That's true. Uncertainty is everywhere.Think of it in trading terms: Why shouldn't you trade 1 second bars? It's the equivalent to C++.
I learnt C++ myself during my teens. After accumulating computing knowledge for 30+ years, my first question would be what do you need it for? Myself, I use whatever tool makes my job easiest. It both depend on myself, my own knowledge, but also on availability and costs of various tools.
Do you absolutely need access to raw speed and low-level access, ie. do you have specific requirements for such? If not, it might pay off to investigate more options earlier than later.
I try to avoid imperative programming, but ruby has served me well when I need to automate and the language has a clean nice syntax, although the lack of static types can be harsh to newbies. I hear good things about Python too. Don't underestimate the power of prototyping, and then migrating to something else when the needs are clearer. Although, it's hard to realize the best course of action when you have no idea - kind of like trading itself. RR can be used in many aspects of life and learning.
We don't know where the way leads,
Until we've done, walked the walk.
I learnt about "RR" with taleb.
I would use Mathematica for prototyping.
Knew that productivity is low. Tradeoffs.
But I also use X_Trader. No Java there.
Only C++, C#, VB ... .Net languages.
I don't really need C++ as I could wrap Apis
Into easier language I guess. Such as Python.
But C++ looks sexy. Low-level and fast. Even if time consuming.
Ok if it's the only trade off ... I've got time in plenty.
I am young and at home, free from anything.
Don't think it's gonna be a big mistake.
If I put time and effort toward it.
Don't know if it's going to help.
Neither ... Let's see.
I give it a try. Then if it leads nowhere. I'll drop it.
But if I drop it. I'd consider having built a good foundation.
For everything related with programing, according I finish the book.
It might have been the case in mid-90's, but a lot of things have changed since then.