Hacker News Reader

Comment by nimonian | Hacker News Reader

Without having looked at the text in depth, I applaud the idea here - I'm convinced the web browser can help us improve upon maths textbooks printed on paper, but there aren't many 'hypertextbooks' like this in existence (partly, I think, because most people positioned to author a maths textbook want to use latex; and most people who are able to develop a web app aren't positioned to author a maths textbook).

I'm interested to know how the authors funded this, and if they have any data about the impact it has had.

View on HN

For my Linear Algebra (http://joshua.smcvt.edu/linearalgebra/), the thing that has held me back has been that the tech changes every few years. I can't keep revising something that I give away for free. In contrast, the LaTeX in which I wrote the book is 98% unchanged since the early 90's.

Perhaps the tech is settling down, I'm not sure.

Wow! Your textbook looks phenomenal. Eyeballing the table of contents and a few chapters, this arrangement of material looks exactly like what I'm advocating throughout this thread. Gaussian elimination and systems of linear equations before vector spaces, a coverage of fields in tandem with vector spaces, some treatment of sets and functions in appendices, generous usage of matrices to supplement theory...

And you include solutions for your exercises? This is fantastic! Do you have any plans to write something like this for functional analysis?

Thank you, that's very kind. Working on Theory of Computation. Don't know any functional analysis, sorry.

This is great news. Please model the Theory of Computation book after your Linear Algebra book (by including full solutions). This is an excellent resource for self-learners.

Do you have a timeline for the release of the book? Would you have any interest in writing an Abstract Algebra text?

> Do you have a timeline for the release of the book?

Several years. I have a very early draft, no doubt full of errors, that I am teaching from. It is linked-to from the linear page.

> Would you have any interest in writing an Abstract Algebra text?

Sorry, no. But all my stuff is Free and you could take the .cls file, etc. Go nuts.

Thank you for your contributions. I want to echo what others have said. Thank you for providing complete solutions to your linear algebra book. This is an incredibly useful resource for self-learners.

These things take time lol, he might not even want to write any other books after that theory of computation one!

Thank you for your free book. Another great book is Gilbert Strang's book: Introduction to Linear Algebra. Also Boyle: Linear Algebra and its Applications.

Perhaps your free book can be used to develop some evolving tool for the student. For example: the student take a test and then the book select the best exercises and topics for that student.

Any interest in doing that? Serious question. There are some platforms such as WeBWorK http://webwork.maa.org/ that seem to me to be promising targets.

A lot of work for a null reward. So not really.

So, you want him to do lots of work for no reward! How progressive!

Solving problems like this is one of the motivators for my nascent BubbleOS project — we need a stable software base that isn't upgraded every couple of years to replace known security holes with unknown ones, also introducing unknown incompatibilities. Executable code is clearly the best representation for many kinds of knowledge, but our existing systems software isn't up to the task.

Thank you for your book. I applaud all sorts of free texts.

I notice that there is an option to buy your book, and perhaps there are other options to somehow get money to you. I was wondering if you felt comfortable revealing approximately how much money has come to you as a result of your (donated) labours?

I occasionally get contributions via the PayPal button on the book's web page. I use them, among other things, to fund going to conferences. I say "conferences" but really there has only been one (very, very good) conference on Inquiry-Based Learning at Westfield State, in connection with my IBL Proofs text. To be fair, the contribute button has only been up for a year-ish.

I sell hard copies on Amazon. Well, really Orthogonal Publishing L3C sells them. This is a limited-profit company that Lon Mitchell set up to publish Free texts. Naturally we round the price up rather than down so there is some money. Since Amazon's price, $21.95, is not far from the price of media, you can guess that there is not much money. I'm reluctant to quote exact numbers but the dollar amount is not much, especially per hour.

However, my school, St Michael's College, has the wisdom to give me professional credit for the work so that counts for a lot.

Thank you for letting me know. It's good to hear that St Michael's College values your work, too.

Cheers.

Thank you for including the solutions to the exercises. This truly is godsend. You have no idea how much it is helpful for those of who try to learn outside an educational institution.

It is very hard to feel confident in learning when I have no way to gauge if I am on the right track or not.

Thanks again.

Perhaps this is a reason why there are some nicely designed apps with not-so-great content, and lots of books with legendary content: the whole ecosystem around app development isn't very friendly to the long-term process of writing a classic text.

It's also hard to focus on design, coding and content solo, and writing great, cohesive content tends to be a private endeavour. I'm not surprised to see immersivemath.com is written by a team.

I know (through fond memory!) your texts have some interactivity in hyperlinking to solutions and I found that very useful, but making it work must have taken time and been a bit of a headache. Extrapolate that to writing a webapp and it's easy to imagine you would never finish.

And yes, the words 'legendary' and 'classic' are direct compliments to you: your Lin Alg sits next to Spivak's Calculus on my bookshelf. They are both immortal.

Thank you, I appreciate it.

