Hey everyone! I really like how websites like Brilliant try to teach in an interactive way. But it’s like $160 (or more depending on the region) a year so I’ve been making a free one.
It’s called OpenAha. right now there’s 140+ courses and 1000+ lessons in math, physics and chemistry. no account or ads.
A few to try if you only have a minute:
- untangle a graph: https://openaha.org/courses/graphs/planar/?step=1
- draw the house in one stroke: https://openaha.org/courses/graphs/konigsberg/?step=3
- program a little robot: https://openaha.org/courses/coding/sequence/?step=6
- Two pendulums passing energy back and forth: https://openaha.org/courses/waves/coupled/?step=3
This project is mostly implemented with claude: I planned every course, then wrote the lessons with AI. so far only about one third of the pages have been carefully reviewed by a person. That’s why I’d really like people to poke at it. If a lesson is confusing or sth’s wrong, make a PR or hit the flag button in the lessons. Also feel free to contribute on github to make this even better.
code: https://github.com/open-aha/open-aha.github.io
Edit, to clear something up: OpenAha isn’t affiliated with Brilliant and doesn’t use any of their content. I don’t even have a Brilliant account. Brilliant just uses the style of learning I liked (try things first, explanation after). every lesson was written for this project, and we’re improving it as an open source community.
And yes it’s not finished, there will be mistakes. That’s what the flag button on every step is for, every report gets fixed in public on github
I hate how every single Open Source project recently is just worthless AI slop.
Removed by mod
I get it mate, there’s a lot of low effort stuff out there right now. I did use AI a lot, but this isn’t a one prompt thing, and it sure isn’t done yet. I’m fixing lessons as people report problems. if you try one and it’s actually bad, easily use the report button and we’ll fix it.
I can imagine that with this contribution system, it’ll become WAY better than Brilliant. Take a moment and imagine that for a second. We’d have the best interactive learning software that’s also open source and free for everyone.
You can’t educate people with data that could be wrong and expect the people learning to report problems. They may have no idea if the graph is right or wrong and continue learning from something that was essentially puked on the page from whatever unknown data source the LLM had. Do NOT teach the next generations with unverified information. That much should be obvious
I’ve already read through at least 30% of all the pages myself. I don’t think we’ll have problem verifying stuff. And how’d I know that other non opensource websites don’t have mistakes? I’ve seen many such problems in the biggest names like leetcode and khan academy.
I’ve already read through at least 30% of all the pages myself
wait, you posted this here when you’ve looked at less than 100% of it yourself? really?
this is incredibly disrespectful of other people’s time. please update the body of the post to disclose that fact 🤦
Removed by mod
You are single-handely proving why teachers matter.
I tried your first example, the planar graphs one. It made some sense until the “How many lines fit?” challenge which asks “What is the largest number of edges a planar graph with 10 vertices can have?” - the earlier steps do not remotely explain how to figure out the answer to this question, nor does the hint given when an incorrect answer is entered.
And it’s the same with the next challenge after it: “A football is stitched from pentagons and hexagons, with three patches meeting at every corner. Engineers design geodesic domes and virus shells the same way, some with hundreds of hexagons. However many hexagons there are, how many pentagons must such a closed ball have?” - there is nothing earlier to prepare someone to answer this question (except for the fact that this challenge is titled “Always twelve” and the answer is in fact 12 😂).
I’m curious about your background and plans for this project. Are you an educator? And/or do you have connections with educators who plan to use this with students? Is this a solo project so far? What is the likelihood that you’ll get bored and abandon this project in a week or a month from now? It is difficult for me to imagine many strangers on the internet volunteering their time to help you develop this when your website doesn’t preemptively provide answers to these and other questions.
tbh I am very tempted to remove this post from [email protected] 🤔
It doesn’t look like you’re a moderator, according to the sidebar. How would you be able to remove it if you wanted to do that? I am not being snarky. Genuinely I’m confused.
