I'm launching UneeBee, an open-source tool for creating interactive courses like Duolingo:<p>GitHub repo: <a href="https://github.com/zoonk/uneebee">https://github.com/zoonk/uneebee</a>
Demo: <a href="https://app.uneebee.com/" rel="nofollow noreferrer">https://app.uneebee.com/</a><p>It's pretty early-stage, so there's a lot of things to improve. Everything on this project is going to be public, so you can check the roadmap on GitHub too: <a href="https://github.com/orgs/zoonk/projects/11">https://github.com/orgs/zoonk/projects/11</a><p>I'm creating this project because I love Duolingo and I wanted the same kind of experience to learn other things as well.<p>But I think this could be useful to other people too. I'll soon launch three products using UneeBee:<p>- Wikaro: Focused on enterprise. It allow companies to have their own white-label Duolingo. I think this is going to be great for onboarding and internal training.<p>- Educasso: Focused on schools. It will allow teachers to easily create interactive lessons, compliant to local school curriculum. I want to make it in a way that saves teacher's time, so they focus more on their students rather than lesson planning.<p>- Wisek: Marketplace for interactive courses where creators will be able to earn money creating those courses.<p>I'm not sure this is going to work out but, worst case scenario, I'll have products that I can use myself because I'm a terrible learner using traditional ways. Interactive learning is super useful to me, so I hope it will be to other people too.<p>If you have some spare time, please give me your brutal feedback. I really want to improve this product, so no need to be nice - just let me know your thoughts. :)<p>PS. I'm also launching it on Product Hunt: <a href="https://www.producthunt.com/posts/uneebee" rel="nofollow noreferrer">https://www.producthunt.com/posts/uneebee</a>
This is awesome! I had the same idea in high school and wrote a whitepaper about it.<p>Personally, I think Duolingo shouldn't exist in its current form. The courses are all created by teams of volunteers and only able to be contributed to by those volunteers, which slow downs course creation and bug fixes to courses. Additionally the data is all held hostage by Duolingo, which prevents you from integrating it with Anki for example. I don't think Duolingo should be getting revenue when most of the work is done by volunteers. I think a decentralized open source version should exist, much like how anyone can modify and upload Anki decks, people should be able to modify and upload "Duolingo-like" courses.<p>Re: marketplace:<p>I think this is a great idea! There are Anki decks that are sold by creators too, like Spoonfed Chinese. I think this is an actually fair way to implement learning like this.
What are you offering that's different from already existing solutions like Moodle or Opigno[1]?<p>For example with Opigno you can already build an open-source based education platform with content authoring with H5P[2] (HTML-5 based interactive content)<p>I've used Opigno LMS in the past, it's Drupal based so it's easy to deploy/maintain and basically have all features you advertise.
Schools especially may use a learning platform for 10+ years, at least when authoring with H5P they are not stuck in a single platform anymore. What's your solution for this problem?<p>The loss of flash left a lot of courses or interactive learning content unusable, that's a mistake I hope new platforms don't repeat.<p>[1] <a href="https://www.opigno.org/" rel="nofollow noreferrer">https://www.opigno.org/</a><p>[2] <a href="https://h5p.org/" rel="nofollow noreferrer">https://h5p.org/</a>
Notes from for LitNerd (YC S21) re: IPA, "Duolingo's language notes all on one page", Sozo's vowel and consonant videos, Captionpop synced YouTube videos with subtitles in multiple languages,: <a href="https://news.ycombinator.com/item?id=28309645">https://news.ycombinator.com/item?id=28309645</a><p><i>Spaced repetition</i> and <i>Active recall testing</i> like or with Mnemosyne or Anki probably boost language retention like they increase flashcard recall: <a href="https://en.wikipedia.org/wiki/Anki_(software)" rel="nofollow noreferrer">https://en.wikipedia.org/wiki/Anki_(software)</a><p>ENH: Generate Anki decks with {IPA symbols, Greek letters w/ LaTeX for math and science, Nonregional (Midland American) English, }<p>Google translate has IPA for some languages.<p>"The English Pronunciation /
International Phonetic Alphabet Anki Deck" <a href="https://www.towerofbabelfish.com/ipa-anki-deck/" rel="nofollow noreferrer">https://www.towerofbabelfish.com/ipa-anki-deck/</a><p>"IPA Spanish & English Vowels & Consonants" <a href="https://ankiweb.net/shared/info/3170059448" rel="nofollow noreferrer">https://ankiweb.net/shared/info/3170059448</a>
Good luck. Since you want brutal feedback, at this early stage, it looks boring and tedious to enter data. I’d concentrate on the course output and define the input format as JSON or CSV. Why? Duolingo isn’t great because of the content, but because of great navigation (and points gamification) so the courses should work great on small touch screens. How to get the data is is something to worry about once the interface for the learner is good. Also, it’s easier to get formatted course input, because you can just ask chatgpt to generate learning material in a specific format. (For my own out of Duolingo experience I do exactly that: I defined a learning csv format and ask chatgpt to give me learning examples in this format) - also I’d stay away from animating images unless you get good at it, because animations distract from learning (and duolingo is turning into an animation studio and spending the resources to get animation right, so anything you can do with low budget is going to look cheap)
That's a nice looking app! I have a similar mindset about duolingo. My interest is language learning and I've developed my own duolingo clone, but perhaps not as fully featured as yours <a href="https://codinginthecold.alwaysdata.net/kuku/" rel="nofollow noreferrer">https://codinginthecold.alwaysdata.net/kuku/</a><p>I've also just finished an early version of an app that facilitates creating "parallel texts".
