I personally have used
Julia for a good number of professional projects, and while, as Dirk mentioned, this is purely conjecture, I can give some insights on where Julia really stands out. The question of whether or not these reasons will prove enough to have
Julia succeed as a language is anyone's guess.
- Distributed Systems: Julia is the easiest language I've ever dealt with in terms of building distributed systems. This is becoming more and more relevant in computing, and will potentially become a deciding factor, but the question of whether or not
Julia'a relative ease decides this is up for debate
- JIT Performance: Julia's JIT compiler is extremely fast, and while there is a lot of debate as to how accurate these benchmark numbers are, the Julia Website shows a series of relevant benchmarks
- Community: This is an area where
Julia just isn't quite there. The community that is there is generally supportive, but not quite as knowledgable as the
python communities, which is a definite minus.
- Extensibility: This is another place where
Julia is currently lacking, there is a large disconnect between the implies code patterns that
Julia steers you toward and what it can actually support. The type system is currently overly bulky and difficult to use effectively.
Again, can't say what this means for the future, but these are just a couple of relevant points when it comes to evaluating
Julia in my opinion.