I agree with you that the biggest weakness of the project is there isn’t enough clarification of conceptual topics and there’s no place to get more info on things. There’s a suggest an improvement flag also, but it’s in a strange place (not near the topic/question) so it’s a bit unintuitive for suggesting the conceptual explanations are weak.
But I also think that if the person who made this project wanted to try to correct that, they probably could. I hope the developer answers some of your questions. I think the “What are the chances you’ll abandon this project?” is a valid question. Because this project needs work to be more user-friendly and if people give feedback and it doesn’t end up being implemented, then the people giving feedback were wasting their time.
It doesn’t look like you’re a moderator, according to the sidebar. How would you be able to remove it if you wanted to do that?
i’m an admin on the instance this community is on.
Thank you for your feedback. Honestly many of the best IMO books I came across challenged readers in similar ways, but that style isn’t suitable everywhere of course. And yes, so far I’ve found and corrected many such stupid mistakes 😄 That’s why community contribution will be really important here.
Regarding my background, my full name is publicly visible on github. You can search that up to see some of my background, education and activities. I have talked to some teachers about this and used their feedbacks, but there’s still a long way to go. I can’t foresee the future, but I’m serious about exploring this projects potential.
That’s why community contribution will be really important here.
requiring a github account to contribute is a major friction point; i highly recommend that you build something to allow people to give you feedback without requiring them to make an account with microsoft or anything else.
also, the website says:
AI tools helped us build this many lessons quickly. What turns a good lesson into a great one is human insight: the moment a step confused you, a clearer example, a mistake only a careful reader spots. OpenAha is open source, and every suggestion is read.
but you go much further in this lemmy comment and acknowledge:
I’ve already read through at least 30% of all the pages myself.
I think you really need to clearly disclose early - not only in this lemmy post but also on the website itself - that most (or any at all, if/when you’ve read most) of the lessons have not yet been read by any human at all.
Asking teachers to try this with their students without disclosing this crucial fact is not really OK.
I agree with this. I like AI and really don’t like people labeling anything AI-created as slop. But in order for something to not be AI slop, the developer needs to review the output and provide feedback to the model. If this is “draft 1” of the content and the dev only reviewed parts of it, I understand dev’s excitement and wanting others to see it, but it seems like dev should be beta testing draft 1 with the model and doing some revisions before asking others to check it out.
Because some of the phrasing is unclear enough in its explanations that a typical person might not be able to follow it. And I feel like if dev had read more of it, dev probably would have noticed that…
I agree. I’ll make the modifications later. Thank you.
I agree. I’ll make the modifications later. Thank you.
Thanks, glad to hear it.
How much later, though? If it’s not soon, I think I will delete this post for now (i’m an admin on the home instance of this community). You can still edit it while the post is deleted but only you and mods and admins will be able to see it; send me a message after your edits are made and I’ll undelete it.
To be clear what I am asking for is an early/prominent/clear disclosure, both on the website and in the body of this lemmy post, of the percentage of the lessons which you estimate have not yet been carefully read by you (or presumably anyone else).
Thank you, and I hope this isn’t too discouraging!
Regarding my background, my full name is publicly visible on github. You can search that up to see some of my background, education and activities. I have talked to some teachers about this and used their feedbacks, but there’s still a long way to go. I can’t foresee the future, but I’m serious about exploring this projects potential.
I'm going to interpolate that answer into answers to my questions, let me know if i've gotten anything wrong:
Are you an educator?
No.
And/or do you have connections with educators who plan to use this with students?
None with plans to use it.
Is this a solo project so far?
Yes.
What is the likelihood that you’ll get bored and abandon this project in a week or a month from now?
High.
I’m sorry to be so negative here; from what i can tell what you’ve built seems quite impressive for a solo project in such a short time. Maybe I’m wrong and this is the next Khan Academy, but I am sadly not optimistic about this evolving into something which is useful for many people.
