[layer8]

Those aren’t real keys in the sense of having their own keycodes, instead they just map to some other key combination. The Copilot key will presumably be similar (?). These “new keys” are unfortunate, and a regular key that you can remap to whatever you want would be more useful.

[o11c]

I for one am highly annoyed at new keys generating a combination, since that means they can't be mapped to some more useful function.

... though I've just realized that with Wayland, detecting key events is quite unreliable since half of them are eaten by the compositor ... oddly, they aren't even reliably eaten ...

[k8svet]

This so flagrantly stupid. It's easier to make every KB manufacturer integrate special firmware to have one keypress register as another specific one, rather than plumbing it in as a defined new key that Windows registers a handler for???

[saltminer]

Not defending MS, but keyboards already have controllers and have since at least the original PC (probably much earlier than that but I'm not very familiar with old mainframe terminals). Considering how many SKUs already exist for different layouts and the like, this isn't that big of a burden - at most, this requires updated firmware and pad printing layouts, and firmware already changes a decent amount as controllers and membrane layouts are modified for new SKUs and cost-reduction. And it's not like they have to update their existing stock or retool their lines right now, the phase-in will likely take some time.

Also, unless I missed something, I don't think the article says one way or another how this will be implemented. It could be a new scancode or emulate a key combo.

[layer8]

That’s what MS did with their own keyboards. Maybe they are actually defining a new key code for the Copilot key? It’s stupid either way.