I don't know.
On the one hand, I think you're right, learning everything wastes your time, if you want to do something, find the tools that make doing it simple for you and be done with it.
On the other hand, there are many people out there who do things the old way and don't want to change it, even if it has considerable down-sides doing so.
"We don't need the Cloud!"
"Web UIs are too slow, we need to do it in C++!"
"Monoliths are much easier to maintain!"
"Mainframes are more efficient!"
I agree, but the comments here are generally pretty reasonable. There are problems on both ends of the spectrum. Landing in a sensible middle area is pretty important.
It's really about context more than anything. Monoliths are easier to maintain when they encapsulate less complexity than is worth the microservice effort overhead. Asynchronous web UIs can be unacceptably slow for some purposes. Breaking out a planet-scale distributed datastore when you have a few hundred megabytes of data tops is stupid. Framing an issue as a case of extreme, flawed, and diametrically opposed perspectives dueling around a sensible center tends to be overly simplistic.
Yes, and it's not easy.
I for one would say serverless is the future and blockchains are a fad.
But maybe it's the other way around?
I think these things are becoming important, but we are in fairly painful early days in both.
I don’t think there’s anything wrong with hanging back and catching on once some of the dust has settled.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.