… but speaking of Khan Academy, why not try to build things with them instead of starting a new thing from scratch? (i don’t actually know much about them besides that they’ve been around a long time and have a good reputation from what i can tell)
Since all the lessons here are CC BY-SA, anyone, Khan Academy included, can use them as long as they keep them open. I’m also aiming to build this project in a way that it doesn’t really need me to stay alive. It’s a static site with no servers or accounts, so it costs almost nothing to run. If by any chance I ever stop, anyone can copy the repo and keep it going, lessons included. That’s kind of the point of making it open source.
How can it not be on a server? Does that mean github hosts it and you just own the domain? (I am reading your privacy policy.)
So it involves both Github and Cloudflare, which means really not only is an IP being logged, but also other things like packet latency, mtu, packet header characteristics, and an overall network fingerprint.
This is also the same Microsoft that helped use Microsoft Azure servers for various war activity, so this does mean data about anyone marginalized who uses this will be captured and used however… Most people don’t care about data privacy, but still it isn’t ideal.
They mean you can git clone it and use it locally, offline on your computer or on your own server for others to use over the internet (or over some non-internet offline network).
When you use it on the web, there is a server serving it (the static files) to you.
Are you sure? The privacy policy says “This site is hosted on GitHub Pages by GitHub, Inc. (USA). When you visit, GitHub logs your IP address for security purposes (Art. 6 (1) (f) GDPR: our legitimate interest in a secure, working website). The site cannot be delivered without it. GitHub keeps these logs as long as it needs them for security and publishes no fixed period. GitHub takes part in the EU-US Data Privacy Framework; see the GitHub Privacy Statement.” I think the server itself may actually be GitHub somehow?
Are you sure? […] I think the server itself may actually be GitHub somehow?
Yes, as i said:
When you use it on the web, there is a server serving it (the static files) to you.
And that server is github pages - one of microsoft’s free static web hosting services.
This is might seem inconsistent with what OP said:
It’s a static site with no servers or accounts, so it costs almost nothing to run
… but it isn’t really imo. You always need some kind of server to put something on the web for the world (using a normal web browser without installing something else) to be able to access it; what they mean by “no servers” here is that (1) there is no logic on the server, the server just serves a static web page, and (2) you can easily get the static files yourself and serve them yourself (or browse them from your filesystem without even serving them).
I like it! But I probably wouldn’t use it right now because I’m studying other things.
I did test it briefly.
My experience was that for subjects I understood already, I could do the tests and follow along easily.
Then I tried Electrochemistry & Batteries.
So a few things. First of all, I can’t easily go back and see the work I previously did, so I am doing this from memory. I could go back with a reset. I’m not sure if that’s a good user experience that I can’t go back to review previous ideas and concepts.
Oh wait, now I am looking at the course and it started over. I didn’t ask it to reset my progress. Hmm. That may be a bug?
One thing I didn’t like is that if I didn’t understand a concept, I couldn’t easily get more information on it or additional help.
For example, in this prelude to a question:
Oxidation numbers
For molecules, chemists track electrons with oxidation numbers: pretend every bond is ionic, giving the shared electrons to the more electronegative atom. Rules of thumb:
free elements: 0; simple ions: their charge, oxygen is usually −2, hydrogen +1 (with non-metals), the numbers add up to the overall charge.
An increase in oxidation number is oxidation; a decrease is reduction.
I was confused by this. I haven’t studied chemical bonds in a long time. I don’t know why I would pretend the bonds are ionic, and I don’t really have memorized which atoms are more electronegative. So I read this and sort of get it, but I’ve also studied this a long time ago…
Now for the question:
Manganese in permanganate What is the oxidation number of manganese in the permanganate ion, MnO4−MnO4−?
Okay… so we just went over oxidation numbers. I have to pretend there are ionic bonds (I don’t know what that is exactly from memory, I know it’s a type of chemical bonding). And electrons move to the more electronegative atom. Electronegative atoms wasn’t exactly explained clearly here. I don’t really know how to determine if Mn is electronegative. I see O4 has a negative next to it. Is this negatively charged Oxygen? Does that imply negative electrochemical properties?
