Well, one context that I left out was that the course was pretty simple. We learned some basic loops, graphing, matrix operations, and writing some basic scripts to solve some problems. If you need a higher level functionality, then you’d probably struggle with GNU Octave, I don’t know.
So it can be done, simple as that.
I’ve just discovered Thonny! I’m not sure of the exact advantages over just vanilla Python though. Maybe because it’s an IDE.
It really depends on the course, but I think for general undergrad stuff, Python should be capable for most things.
As another commentor said, it kinda depends on what is the purpose of the course. If the purpose was to actually teach you the MATLAB ecosystem, then yea, sure, teach it all you want, but the institution has to provide the software.
But for an intro course? The students should probably be able to just use what they want.
Yeah, then other languages should be allowed as well.
I agree with that. It’s similar to Photoshop or Premier Pro. Sure, you could maybe, perhaps use open-source alternatives. But you’ll have to get used to a different set of (usually separate) software, dissimilar to what people all over the world uses.
Even though I’m generally for open-source software, I know that in heavy duty use, highly niche specialisations, and in industries in general it’s difficult to find equally competent software. That’s why I put emphasize on my specific situation, where it’s an introductory course. Heck, we ended up doing what could be done in Python anyway.
I’m not sure what would have happened had I insisted. I imagine that they’d probably ask us to obtain it on our own though, based on my memory that they were insistent that everybody must have it.
I see. That’s a bit rough that we require proprietary software to graduate.
I think that’s ideal! It’s supposed to be a lesson on numerical methods, not MATLAB.
I’m not sure how it works in the US but where I’m from, the way lessons are conducted are typically like this:
So I’m personally unfamiliar with the “shilling” of textbooks which cost up to hundreds of dollars for practically the same content, which, from what I’ve heard, is quite common in US colleges. This seems to be a very strange concept to me.
I think I get that as well. I used to talk quite a bit about open-source to my friends, but looking back, it seemed quite preachy (maybe because I was quite young at the time), and it never really changed anything. This is especially the case since open-source (or free software) is a philosophical approach to technology that many people might be unfamiliar with or simply don’t care about. I just simply use open-source software, supports devs/foundations, and only will talk the necessary bits if someone asks me about it.
I mentioned it to a couple of friends, but I think I didn’t get it across well to them that GNU Octave is supposed to be syntactically compatible with MATLAB. Also, they’re more comfortable using established software since everybody else is using them anyway.
Speaking about numerical analysis courses, I feel like one should be able to choose what programming languages they wish; the course should just aim to teach the fundamentals/principles of numerical methods, not what language to use. I get that it is much more convenient to streamline software choice, but still, why not use Python over MATLAB for undergraduate introductory courses?
I’m not sure about that since I’m not in any field that requires MATLAB at the moment. However, my specific case is for undergraduate introductory courses, and perhaps even at schools. To go even beyond this conversation a bit, any numerical / computational / algorithmic principles should probably be taught using Python. I had another numerical methods course where students can use any language they want, either C or C++ or Python. So I know it’s possible.
It’s probably the main reason why I think most open-source software will never be able to replace their proprietary counterpart: the fact that proprietary software are typically developed for either massive or highly niche industries, and so they are funded and are basically now integrated inside the ecosystem of such industries. As people use it more and more, Mathworks will develop more toolboxes with hardware integration, until it basically becomes the de facto software for that purpose (e.g. computation). I’m all for open-source software, but I don’t see a way out of it. Big companies with mega budgets can always improve their software, far outpacing any alternative open-source projects.
I don’t use MATLAB nor GNU Octave for my work, but I imagine that the hardware that I’ll operate on probably require MATLAB, and so there’s no incentive for me to use GNU Octave, especially if it has poor hardware support or lack of toolboxes or whatever such issues. This is a natural consequence of open-source alternatives being built from scratch typically with volunteers. That’s insane to me that GNU Octave is still somewhat usable for some basic computational work.
It’s $1K per head?
Piracy is so widespread in public universities here that nobody thinks about it as being wrong.
That’s interesting to me. So it has spread towards the public institution level, where many or most people think it’s just normal.
It’s literally the reason a lot of us in south America can scape poverty.
Probably one of the biggest examples of justified piracy. I’m not sure if it’s fully justifiable, but it is really hard to deny its benefits. One thing though, piracy as a means may be justified, but I’m unsure if it’s for the ends.
Out of curiosity, what is your field of work?
I’ve seen people mention this before. This will be troublesome if you require massive computing speed, otherwise it is still acceptable, since you’re basically using a MATLAB clone.
This is interesting! Thanks for sharing!