It started as a language targeted for numerical computation.

But as usual, many developers resist to "use the best tool for the job" instead of "a language to rule them all", and Julia is now starting to have libs for everything.

Not that is bad, quite the contrary. The language is quite nice and already achieves C like speedups for many of its features.

So I can easily see it getting a place in the mainstream.

it's surprisingly good. it feels much more like a general language than matlab does. i have been using it privately as a "faster python" and have been very happy with it (but for work that is arguably still mathematical in nature - generating procedural art and looking at block ciphers).

in fact, i didn't realise it was intended to be a matlab replacement at first. i just looked at the specs and thought it looked interesting (i still haven't used the interactive dohickey (ijulia)). if you come to it from that pov, the only thing that seems odd is the 1-based indexing (and if you're old enough to have used fortran then even that doesn't feel so strange).

I think its fair to say that it shines for numerical computation, but you can perform other tasks with Julia with ease.

For example, the idea of writing some number-crunching as a webservice with MATLAB or Octave is not an option, but with Julia is totally possible (and has been done more than once already) - and it builds on strong foundations (e.g. libuv). Heres an example (disclaimer: my own site) http://iaindunning.com/2013/sudoku-as-a-service.html