How does this project compare to LibreLingo ( <a href="https://github.com/LibreLingo/LibreLingo">https://github.com/LibreLingo/LibreLingo</a> )? Why would I choose UnueBee over LibreLingo?
This looks like a great idea with a lot of hard work ahead of you.<p>You mention a few different areas it could work in (education, enterprise, consumers), but really you need to pick one and really focus on it - you will dissipate your focus and energies if you try to serve even two of these markets at the same time. Might be worth picking something far away from languages since Duo and other apps have that sewn up already. It would be hard to compete.<p>I like the education/school idea but you'd have to make the product free and focus on curricula and really serving those well. Perhaps target it at teachers and make it easier for them to construct lessons.<p>Need to make sure it targets the device that your markets will be using, which probably means phones rather than desktop or ipad.<p>Feedback:
1. Password requirements are a little annoying, perhaps loosen those a bit.
2. First examples are way too broad - pick an area, focus and try to execute just for that area, so that people see immediately what it is for (e.g. make your lessons as a teacher) and you have a target customer you can go after.
3. Your examples need a <i>lot</i> more work and polish, I did the launch to moon base alpha - the pictures don't add anything IMO, and I didn't feel like I really learned much because the questions were too fuzzy and the answers were not informative, you need to build in a better feedback loop and probably pick examples which are <i>far</i> more clearcut.
4. There is some very weird/wrong content IMO e.g. use a 'gravitational slingshot' instead of burning fuel - the vast majority of spacecraft burn fuel to change course, including the rare cases where they use a gravitational slingshot.
5. Content is king - I get this is supposed to be some quick examples you threw together but you need to step back and consider what impression this gives of your product - it makes the whole thing seems slipshod and ill-thought out. You need some real, very well thought out content here to make it seem worthwhile. The user should feel like they learning something in your examples.<p>So it looks like you have the basics here for your goal, but you need to focus on content and make sure you provide good feedback when the user gives a wrong answer as this is the biggest point of friction when learning. Also consider how someone like a teacher would give this to kids and how the kids would react to it - I'd recommend trying to connect with a teacher and get their honest feedback, that sounds like a promising direction even if initially they'll tell you it'll never work, they might give you some good pointers.
This looks great. I am involved in a crypto community and could see using this to help onboard users to our ecosystem. Get them familiar with how to use the various applications and learn about the technology behind the chain.
if you're targeting enterprise clients, they may insist that your course fit in to their existing LMS (learning management system). A bunch of large corporations use these old and stodgy technologies and require learning materials to support the ancient SCORM API or the slightly less ancient xAPI API. You might find, if you go that route that you spend more time dealing with the complications of improperly implemented APIs, poor developer documentation for integration into those systems, organizations who don't really have people with the technical knowhow to help with the integration issues, Subject Matter Experts who aren't actually experts at anything, etc., and less time developing your own technology.
I know it’s not done to comment on the page design. But here it is anyways: the page is nonscrollable on my phone and the images are much wider than my screen. So I cannot see the screenshots. Even when I rotate to landscape, I can only see a part of the image.
Looks really cool!<p>Just a quick note - The 'Live Demo' probably shouldn't be locked behind a login unless there's a good reason for it.<p>More broadly, do you plan to have general free courses? i.e. would I go to your site to hunt around for something interesting to learn, or are you always behind the scenes, and some third party is advertising their course to me (which uses this as a backend)? Just wondering about the marketplace; how/why I would end up on the site in the first place.
Duolingo is cool, but I really think that having students produce something other than answers to multiple-choice questions would lead to higher quality learning results. It should be much more possible now especially, with all of the AI advances.
Re Educasso. what kind of schools are you thinking of public private charter? Also do teachers create their own content? I thought the content always came from a tie up of the school to a publisher that supplies all of it. It's hard to supplant publishers in grade school IIUC.
" your brutal feedback. "<p>I did not like Duolinguo, it was too playful for me.<p>I prefer <a href="https://clozemaster.com/" rel="nofollow noreferrer">https://clozemaster.com/</a>
Cool idea.<p>Would you target "first-time" creators, or experienced ones?<p>And if experienced, why would they take a bet on Uneebee vs. existing platforms?
As a founder i can give you one advice you did not ask for:<p>Stick to one market and one solution. do Enterprise, do Schools. Don´t do both. It takes an insane amount of capacity and resources to penetrate a market, and it really does not matter how much you have - trying to be in two places at the same time will fail. If you care for both, start with one and get it done.<p>Also: i strongly suggest to stay away from a marketsplace. Pulling off a working marketplace is unicorn-level hard. Spend a day to research how many successful marketplaces exist and then another day to find for every successful marketplace the failed ones - getting a startup right might be 1:100 chances, getting a marketplace right is more like 1:10.000.