It's true that things went from embedded Java applets (or, eh, ActiveX objects, thanks Microsoft!) to embedded Flash applets to... HTML5/Canvas, but the latter does seem finally stable-ish, especially once web assembly is widespread (which it kinda is, already: https://caniuse.com/#search=wasm)

The technological endurance of LaTeX is legendary, though

Forgive me but the page says "WebAssembly or "wasm" is a new portable, size- and load-time-efficient format suitable for compilation to the web." It is the "new" part that makes me shy.

It IS newish and it is fair to be cautious of new things, but this one may stick around for awhile, as it's mostly just a modification/refinement of what Emscripten was already doing successfully https://github.com/emscripten-core/emscripten acting as a build target for the vast majority of C/C++ code out there, and Emscripten is not at all new. Here are some demos/examples to show what it's already accomplished: https://github.com/emscripten-core/emscripten/wiki/Porting-E...

And JavaScript itself (for better or worse) isn't going anywhere, it is literally "too big to fail" at this point.

I'm a web developer and have been following this space since the Web came out, these are just my best educated guesses. I would deem "developing with a target of HTML5/Canvas/JS/emscripten/wasm" to be future-proof for the foreseeable future (the next 10 years).

BONUS: You may already know about being able to convert LaTeX to HTML5 via JS, but just in case you haven't, this is pretty cool: https://latex.js.org/

Thanks. I was unaware of the web page. I've saved a bookmark.

I'll trade you. I perceive that the standard for conversion at this point, which you may not know about, is https://www.tug.org/tex4ht/.

(Sorry, do you know the author of the latex.js package? I would like to contact him.)

I found his email in the package information: https://github.com/michael-brade/LaTeX.js/blob/master/packag...

Your book is genius. Really. Thanks for making it free.

Your book taught me a lot. I got stuck some of the way in (dropped out of education at 16 - kids and full time job don't offer much opportunity for the focus it deserves). It's the furthest I've progressed with this topic and helped me understand in ways other books didn't. Thank you!

It always makes it a good day when I get a note like yours. Glad it helped.

I'm also working on an education related project. With projects like this....the truth is, it is mostly a labor of love or passion for the subject that drives you. There isn't much money to be made (even to just support yourself) in creating free courseware. If you want to make money you are better off just working for a traditional university.

You can even see it for yourself. This guy started this project 3 years ago. Fast forward to today we can see that his youtube channel has 500 subscribers and most of his videos have around 1k views...this is after 2-3 years of each video being uploaded. Basically, no one is using this site. It is no surprise that we see so many of these education type projects being half built. The developers eventually toss in the towel.

Personally I think we need a combination of both to really change education though - open educational resources and the associated tools to learn the subject better. Preferably free but also low cost might work or supported by donations.

Right now the big publishers pretty much control the entire market. They own the content and they create the tools. Then they spoon feed it to university faculty (by making tools to basically "drag and drop" the content for teachers) and the student/taxpayers ends up footing the bill for it.

This resonates with "There are no projects like side projects" https://crastina.se/theres-no-projects-like-side-projects/ I wrote a few years ago (and it is still my credo).

Since I create educational projects (luckily, Quantum Game with Photons http://quantumgame.io/ was partially supported by an outreach grant; all other writings and stuff - not so much), I would love to see some way to support such endeavors.

Yup, that write up definitely hit the nail on the head. I could work on more lucrative side projects but there are just some side projects that you have to do because it just has the potential to be great and do real good in the world. Mostly what I see these days is a void in the educational market - you have a traditional educational system practically resistant to changing and then you have all of these emerging technologies. The emerging technologies go on to change other industries....but education, nothing changes, or very slowly. That's where my side project ideas some into play. I'm working on making a course right now in a traditional college level subject then adding a bunch tools (student data analysis, simulation games, interactive lectures etc.) to help students learn the subject better. Simply put, I want to take traditional college courses and then apply the latest technology to teach that subject. Let's create the world's best online course, create a bunch of really useful tools to learn the subject then scale it up and give it away to everyone for free.

Also I looked at your game and I really love it. It is so much fun (even though I don't know much about quantum physics). Now just imagine if we have a traditional college-level subject, such as quantum physics, and then use tools like your game to help people learn better. You still learn all the usual topics but the overall learning experience is made SIGNIFICANTLY better by having these extra tools to use to learn. Your game is an educational game but using the same technology used to create that game could also be used to create all sorts of things - experiment simulation, interactive graphics etc.

Sadly, innovation in teaching is neither a priority in academia (since it is not research) nor in the industry (since it is this kind of activity that ).

I wouldn't say there is no innovation though. There are science-based games (https://github.com/stared/science-based-games-list; it was here some time ago: https://news.ycombinator.com/item?id=14661813). I started collecting interactive machine learning, stats and maths: http://p.migdal.pl/interactive-machine-learning-list/ (a lot of things; of course, Immersive Linear Algebra is there). Most of these projects are side projects, though.

