One day we'll be the old men yelling at internet clouds while sitting behind an old PC or some kind of open source terminal running an open source OS and weeping for the past.
This is also as far as I know still true for a lot of other productivity-type uses of computers -- even though mobile and tablets are also used in some of this space, they often seem more auxiliary. Most of the serious professional tools for, say, business (eg spreadsheets, accounting, inventory, etc.) or even artistic uses (music, photo editing, etc.) also are primarily keyboard / mouse oriented. Tablet oriented tools definitely exist (especially in the art space), but a lot of times its not a primary use case, except perhaps in communication tools.
Mobile space is a walled garden, designed more simply for limited real estate and touch, with more difficulty to poke around (there is no real equivalent in the mobile world to open a web site, hit F12, and poke around). That's fine -- PCs / laptops haven't disappeared, and if a kid showed an interest in programming, you'd get them one. :)
Now, the original poster may have been referring to the new world of social. I did more enjoy the long form of social (eg blogs) that popped up in the 00s. Versus the current social world, dominated by only a few firms that are highly motivated to manipulate your feed, tweak your emotions, and optimize the dispersion of ads (and other low-quality tabloid junk). It's my "old man yells at cloud" pet peeve, I suppose. Not exactly a mobile device problem, though.
How many programmers and developers are there vs consumers? The traditional keyboard and mouse interface is dying in consumer space. Sure it will live on in business and content creation, but the majority of development going forward is pushing people onto mobile reinforcing it as the standard for interfacing to a computer.
But here in 2018, we really ought to all have a pretty clear idea of why touch screens are in no danger of displacing mice and keyboards 100%. It's pretty obviously not going to happen, and not particularly mysterious why.
In other news, phone-based gaming is also obviously not going to kill dedicated consoles anytime soon.
In other, other news, see also why touch screens need not fear being entirely replaced by voice-based computer usage. In much the same way that touch screens didn't, can't, and won't completely replace other higher-bandwidth ways of interacting with computers, voice has those same problems w.r.t. touchscreens even more so. You can get a nice boost on the bandwidth in the computer input direction, but you get annihilated on the bandwidth coming back out compared to a screen. There is no other computer IO mechanism that so thoroughly lacks bandwidth back to the human as a voice interface, short of a plain flashing light on a physical console. (Which, actually, if the human knows Morse code or something, could damn near keep up with the voice interface on the bandwidth front. That's how bad voice interfaces are on the output bandwidth.)
Productivity's less niche, but it's becoming more and more realistic to imagine someone using a tablet with attached keyboard as their sole computing device (besides the omnipresent smart phone, I mean).
I'd say that the input method doesn't matter so much, although it's interesting to talk about how the commonly used devices are shifting. I'm more worried about native software development sliding into impossibility and content-creation becoming an afterthought.
We'll be the last guerillas in the long-lost War on General Purpose Computing.
You can have lock-out without the TPM anyway - like the pre-Fingerprint reader iPhones.