Honestly who even needs octave unless you are trying to run someone else's Matlab? Numpy and ipython notebooks are superior to both in literally every way.
this post was submitted on 03 Mar 2026
237 points (100.0% liked)
Free and Open Source Software
22016 readers
69 users here now
If it's free and open source and it's also software, it can be discussed here. Subcommunity of Technology.
This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.
founded 3 years ago
MODERATORS
Matlab's moat is that their shit mostly* just works, and most people uaing matlab arent the same ones writing the check. Basically no dependency hell, no random broken libraries, no 30 different 3rd party options that for the same thing. If matlab has it, it almost always works, as expected, and they'll sell it to you and give you support if you have a problem. Stay inside Mathworks domain, you'll have a pretty good time. Basically I'm saying matlab follows the zen of python better than Guido does
As someone who has swapped from matlab to python, mathworks puts in real work from all the money they pull in. Shits expensive, but you get like... 50% of what you pay for. Even better if someone else pays. We did it for the money savings, but it definitely cost us extra dev time doing dependency management and version upgrade testing, and all kinds of little things.
*I got some issues with how they changed how figures are rendered, and that generally was causing issues during the changeover.
The moat is also external libraries, documentation, teacher materials and youtubers from India making incredibly specific videos on your exact niche subject.
In Octaves case many of these start to be covered, I did some courses back in Uni fully with Octave, but I couldn't do all.
I havent used Matlab (or Octave) for 20 years. I assumed by now that Python would have caught up and overtaken with all of the scientific/mathematic/computing libraries.
Not by a long stretch. There is one single toolbox that I use which requires MATLAB to run. The toolbox itself is under GPL, and it's fantastic work, but the authors don't have the time and resources to port it to pythonz and it's the only reason why I (begrudgingly) use MATLAB at all.
I'm in the exact same position.
Locked in software also stifles innovation of the soft itself.
But as usual, FOSS must be done by sweat and tears without pay, so it usually takes a while.
That's the thing. I think FOSS' main competitive disadvantage is the lack of money. Because money gets things moving/written, fast.
Looking at the R community, it seems like this isn't necessarily true. You can find so many resources to very niche problems. The main difference seems to be that there are many people working with R because large companies like Google use it. The same is probably true for Python as well, but I'm not as familiar with it.
I believe the only reasons MATLAB persists are Simulink and a C code generation engine that outputs embeddable code that conforms to some regulation or other in the engineering world. Does gnu octave have similar features? I've spent a while looking fruitlessly for python-based alternatives...
Wonder how Rust changes that game - honestly easier to learn than MATLAB....
Coming from a more science/engineering background than a cs/programming one and having tried both, I wholeheartedly disagree. Or at least it's a higher floor and lower ceiling situation.
And from an engineering industry standpoint the transition cost (or at least compatibility) from generated C, and the lack of a simulink-like block implementation/visualization and code generation pipeline still make it a hard sell. The block thing isn't just for comfort in those circles, it's to do with industry standards for system definitions and representation.
Don't get me wrong I wish Rust changed that game: I hate MATLAB with a passion; I'm just pessimistic.