Also - I had a plan to extend the quantum game (it is nice to hear you enjoyed it!) to more particles... and then, a full course, with experimental widgets (e.g. showing interference, then explaining it, then giving space to play with it). With some framework (maybe $, but I care even more for collaborators and deadlines), I would love to continue that.

What are your projects?

> partly, I think, because most people positioned to author a maths textbook want to use latex; and most people who are able to develop a web app aren't positioned to author a maths textbook

Luckily we have interactive notebooks (like Jupyter) and MathJax for LaTeX. You can have all the beauty of LaTeX typesetting in your browser and make it highly pedagogical with interactive computation.

+1 for Jupyter and Miniconda. I was at a Python meetup recently and one of the seminars was intro to deep learning. It is very easy to setup and quite seamless. Great learning tool as well, because you have output side by side with the teacher’s code. Many modern IDEs still struggle with this (Lighttable being a notable exception).

Most notebooks have Python code in them. How to make that interactive in browser? Are there smooth solutions out there, be it client-side or some hosted service? Or does one have to write all the code in JavaScript?

The only reason I was able to learn any math in college is because you don't need a computer to do it, so you can avoid the distractions. A browser, the source of all the distractions, seems like a really bad place to learn math.

I think it's also the greatest tool for learning there is, so it's worth it to over come that tendency and develop discipline when using it for learning.

Just because some people choose to get distracted doesn't mean the whole technology is worthless.

I mean there are a million and one useful resources on the internet that enhance learning and just ot getting distracted is also an option

as an undergrad who's currently downloading mathematica to play with .cdfs instead of actually studying them, this hits home

I see your point! There is something very focused and calming about having a book open with no screens around. I hope it is possible to recreate this flow state in a browser (because I plan to try over the next few years). I'm imagining a very minimal UI, 66 characters per line typeface, a distraction free mode (full screen kiosk, a la Sublime Text), a 'dusting' of gamification maybe. I have spent hours on Project Euler before without switching tabs so I believe it's possible to focus at a computer, though I agree that a lot of the positives of the textbook experience don't easily translate; but I still believe that, with respect for the learner, their working memory and attention, a great in-browser learning experience can be achieved.

To an extent I disagree. Yes, drawing diagrams can be useful, but simply solving lots of problems is honestly the best way to learn maths. I also much prefer reading from a physical book. Im not sure simple animations will improve things massively.

I don't mind if a book skips some details and I have to fill it in myself, for example drawing a diagram to help understand a key point.

Perhaps in school most people treat maths more mechanically, and exploring requires more mathematical maturity. I don't think some basic animations will improve that directly.

That being said, people like 3b1b on YouTube are in the intersection of people who can animate and know the maths. They certainly produce things which spark the imagination.

I think examples rather than problem solving is the way to go. Some very advanced topics are arcane enough that they really do need hand holding.

Compare Dirac's book on general relativity with one people actually read(To learn from) and you'll see why hand-holding and diagrams are helpful

Two of the authors, Kalle and Tomas, are at Lund University, in sort of the perfect fields for this project - Computer vision and computer graphics (Tomas wrote _the_ book on real-time rendering). I think Jacob is a graphics guy as well.

My guess on the funding is spare time and spare university resources.

>I'm convinced the web browser can help us improve upon maths textbooks printed on paper, but there aren't many 'hypertextbooks' like this in existence (partly, I think, because most people positioned to author a maths textbook want to use latex;

We have much better tools right now such as Jupyter notebook which as accessible to everyone. However, your concept of hyper textbook is intriguing.

What do you mean by that? Can a webpage be an hypertextbook? Then aren't all online books already hypertext books?

Anyways, I am working to build a different medium of learning called Primer. The courses on Primer are basically books that teaches you through conversation. It offers much more than that though. Creates latex pdf, generates flash cards , implements SRS etc. Can you check if Primer matches your description of Hypertextbook.

www.primerlabs.io

Interesting concept, Primer. Just a note: Scrolling is rather erratic on my iPad.

I am aware of that issue. I added parallax and regretted it.

I will change the front page once I launch in March.

The website works well in Dekstop only.

Thanks!

I saw the word hypertextbook somewhere online once and it just stuck in my head. For me, it has come to mean a textbook (so, beautifully typeset, well organised and written, easy to tell where I am, distraction-free) enhanced with code (keep track of learning, tasteful interactives, maybe video, a smattering of ai perhaps). A webpage/app lends itself to this nicely. If I'm honest, the idea is a little inchoate but my instincts tell me there's something to it.

Primer looks like an interesting system - I'm on the mailing list, good luck!

You can write texts in LaTeX, and post them online using Ximera!

https://ximera.osu.edu/mooculus

You can also literately program in Lean, and get out math theorems like http://pat.perso.ens-lyon.org/M1P1/

Reply on news.ycombinator.com