There are so many unwarranted ideas here... I stopped reading half-way because of it.

Eg. JIT compilers are rare because they are difficult to write. JIT compilers come with tradeoffs, they aren't an absolute boon. JIT-compiled code isn't good for short-running scripts, for example. Even normally JIT-compiled languages try to special-case parts of the code that would not benefit from compilation and interpret them instead. Another huge downside of JIT-compiled code is debugging it. JIT compilation is better for languages that don't deal with memory allocation directly, and, in general, expose fewer low-level primitives to the language user. It would be a fool's errand to make a JIT-compiled C... I bet someone tried it, but, seriously...

Another one: a solution to slow search is building an index... Boy do I have bad news for you... Even a beginner DBA knows that an index can be a curse or a blessing. There's no one size fits all solution to this.

And the author continues to misrepresent similar problems only to make his solution make sense.