5 ms·

I agree that Julia can't stack up to Python in terms of a data science / statistics domain. But in everything people use MATLAB for, Julia actually has the upsi

I agree that Julia can't stack up to Python in terms of a data science / statistics domain. But in everything people use MATLAB for, Julia actually has the upside in many of the qualities discussed in the article: Community, Package Ecosystem, annoying licensing bureaucracy etc etc.

socialdemocrat 6y ago
The best bet for Julia might be to be a Trojan horse in the Python ecosystem. By making the best of breed packages in specific domains it can become the first pick in the Python community rather than C/C++ based packages. If Julia is the engine that drives all the critical parts in Python rather than C/C++, then it has a way to get the foot in the door. People will stop and ask: Why am I using Python if I could just use Julia directly?
- andi999 6y ago
  Because you want a snappy repl?
  - FridgeSeal 6y ago
    Every major release has improved performance in the language and its use, and it doesn’t really take that much before most of the stuff you do is already compiled and you get a repl experience far superior to pythons.
    - michaericalribo 6y ago
      That may be true for elaborate analyses, but doesn’t really address exploratory analysis that changes dramatically from command to command. The use case for REPL-driven data science is experimentation, not performance
- phillc73 6y ago
  The same with R. There's already the excellent JuliaCall[1] package, for embedding Julia code in R. Listed in the README are some R packages, using Julia code through JuliaCall. [1] https://github.com/Non-Contradiction/JuliaCall https://github.com/Non-Contradiction/JuliaCall
- jakobnissen 6y ago
  To make Julia compelling as library code, it would have to be statically compiled. Having a massive laggy Julia runtime embedded in a library seem like a complete no-starter to me. I'm convinced Julia will eventually be able to be compiled statically, but doing that will likely make it feel decidedly un-Julia like (e.g. no dynamic dispatch, no type inference failure), to the point where a library maker would probably just want to use an actual static language instead.
zhdc1 6y ago
Julia has advantages over Stata and SPSS as well. The main thing holding it back is its ecosystem. R has better libraries and a much better IDE (RStudio), while Python is still the best option for machine/deep learning (and general programming). However, Julia has much better performance. R is arguably easier for someone with little coding experience to learn, but Julia isn't that much more difficult, and it's much more intuitive to code in than Python. Julia + RStudio + CRAN/BioConductor would take the cake.
- dagw 6y ago
  Julia has advantages over Stata and SPSS as well. I can't speak for Stata, but literally everybody I know using SPSS use it because of the easy to use GUI that allows quite complex analysis with basically zero programming.
  - zhdc1 6y ago
    I don't intend to downplay either Stata or SPSS for exactly that reason - they're very good at what they do. However, they have a limited scope (which isn't a bad thing), and if you want to go beyond it, you have to turn towards other options (e.g., R for extensibility, or Julia (or C, or whatever) for performance).
- superbcarrot 6y ago
  > R is arguably easier for someone with little coding experience to learn, but Julia isn't that much more difficult, and it's much more intuitive to code in than Python. Having some experience with all three of these languages, I find Julia much harder than R and less intuitive than Python. The relatively clean syntax isn't enough to make Julia an easy language.
siproprio 6y ago
Except plotting things

socialdemocrat 6y ago

The best bet for Julia might be to be a Trojan horse in the Python ecosystem. By making the best of breed packages in specific domains it can become the first pick in the Python community rather than C/C++ based packages. If Julia is the engine that drives all the critical parts in Python rather than C/C++, then it has a way to get the foot in the door. People will stop and ask: Why am I using Python if I could just use Julia directly?

andi999 6y ago
Because you want a snappy repl?
- FridgeSeal 6y ago
  Every major release has improved performance in the language and its use, and it doesn’t really take that much before most of the stuff you do is already compiled and you get a repl experience far superior to pythons.
  - michaericalribo 6y ago
    That may be true for elaborate analyses, but doesn’t really address exploratory analysis that changes dramatically from command to command. The use case for REPL-driven data science is experimentation, not performance
phillc73 6y ago
The same with R. There's already the excellent JuliaCall[1] package, for embedding Julia code in R. Listed in the README are some R packages, using Julia code through JuliaCall. [1] https://github.com/Non-Contradiction/JuliaCall https://github.com/Non-Contradiction/JuliaCall
jakobnissen 6y ago
To make Julia compelling as library code, it would have to be statically compiled. Having a massive laggy Julia runtime embedded in a library seem like a complete no-starter to me. I'm convinced Julia will eventually be able to be compiled statically, but doing that will likely make it feel decidedly un-Julia like (e.g. no dynamic dispatch, no type inference failure), to the point where a library maker would probably just want to use an actual static language instead.

zhdc1 6y ago

Julia has advantages over Stata and SPSS as well. The main thing holding it back is its ecosystem. R has better libraries and a much better IDE (RStudio), while Python is still the best option for machine/deep learning (and general programming). However, Julia has much better performance. R is arguably easier for someone with little coding experience to learn, but Julia isn't that much more difficult, and it's much more intuitive to code in than Python. Julia + RStudio + CRAN/BioConductor would take the cake.

dagw 6y ago
Julia has advantages over Stata and SPSS as well. I can't speak for Stata, but literally everybody I know using SPSS use it because of the easy to use GUI that allows quite complex analysis with basically zero programming.
- zhdc1 6y ago
  I don't intend to downplay either Stata or SPSS for exactly that reason - they're very good at what they do. However, they have a limited scope (which isn't a bad thing), and if you want to go beyond it, you have to turn towards other options (e.g., R for extensibility, or Julia (or C, or whatever) for performance).
superbcarrot 6y ago
> R is arguably easier for someone with little coding experience to learn, but Julia isn't that much more difficult, and it's much more intuitive to code in than Python. Having some experience with all three of these languages, I find Julia much harder than R and less intuitive than Python. The relatively clean syntax isn't enough to make Julia an easy language.

siproprio 6y ago

Except plotting things