4 ms·

I work in science. We agree that testing would be beneficial, but nobody codes well enough to actually get it done. To all language designers, there is a HUGE

by tel 15y ago

I work in science. We agree that testing would be beneficial, but nobody codes well enough to actually get it done. To all language designers, there is a HUGE space for a better scientific language. Make it easy for Matlab users to understand, but include better encapsulation and library support. Tie in testing and proving from the core.

densh 15y ago
You might want to check out Julia[1] programming language. I have absolutely no idea how good or bad is it but considering the fact that it's so young -- you are still able to influence it's development (e.g. suggest better testing capabilities) if you really wanted to. [1] http://julialang.org/ http://julialang.org/
- tel 15y ago
  The webpage is down, so I'll take a look at it later. I'd like to see some simple examples of things as well. It should be a one-liner to import a CSV file and do a least squares regression on different columns. It should also be a one liner to open an image, compute it's 2d FFT, and display it. It should also be one-liners to do numerical quadrature integration, compute the solutions to some ODEs, and maybe even backpropagation training of a neural network with just one layer.
achompas 15y ago
What about NumPy + unittest + all other Python libraries?
- tel 15y ago
  NumPy is making pretty strong headway in many communities. It's still not Matlabby enough for the majority, though. It doesn't make a clear improvement, so I think many see it it as just poorly replicating the features of Matlab for free. It's also pretty notoriously difficult to install, especially if you want LAPACK/BLAS. I wasn't able to get it running on many of our servers for that reason and had to revert to Matlab.
  - achompas 15y ago
    It's also pretty notoriously difficult to install, especially if you want LAPACK/BLAS. I almost said "no way!" then remembered how difficult it was for me to get Scipy 0.9.0 installed and verified via tests. Still, it's like Matlab ("it" being Num/Sci/Matplotlib) plus all of Python. That's a significant improvement over Matlab if you can get it installed.

by tel 15y ago

densh 15y ago

You might want to check out Julia[1] programming language. I have absolutely no idea how good or bad is it but considering the fact that it's so young -- you are still able to influence it's development (e.g. suggest better testing capabilities) if you really wanted to. [1] http://julialang.org/ http://julialang.org/

tel 15y ago
The webpage is down, so I'll take a look at it later. I'd like to see some simple examples of things as well. It should be a one-liner to import a CSV file and do a least squares regression on different columns. It should also be a one liner to open an image, compute it's 2d FFT, and display it. It should also be one-liners to do numerical quadrature integration, compute the solutions to some ODEs, and maybe even backpropagation training of a neural network with just one layer.

achompas 15y ago

What about NumPy + unittest + all other Python libraries?

tel 15y ago
NumPy is making pretty strong headway in many communities. It's still not Matlabby enough for the majority, though. It doesn't make a clear improvement, so I think many see it it as just poorly replicating the features of Matlab for free. It's also pretty notoriously difficult to install, especially if you want LAPACK/BLAS. I wasn't able to get it running on many of our servers for that reason and had to revert to Matlab.
- achompas 15y ago
  It's also pretty notoriously difficult to install, especially if you want LAPACK/BLAS. I almost said "no way!" then remembered how difficult it was for me to get Scipy 0.9.0 installed and verified via tests. Still, it's like Matlab ("it" being Num/Sci/Matplotlib) plus all of Python. That's a significant improvement over Matlab if you can get it installed.