Josh Emerson was talking about interfaces on the web at Dot York, these are my notes from his talk.
The device is the lens thorough which we view the web. We now have a spectrum from watches to TVs: small screen that's close to you to a large screen that's a long way away from you.
We've also got a plethora of inputs: touch, mouse, keyboard, sensors...
As we get more diverse we need a common language for interfaces that allows users to switch between them without relearning. Google have the the "Paper" design language. Apple have different interfaces tailored for different devices. Windows adapts to the input method to change the interface.
We need affordances to provide indications otherwise we are expecting everyone to think like we do.
With IoT the interface may be another device.
With Cortana developers can hook into the API to provide voice controls. The way the web works allows us to provide a voice interface: screen readers. Cortana is now built into the next Internet Explorer.
In the future we may have a goggles and gloves* interface: we might not have to choose whether to look at the screen or the surroundigs.
[Goggles and gloves is an old Cyberpunk term for a net interface that can be used by those without a neural jack. These days the gloves are virtual from things like Kinnect and Leap.]
[Note to self: we can't just leave people behind because they don't adopt a new technology - we have a social responsibility to provide for people to use our services on their terms, not enforcing our terms as the barrier to entry.]
Memes: Ideas are competing for survival.
Native form elemets are the most responsive and progressively enhanced UI elements - the appearance is defined by the device and the device can optimise it better than the web developer ever could.
CSS seperates out content and presentations.
Future friendly manifesto.
Browsers [user agents] need to think about how to best present web sites. It doesn't have to be the same.
Dao of web design.