HN Academy

Author here. I made this demo and a related matrix-matrix multiplication demo [1] back in 2015 for Robert van de Geijn's Linear Algebra: Foundations to Frontiers MOOC class [2]. In the light of Spectre attack and recent browsers' changes to reduce precision of timers, I remembered of this project, and decided to check if it still works now, 3 years later. Surprisingly, it still works well!

The source code is available on GitHub [3].

[1] https://maratyszcza.github.io/laff-demos/dgemm.html

[2] https://www.edx.org/course/linear-algebra-foundations-fronti...

[3] https://github.com/Maratyszcza/laff-demos

⬐ madez

Could you make the website viewable without JavaScript?

Edit: I think the downvotes are unjustified. For clarification, if that wasn't clear by context, I don't expect to get the JavaScript test results from my computer while viewing the website without JavaScript. Demanding that would obviously be nonsense. Rather than that I assume there is information on that website that is interesting to read even without using JavaScript personally. Or is using JavaScript now a requirement to learn about JavaScript?

⬐ StavrosK

I guess you'd have more luck if you had asked "can you make some examples accessible for those of us who don't run JS?".

⬐ madez

I didn't know what to expect to see. I vaguely assumed to see some text and data. That's why I carelessly formulated my question like I did. Still, I think the difference is small and my question wasn't extraordinary.

⬐ Cyph0n

Is there a particular reason why you can't enable JS yourself? Metered connection? Low-end machine?

⬐ Tehnix

Metered wouldn't really matter would it? Scripts are still downloaded, right? (never had JS off, because the web)

⬐ swsieber

Metered could matter - it's very easy to disable downloading of any non-inlined JS (like with uBlock)

⬐ madez

I consider it an unacceptable form of code deployment. It's unsafe in the computing sense, and leads to a ecosystem where users are less and less in control of the software they use.

⬐ Mister_Snuggles

I use uMatrix, this site just shows up as a white page until I allow it to load some JS from a third party. Once I allow the JS, I see a graph that gets built without any explanation of what that graph means.

⬐ detaro

There's only a graph of the results, so you are not missing any other content.

⬐ madez

Thanks for the info.

I agree. As far as calculus goes, I am more enamored with books like Spivak's (http://www.amazon.com/Calculus-4th-Michael-Spivak/dp/0914098...) that take a proof-centric approach to teach calculus from first principles.

Incidentally, for those who want to learn linear algebra for CS in a mooc setting there are 3 classes running at this very moment:

https://www.edx.org/course/linear-algebra-foundations-fronti... (from UT Austin)

https://www.edx.org/course/applications-linear-algebra-part-... (from Davidson)

http://coursera.org/course/matrix (from Brown)

The first 2 use matlab (and come with a free subscription to it for 6 months or so), the last python. One interesting part of the UT Austin class is that it teaches you an induction-tinged method for dealing with matrices that let you auto-generate code for manipulating them: http://edx-org-utaustinx.s3.amazonaws.com/UT501x/Spark/index... .

And of course there are Strang's lectures too, but those are sufficiently linked to elsewhere.

⬐ pakled_engineer

My calc I course in university was applied calculus without a text. I had to go back and redo single variable by reading Spivak (and Polya's How to Solve It) to figure out the proofs in Concrete Math by D. Knuth, et. al.

Learn linear algebra while writing a linear algebra library using the latest techniques (starts at the end of this month): https://www.edx.org/course/linear-algebra-foundations-fronti... I think there's also a non-mooc version at http://www.ulaff.net/

Unfortunately, I had to give up on the course last year because my math background is even more limited than yours (you'll need to know how to construct proofs). So time for me to learn calc, I guess. :)

⬐ rifung

I don't believe Calculus is a prerequisite to Linear Algebra. Constructing proofs is something that's used in all branches of math as far as I know. If proofs are the only thing stopping you then I'd recommend just trying your hand at a few to get the hang of them. I can't imagine they'd be taught in a Calculus course either.

⬐ muraiki

Oh, I realize Calculus isn't a necessary prereq to linear algebra. But my impression is that people normally learn proofs alongside Calc than learn proofs alongside Linear Algebra. To be honest, I was writing more out of a desire to be humorous than accurate; sorry!

Anyways, I've picked up "The Haskell Road to Logic, Maths, and Programming," so hopefully I can learn proofs alongside programming and logic, which is probably the best route for me at this time.

Another upcoming edx course that might be of interest to people on here:

https://www.edx.org/course/utaustinx/utaustinx-ut-5-01x-line...

    Linear Algebra - Foundations to Frontiers

    Learn the theory of linear algebra hand-in-hand with the practice of software library development.

(can't post a thread of its own, as was submitted too recently)

⬐ muraiki

I'm very much lacking in math education and this seems like it would be a great approach for learning linear algebra. Do you know how this compares with "Coding the Matrix: Linear Algebra through Applications to Computer Science"? The two approaches sound very similar.

⬐ None

None

⬐ ambler0

I signed up for "Coding the Matrix" and didn't finish. Before that, the last math I had studied was Calc 1 during my freshman year of college, over a decade ago. My impression was that it seemed like a good class and a good professor, but it moved quite quickly and was hard to follow for people lacking a background in higher math. I did get very comfortable with writing comprehensions in Python, though.

I still want to learn Linear Algebra, myself, so if anyone has suggestions, please post them.

⬐ ivoflipse

You could buy the book, which isn't that expensive and covers everything from the video's as far as I can tell.

That allows you to study it more at your own pace. I'm taking the Visualizing Algebra course from Udacity, after working halfway through Coding the Matrix, because I found my algebra skills to be lacking.

Too bad its no longer possible to get a certificate on Coursera though, now I have to wait until the next iteration.

⬐ muraiki

Thank you for this feedback. You have most likely saved me a lot of angst. :)

⬐ VLM

Hit archive.org or whatever for "Strang Linear Algebra" and you'll see the MIT OCW videos for free. The quality is, um, very turn of the century, but you're watching to learn, not critique video codecs.

If you want to spend money, from my bookshelves:

Strang (the guy in the videos above) knows one or two things about Linear Algebra. His textbook is legendary. Bring lots of $$$, like three figures.

"The Manga guide to Linear Algebra" Yes, that is exactly what it sounds like. I think if you have to start somewhere, maybe this is it. Cheap. $

"Matrices for Engineers" by Kraus. There's about 50 textbooks along the lines of linear algebra matrices for engineers programmers using $math_application or $calculator or $chicken_entrails and similar title permutations. You'd think every engineering program in the nation is required to use a different text. This particular text was pretty good. If I recall correctly, reasonably priced $$.

I would suggest reading them in the order of the comic book, the engineer book (or any of the dozens of equivalent college texts), and Strang. Coincidentally thats also order of price.

At one time I understood everything in the comic book and the engineer book. That was a long time ago. Strang mystified me in parts. So I'm not going to pretend to have THE perfect answer. It is entirely likely in the last decade someone has written the Uber text to replace them all. Probably a new edition of Strang is out by now.

I would estimate the effort required to be about one programming language.

Best of luck to you.

⬐ nextos

IMHO the uber text for intro LA is Axler's.