed. The PD spec mentions requirements for charging-only cables, these can actually be certified, if I understand correctly. I also think it's not necessarily a bad thing since it minimizes the risk of 20 volts short circuiting into D+/D- lines of a sufficiently abused cable.

Most USB-C devices and cables are USB2. Most USB-C phones are USB2, and there are a lot more phones than there are laptops with USB3 USB-C.

It is probably better to think of USB3 as extension that requires special, shorter cables. I like to think of them as charging cables and data cables.

I probably have two USB3 USB-C cables for MacBooks (and I found one that is 6ft long) and dozen USB2 USB-C cables for charging phones and MacBooks.

I recall hearing Alex Lindsay talk about only ever buying Thunderbolt cables now because it’s so uncertain of the USB3 cable you’re buying will actually do everything you want it to do.

True but many times helpful since that is the only way to get some relative cheap usb-c cables rated - and working -for 100W charging

I’ve been messing around with implementing a USB C port on a board I’ve been working on. The chip I’m talking to is only USB 2.0 so there’s a whole pile of things to figure out related to getting any given host to talk right to my thing. It’s cool, but complex.

A lot of Appple's iOS charging cables are USB C to USB C now that all/many iPads are USB C