I didn’t know, guessed, got it wrong, and then saw the answer:
Explanation
x+4(−2)=−1x+4(−2)=−1, so x=+7x=+7, manganese’s highest oxidation state. That’s why permanganate is such a powerful oxidising agent: manganese is hungry for electrons.
This doesn’t make sense to me. I know the previous screen said -2 was often a number associated metal (“oxygen is usually −2”; it also didn’t clearly define what that is. Is that the charge? Is that the oxidation number?)
After that, the next thing is:
Step 5 of 6 Try it Balance a half-equation In acid, purple permanganate is reduced to colourless Mn2+Mn2+. Balance the half-equation: oxygen first (with water), then hydrogen (with H+H+), then charge (with electrons).
I was able to figure out the right answer by trial and error, but I don’t actually conceptually understand this.
So my main feedback is that this may help people who have studied things before or want to look up things on their own, but for people who don’t understand terms or need remedial help on a concept, there’s not really a way to reinforce lessons or clarify topics.
I don’t know if every lesson is like this, but the one thing that could probably help this the most would be something like a button for “I still don’t understand” that has longer explanations, more examples, and more similar problems. I’m somewhat flexible when learning and don’t need to understand everything right away… but I need to be able to get conceptual clarification on something before trying to build on weak conceptual understandings when I’m studying a topic.
I am sure I could step out of the ecosystem and just ask an LLM why I am not understanding certain parts, but that definitely takes me out of the flow.
I am also not trying to re-learn chemistry or batteries right now. (But I’m pretty sure I never learned how lithium ion batteries worked and so if I could have followed this lesson and then learned more about new battery technology, I would have found that interesting. But I had to stop because of conceptual confusion, and I don’t want to take the time to have an LLM fill in the gaps. But for me to use this, I would probably need to be able to log in, have my progress saved, and harder modules might need more links to various terms. I did skip ahead and I know that I could have covered chemical reactions first. I did try part of that module and it felt too easy.
For me, if I wanted to re-learn chemistry right now so I could learn more about science and improve my science abilities, I would likely just have an LLM teach me directly because I think it could just be faster.
For people who might be interested in using this, for example, a teenager who wants to take some community college courses in lab work because they want to become a genetics researcher and they want to pre-learn some of the material so the course will be easier later, I just don’t know if the explanations are long enough or clarify things enough. Those explanations felt really sparse. I like the structure, I like the diagrams and pictures and those are nice. But just for anything slightly harder, there’s no real clarification when there are conceptual misunderstandings.
You could probably create links for terms (like ionic) that help people review what those things are or their definitions. And you could probably create longer deep-dives for people who get a problem wrong or don’t feel comfortable moving on and then it would work better.
Best of luck with this project!
I like all the comments debating that LLM produced education material is garbage but ignoring the fact that the “educational” examples OP provided in the three links are garbage games at best.
i mean bro, it’s a one person open source project. i’ve seen tons of errors in duolingo througout the years and they’re a million dollar company… the idea is to contribute and make the bad parts better, but the foundation and the idea of the project is really good if you ask me.
I agree, and I don’t even particularly like the foundational use of AI here. Education should be accessible to all, and it would be cool to see this become a free tool for which OP doesn’t even need AI anymore, because there’s enough contributors and experts.
This is not learning. Learning is reading and understanding verified content. What AI slops out isn’t verified.
If you tell me you’ve copied what’s on Brilliant hence “verified”, it’s even worse. It’s stealing. Do you read pirated books?
For the love of god, take this shit down.
Do you read pirated books?
yes, and you should too - limiting yourself to the books which you can afford to buy (and which are even available for you to buy) doesn’t make any sense.
o7
Do you read pirated books?
