Will it, though? I feel like we're in the part of history right now where - when it comes to computers anyway - we're going with what's cool rather than what works. I'd say in general touchscreens have continued to be an utter failure as a general-purpose input device, and that's why we only see it on toy devices. Anything where what you type matters still uses some kind of mechanical feedback.
Maybe with haptic feedback that'll change, but for now any attempt to move mechanical keys to touch-screens are, for me, a net negative just because of the basic facts of reality.
Look at how volume and brightness control is done on a Macbook. They took function keys and repurposed them into crude up and down buttons. To get fine control you have to hold down the Shift and Alt keys. If you want to actually use the "function" part of the function key, you have to hold down the Fn button.
A touch control there could present a more intuitive interface--tap the control you want, then use a touch slider to adjust with great precision.
Another idea I saw in a blog post was to put a scrubber control there during video editing. Currently the trackpad has to serve double duty controlling the features of the application, and controlling the scrubbing of the content.
The function keys are a UI mess. Each serves double or triple duty, and the duties sometimes change with OS releases. For example, the Dashboard key on my Macbook Pro no longer activates the Dashboard.
This is just like the situation on smart phones when they all shipped with physical keyboards. Each tiny button on my Blackberry had at least 2 different functions. As Steve Jobs said, "we solved this problem with a bitmapped screen and a pointing device."
I call it a toy, you express indignation and then use children as an example.
The way children use toys is not the way adults use tools. Touchscreens have proven great for "consumption" devices like phones, and have proven pretty terrible for most everything else.
I'm not being too hard on them I should say, but anyone who thinks traditional input devices will be replaced in general purpose computing with touchscreens ignores decades of history and basic physical reality.
How does your hypothesis square with the physical reality that a declining share of users use non smartphones for ANY type of computing activity, and sales of touchscreen only devices exceed sales of keyboard controlled devices by an order of magnitude (with that gap growing every year)?
Maybe with haptic feedback that'll change, but for now any attempt to move mechanical keys to touch-screens are, for me, a net negative just because of the basic facts of reality.