4 ms·

Python is my usual go-to language for technical computing, bit Julia is a lot of fun! I'd be curious to hear more from folks who have made use of it in "product

by thearn4 12y ago

Python is my usual go-to language for technical computing, bit Julia is a lot of fun! I'd be curious to hear more from folks who have made use of it in "production" (sort of a tricky label in scientific programmiong) code.

oostevo 12y ago
For what it's worth, I tried and failed to convert my team of scientists/programmers (heavy on the science; not so much on the programming) to Julia from MATLAB/IDL/C++ for a project we were kicking off ca. 1 year ago. It's a really, really nice language that I find more consistent and easier to grok than MATLAB, but the deciding factor was the lack of the equivalent of MATLB toolboxes with all of their pre-rolled code that made our development as fast as it was. Edit: I should have mentioned: the Julia add-on packages are multiplying quickly, and are starting to look mature! http://pkg.julialang.org/ http://pkg.julialang.org/ I'm looking forward to them overtaking Mathworks on this front too.
- jordigh 12y ago
  Octave dev here, which packages are you looking for? I'm curious if it's as sophisticated as Simulink and Mupad, which are completely different programs that happen to be associated to Matlab, or if it's just the image package that we have almost all in Octave already. Where's the demand? Where is your demand? :-)
  - oostevo 12y ago
    This was at an organization that I don't work for anymore, and my recollection is a bit fuzzy. I think it was Computer Vision, Signal Processing, Wavelet, and Image Processing. There were also some convoluted, non-technical reasons that we went with MATLAB.
    - jordigh 12y ago
      Could you contact me at jordigh@octave.org for some follow-up questions?
idunning 12y ago
There was a talk at JuliaCon about this exact topic, and they weren't using it for something I'd describe as "scientific programming". Hope it gets released eventually!
msravi 12y ago
I've been using Julia in production for about 4 months now for algorithmic trading (not high frequency) and am very happy with it. I use Julia for the core algorithmic engine while my broker interface is in Python/Java/C++. The main aspects I'd like to see improve are the initial load times for some of the packages and the limitations in multithreading.

by thearn4 12y ago

oostevo 12y ago

For what it's worth, I tried and failed to convert my team of scientists/programmers (heavy on the science; not so much on the programming) to Julia from MATLAB/IDL/C++ for a project we were kicking off ca. 1 year ago. It's a really, really nice language that I find more consistent and easier to grok than MATLAB, but the deciding factor was the lack of the equivalent of MATLB toolboxes with all of their pre-rolled code that made our development as fast as it was. Edit: I should have mentioned: the Julia add-on packages are multiplying quickly, and are starting to look mature! http://pkg.julialang.org/ http://pkg.julialang.org/ I'm looking forward to them overtaking Mathworks on this front too.

jordigh 12y ago
Octave dev here, which packages are you looking for? I'm curious if it's as sophisticated as Simulink and Mupad, which are completely different programs that happen to be associated to Matlab, or if it's just the image package that we have almost all in Octave already. Where's the demand? Where is your demand? :-)
- oostevo 12y ago
  This was at an organization that I don't work for anymore, and my recollection is a bit fuzzy. I think it was Computer Vision, Signal Processing, Wavelet, and Image Processing. There were also some convoluted, non-technical reasons that we went with MATLAB.
  - jordigh 12y ago
    Could you contact me at jordigh@octave.org for some follow-up questions?

idunning 12y ago

There was a talk at JuliaCon about this exact topic, and they weren't using it for something I'd describe as "scientific programming". Hope it gets released eventually!

msravi 12y ago

I've been using Julia in production for about 4 months now for algorithmic trading (not high frequency) and am very happy with it. I use Julia for the core algorithmic engine while my broker interface is in Python/Java/C++. The main aspects I'd like to see improve are the initial load times for some of the packages and the limitations in multithreading.