I never claimed that this is currently a finished product. My argument is that we’ll verify and improve it as a community together. On each page, I placed a button where you can suggest an improvement or report a mistake. And no, It’s NOT a copy of Brilliant. I didn’t use any of their content, and they sure don’t own maths and natural sciences.
shut the fuck up you chauvinist pig, you’re fucking whinning about piracy? go fuck yourself and your priviledged ass you scream about ai but you’re really complaining about people wanting free access to shit. go fuck yourself.
This post is a perfect embodiment of the discourse over AI happening everywhere. AI? Must be slop. AI? Just a tool bro, judge the result. Throw opensource in the mix, just needs popcorn. My two cents, the initiative is laudable, there’s not enough quality free and open source education materials out there, Gcompris for younger kids is good but gets exhausted quickly with age. Now the quality is the question mark when you use a black box like AI, and not talking about copyright that these tools piss on, ironic when producing open source… Obviously you put a lot of effort in it, it has the merit to exist vs the huge effort to create an open source community involved enough in an education project. It’s my two cents, I’m not concluding on anything.
There are textbooks you can grab on shadow libraries and recorded lectures on youtube (though exceptionally hard to find). Edtech is all crap, frankly. It’s even more crap when it’s AI-generated. The best way to learn a subject is still the way people have been doing so for hundreds of years.
Removed by mod
First off: $160 a year for education (about $13 a month) is a steal. I’m not sure why this needs to exist.
Second: ignoring the ethical issues with using Claude so glibly, who is reviewing these lessons for accuracy? Who is making new lessons? Part of Brilliant’s deal is having actual subject experts who are at least reviewing the lessons. No AI model is a replacement for that expertise. You can’t crowdsource that expertise ad hoc, either (the people who use it are the ones learning from it, after all).
Are you just going to hope Claude can clone some facsimile forever?
Impovrished people are exactly the sort of people who may need an open source educational tool to get ahead. Lots of people are impovrished.
First off: $160 a year for education (about $13 a month) is a steal. I’m not sure why this needs to exist.
First off, It being a “steal” for you is not same for everyone els
Second, not everyone around the world can pay for these. Payment processors, their monopoly and shenanigans aside, paying online is not a universal thing and we should stop treating it like it is.
I fully agree with the slop arguments.
do you think everyone lives at the expense of the global south like you? 160 is a fucking lot for a lot of people.
Notice that I didn’t say that $160 wasn’t a lot for people. I said it was a steal for the amount of educational content.
Removed by mod
This software isn’t free. It has a cost. Did you think using the tools provided by billionaires was a gift from them?
Also, it’s hilarious that you think I’m bullying OP. Where in my original comment did I say, “No, don’t use this”? I intentionally skipped over my ethical concerns, because OP is likely going to use it anyway.
They’re providing educational content. You can’t just replace the knowledge of experts creating and checking lesson material with AI and hope for the best; that’s how you end up teaching people wrong information.
Y’all pro-AI goobers are so lost in the sauce, you can’t even take a critique.
Fair questions. for you $13/month is fine, but I personally remember that I didn’t wanna spend this much money years ago when I needed it.
on accuracy: right now it’s me going through lessons plus anyone who hits the flag button, and every fix is public on github. I’m not going to pretend that’s the same as a team of experts, it’s honestly the weak spot and why I posted here. if you know a subject well and want to check a course, I’d really appreciate it. Together we can make this much better that Brilliant :)
$13/mo can be a lot for people, that’s true. There’s certainly a subset where they simply don’t have that extra. I suspect that subset is fewer than people might imagine, but I honestly would rather people have accessible education than not.
For the second part, I think it’s a hard ask for experts to contribute in that way, but hey, people spend their hobby time working on Linux and other OSS, so who knows?
I also have ethical issues with the use of Claude itself (as well as other tools/platforms provided by billionaires), but I think that’s beyond the scope of your original questions.
There are other ML models made from scratch by universities and experts which aren’t beholden to shareholders (and might not be accessible by the public, yet), so if you are determined to continue using AI in this way, keep an eye out for potentially better alternatives.
