This is what I am trying to explain. Nobody grows up surrounded with 1000's of extensions
It doesn't matter if we have 5000 TLDs tomorrow, or just under 300 like today. Because only a few will really have any relevance.
Ask your friends or colleagues to quote all the extensions they are familiar with, usually you will be lucky if they can quote a dozen extensions (including ccTLDs of neighboring countries).
People already know that a lot of TLDs exist, but they don't which ones. Tomorrow will be no different.
Today .com has more than
110 million regs. I think it will continue to grow, but at a slower pace. I don't believe it's going to shrink, for that to happen domain holders would have to ditch their .com and flock
en masse to other extensions. I don't see this happen, as there is no clear incentive.
Even in some remote future it is extremely unlikely that another extension, whatever it is, will attain 110M regs or more. Therefore I conclude that .com is going to remain the biggest TLD.
Likewise .berlin or .bremen will never have the volume of .de (15+ millions regs).
These are niche TLDs.
The TLDs that you mentioned as examples are not all purpose TLDs, they are niche TLDs too.
Unless I run a restaurant I have no use for .menu. .consulting is only good for consulting companies etc.
I think it's unnecessary fragmentation and these extensions restrict your scope.
The strength of .com is that it is very generic in nature and be used for anything. As for the ccTLDs they define markets with clear boundaries and local end users can
identify with them.