（评论）

（评论）
(comments)

原始链接: https://news.ycombinator.com/item?id=40329388

该用户对教科书中多媒体内容的潜力表示热情，但由于专有格式和有限的视觉质量而担心其寿命和可访问性。他们欣赏写得好的书籍，并提到准确性的重要性，尤其是在数学背景下。用户称赞某本线性代数书籍并分享他们的自学经历。他们批评在教科书中添加人工智能的建议，并强调无错误内容的重要性。他们还建议通过过滤、调制、空间计算和基于网络的技术来提高交互性。最后，他们都更喜欢简洁的设计，而不是教育资源中华而不实的演示。

Thank God people are moving towards this modality. I have a fervent hatred for textbooks and publishers. The former are antiquated, static, badly formatted, and often ride with distracting garbage in the margin, or worse inline. It makes actually reading them far more difficult than needs be, with unremarkable asides that may span pages and that are easy to get pulled into. While I understand that they have a purpose, they aren't for everyone so having a platform with the dynamism of a webpage is something that I hope will inevitably lend itself to future development along this course. Not to mention being able to have interactive questions that give quick feedback rather than requiring turning through pages to find out if you're on the right track... And this interactive stuff is just an excellent means to drive meaning from terse and difficult to explain systems.

Cheers to the folks that put this together, a thousand thanks for the travails you've been through to blaze these trails!

My undergrad math professor created one of the first fully online linear algebra texts: http://linear.ups.edu/html/fcla.html It's integrated with Sage, a Python library for studying (among other things) number theory. Another prof at the same university also wrote his own linear book, using a lot more illustrations, but as a traditional textbook.

I see this book as a solid evolution in both directions. Nicely done!

My gut reaction to this sort of thing is that textbooks work great and have a timeless quality. But with a little more thought, they certainly aren't optimal. Being able to see a video, hear a sound, or play with a figure can be powerful learning. I think my hesitation comes from the fact that this stuff just doesn't seem to have longevity (think educational java applets) and tends to be proprietary (shitty Pearson learning hw). How can we get newer media as robust as text and pngs?

It does not have to do with the book or any other medium. It comes from the learner. You are right about the problems mentioned.

1. Stuff does not have longevity. This is because most of the new tools are outside curriculum. This does not encourage students to learn something extra.

2. Proprietary. Books providing add ons have a different goal. They want to sell their books. So the addons or just that. Add ons. They are not complete.

These were some of the thoughts which encouraged us to build [1]. Keep it curriculum focused so kids don't learn anything extra. No books to sell :)

[1] https://books.innings2.com/

If only there was some kind of magic device that has basic input facilities coupled with some kind of simple graphics output.

I jest, but I think a very basic computational platform - think 6502-style - with simple graphics could be standardized upon and anointed as universal educational computation base.

If you don’t touch the instruction set this can go a long way. The outputs of course need to be universal, but we have some experience with outputting stuff on machines. I think we are able to come to some sort of useable situation.

Think PDF or whatever with built-in NES emulator.

Is that what you have in mind?

I hope that this sets a standard for future textbooks/publications. I haven’t been able to grasp several concepts in math unless I was able to properly visualize it, which most modern textbooks do with a terribly compressed and unsaturated JPG.

This looks like a fantastic resource. After a quick scan, I couldn't find any information on how this book was programmed / created. Does anyone know if a particular framework was used, or if this was all coded by hand?

I studied physics-based rendering from a book by one of the two authors (T. A-M), and it was written excellently. I'll have a look at this for sure, as I need a refresher every now and again.

Which "subject" would you normally ascribe linear algebra to?

It looks like this is a first module within the Immersive math system. No doubt there will be more.

Why not look at what is presented rather than what is not?

Cool to see stuff like this which makes math fun for everyone. I love stuff like this because it brings together two things I love, math and programming.

I mean textbooks are cool. But with the tools available to us we should be able to make almost any textbook interactive. It will need effort, pedagogy, programming skills and design. But it's certainly worth it.

Making an effort with this. Started with 6th to 10th grade math. Let's see how this goes.

Hell yeah. Linear algebra was a hole in my math education and has turned out to be far more relevant to my work than the calculus I spent so much time on. I’ve been teaching myself piecemeal on an as-needed basis but this looks like a great opportunity to finally get a cohesive overview.

The biggest barrier to learning is having a real motivating need to learn the thing. Young people aren't as dissolutioned about the uselessness of what they are learning relative to all the demands of adult life.

I've learned amazing things when I had a real reason pop pup, after a decade of idly wanting to learn it.

Looks incredible! All this needs is a little bit of conversational AI magic in the background to filter and modulate the content according to plain-English student questions and its go time.

Note that this was finished in 2019, so now would be the perfect time for someone to polish this up and expand it to the rest of math! Assuming this is threeJS, you could get an open-source file format going for simulations, and even host crowdsourced applications of it to existing popular math textbooks by Figure/page #. I mean, linear algebra is cool, but the market for good free geometry education is limitless

Does anyone know if the big names in math education offer simulations yet, or is it all animations/images/videos still?