Removed by mod
Your AI cannot beat my source of information; there is literally all information that has ever and will ever exist. No human can create anything nearly as advanced as my source of information
My source of information is picking a random book from the library of babel
Removed by mod
There’s petabytes of data there dude
The fact that you think that AI has a database of information from which it pulls demonstrates that you have no idea how it works. And you think we’re the blind ones lmao
As a PhD candidate, you are a doofus. An LLM cannot answer any question regarding my field with any degree of accuracy. Believing that LLMs are PhD level already speaks more to your intellect than LLMs.
Removed by mod
Get a refund
As a PhD year 2 I assure you they work, you freaking removed. I study computer science btw.
A PhD second year in October is someone who took one year of courses quite similar to undergraduate work, lol. As a candidate, someone who has actually done work (prelims, language work, dissertation writing, fieldwork, etc.) beyond taking classes, this is nonsense.
I assure you they work, you freaking removed.
This statement does not make any sense and further proves my first point.
I study computer science btw.
Ahh, that explains it.
Removed by mod
160 for some ai content? It’s not even cheap for humanly curated stuff.
I thought it was weird there wasn’t a negative Nancy(s) in a thread that has any mention of AI. It’s now become a disease to hate on anything AI, even if it was good.
It’s weird that people are so positive towards AI, even when it’s used to generate something with no understanding of the material. It’s now become a disease to be accepting of anything AI, even if it uses the very thing our oppressors want us to use.
😂 the same tired and boring Reddit cliche. Most of you who “hate” it use it religiously, but sitting out here virtue signaling. Gtfo.
Ah, right. Because I secretly love it and use it all the time, as evidenced by your very scientific source, “Trust me, bro.” What a tired and boring Reddit cliché.
It’s ok. We still love you, even if you use AI.
Removed by mod
😂 I want to see that
Removed by mod
They’re wide open 👀
If ummm akshually was a comment.
I have a request: make them multilingual. A lot of people who would benefit from this sort of thing don’t speak English and as a native speaker of a small language, I am unhappy.
Aside from that, great work! A few of the excercises I tried seemed to assume I know a lot of the background already, like what is node, edge, vertex etc.
good points both. I’ve implemented this system where anyone can suggest an improvement and it opens up an issue on github. So feel free to report any improvements you can think of.
translations are on my list, and it sure can be done with ai agents. But the quality and tone have to still be checked by someone.
Interesting, an MIT license project
Yes that’s another ick, GPL would make me consider collaborating.
The code is just the player and not the most important aspect of this project. the actual lessons (the part that matters) are CC BY-SA 4.0, which is basically GPL for content. anyone who reuses or changes them has to keep them open under the same license and credit the project. same license wikipedia uses.
I like the cc by sa but, with an MIT license you can yoink it, paywall it, and never contribute back
Removed by mod
Removed by mod
Honestly I like it. The few examples I tried worked perfectly fine even on mobile. Can I vouch for the quality and veracity of each and every play? No of course not but I can bear this caveat in mind while not parting with a ridiculous £30 a month or £216 if you want to stump up the money upfront. Fuck wanky subscription services in general, I for one am glad you came up with this and hope you will continue improving it. Well done you!
thanks a lot, really glad it worked well for u! if you run into anything weird or want to contribute feel free to open up a PR
Removed by mod
Thank you very much for making this. It is a dream come true for homeschooling since the cost of supplemental resources add up quick!
that’s so nice to hear, homeschooling was one of the things I had in mind.
thanks for this op.
on a side note to those who pass by; this aggressive anti AI sentiment might look like a plague, but there is something good that can come out of it. I am blocking every one of them, without responding to them or acknowledging their drivel.
life is wonderful 😍
Removed by mod
Thanks for coming out. Now I’ve blocked you. You’re welcomed to block me in return.
May we never live in the same universe again.
Removed by mod
