My hope would be that the interfaces of the future would be more accessible to the disabled, not less.
In any case, why do you think that using gripping as a UI action implies that you'd need a strong grip to use it?
I don't think it's fair to make non-disabled individuals use a less productive interface, just as it's not fair to prevent disabled individuals from using the interface of their choice.
I grip my mouse to use its side buttons to navigate forward/backward in a web browser. Human hands are generally good at holding things, and flat touchscreens/trackpads aren't very holdable. Power tools, steering wheels, gear shifts, volume knobs -- all examples of things that are gripped.
I disagree completely. The line of reasoning is "what actions can we perform that have the highest bandwidth, and lowest cognitive load?" And various forms of manual manipulation, like gripping, are an excellent avenue of exploration when answering that question.
Actually, I think this extra thought could help everyone by treating disabilities as just special cases of the general idea that different people prefer different input/output devices and having apps (mostly) stop processing low-level input (keystrokes, pixel hit testing) and switch to handling more abstract "primitives".
The OS would have a set of more abstract primitives: enter text, zoom in, zoom out, rotate, 3d move, 3d rotate, select item, select 2d region, select 3d region, deselect, grab, cancel, undo, etc. You, the user, would need a set of I/O devices that provided one or more implementations of each primitive in the checklist.
One user would send an "undo" to an app with a click of the tongue, another with a hand gesture, another by pressing the ESC key, but the app just gets an "undo" event from the OS.
For 2d region select, for example, one would use a joystick, another uses voice, another uses eye movement. The app doesn't process pixel hit testing to do its own region select, it lets the OS tell it what region of the 2d surface was selected. The app makers don't make the app "accessible" themselves, they just make it respond to the checklist of higher-level primitives.
You could provide at least the core functionality of most apps by this means after which an app could possibly add some lower-level processing in the form of "shortcuts".
When I'm forced to use a watch interface away from my keyboard, I am in a sense "disabled". When driving in my car with eyes on the road, I am "disabled". I'm not a gamer, so relative to a gamer with mad joystick skillz, I am "disabled". Well, fine, just change the approach to software interaction to a more abstract type of interaction that treats us all as being variously abled.
Generally if you want to design for the disabled, you probably are designing for the disabled, rather than using them as fodder for an argument.
No single interface will ever suit all potential users, so any accessible design has to keep this in mind.
Honestly, common sense says the guy is talking about interfaces for the general population, not the disabled. If most people can grip then make interfaces that grip. Don't force everyone to read braille just because a few people are blind.
For the cats.
(I like cats. The problem then was that I had way too much stuff in various cabinets and the cat liked to make a grand tour through them, throwing a lot of stuff to the ground in the proces...)
Taking advantage of major human ability != making it an exclusive requirement to use said interface.
Also, suggestion != claim.
What edge cases? What design? What are you even talking about?
A person merely said "Hey guys, have you thought of this". GP got on their case as if they were against the disabled or something.