EDIT: definitely ThreeJS — love the vector chapter. What this needs is true spatial computing support - not pages with nested simulations, but site-wide (SPA-wide) simulated objects. What if every student in geometry class could have their own simulation on their Chromebook as they read/follow along? I can’t wait.

It really pains me to see someone suggesting adding AI to a book like this. Current AIs are infamously bad at math. The last thing we need is ChatGPT misplacing a minus sign and confusing readers or setting back their understanding by weeks.

I remember a friend reviewing some math before starting grad school was stymied by a typo in her textbook for an inordinate amount of time. It’s really vital that instructional materials avoid errors as much as humanly possible. AI right now ain’t it.

In my last homework the professor omitted a required assumption and I nevertheless proved the false assertion. Extremely embarrassing. This happened earlier in the semester and I correctly failed to finish the homework problem. I am getting tired I guess.

You clearly have no idea how effective an interactive conversation with a text can be. An AI doesn't have to be "good at math" to be useful. People (and programs) who are "good at math" are a dime a dozen. To be useful to a student, a language model just has to be good at answering questions about math.

That part works, right now. Try it. Go to ChatGPT4 and pretend you're a student who is having trouble grasping, say, what a determinant is. See how the conversation unfolds, then come back and tell us all how "infamously bad" the experience was. Better still, ask it about something you've had trouble understanding yourself.

Yes, that makes a big difference.

Many people on HN formed their opinions on the basis of GPT3.x-generation models, though. They asked it a question, they got the nonsensical or hallucinated answer they expected, they drew the conclusion they wanted to draw all along, and by golly, that settles it, once and for all.

(Not the person you replied to, but) I just re-read it, and the "canned retort" still looks completely accurate and relevant. Can you elaborate on why you think that AI's (known, admitted, and inherent) propensity for hallucination _wouldn't_ be disastrous in the context of pedagogy?

If the original comment had _just_ proposed to direct students to locations _within_ the original content ("filter"), it would have been less-impactful - being directed to the wrong part of a (non-hallucinated) textbook would still be confusing, but in the "this doesn't look right...?" sense, rather than the "this looks plausible (but is actually incorrect)" sense. But given that the comment referred to "Conversational AI", and to "modulat[ing]" the content (i.e. _giving_ answers, not just providing pointers to the original content), hallucination is still a problem.

GP's comment has been edited since my post. The original said something like "regenerate diagrams according to student questions". It's obviously a bad idea if you're trying to learn vectors and the entire diagram is flipped over the X axis, for example.

Nonetheless, today's AIs still regularly contradict themselves from one sentence to the next. Even if they're only generating text and "modulating" (which I take to mean rephrasing/summarizing), mistakes can and will happen. I stand by my comment even as it applies to the edited GP.

Filtering Modulating means selecting relevant excerpts. Think "AI for Search", not conversational chat generation. This is what LLM has been exceedingly good at, in the Alpha (DeepMind) series of projects.

+1 for Spatial Computing here -- I see immersive here and just think 2D animations of 3D concepts, good start though it may be, is leaving possibilities on the table. 3D consumed inside a fully 6DOF 3D animated space is a better environment to transfer meaning. These collections of links could be piped into WebXR with just a little tweaking and really be immersive.

Just taking something like the threejs GLTFExporter and combining it with modelviewer.dev on the fly could enable a 'view in AR' button compatible with both SceneViewer and Quick Look (i.e. most mobile devices available today).

maybe the layout just messed up on the resolution I looked at it with, there were boxes all over the place, the spinny thing on the main page scaled out of its bounding box. also I think it would be helpful to have some consistent design cues for which elements are interactive or what you can do with them.

if nothing else I feel each interactive canvas should have some kind of description of what you can do with it so you don't have to trial & error

It’s possible I’m alone in this, but the textbook / vaguely “academic” style of site always comes across as messy to me. That said, I think we’re spoiled by vaporware that invests huge money into style and flashy visuals — as anyone who’s been in school during the Blackboard/Piazza era knows, education software does not have to be pretty to be popular.

One critique that glares in my face is "you changed the textbooks via hackers and now my children's development is your fault". These accusations fly when money (e.g. big textbook starts a media campaign) is on the line. As such, these interactive apps need to be able to handle that level of criticism if in a public school (before college).

These applications probably need to be client side only enabled and a hash that agrees to the hash of the "approved curriculum" for most parents to think that this might be okay, or a read-only USB rental from the school library that gets verified when it is checked back in.

I will not let the government install unknown tech on my computer, but I am willing to download a read only file. I do not want to have my children learning "new math" and be able to point at the technology as the problem "because hackers" or some other BS that has sounds true but has a high amount of effort to disprove coming while chatting with parents at a bus stop or PTA meeting.

https://en.wikipedia.org/wiki/Brandolini%27s_law

（评论） (comments)

（评论）
(comments)