Hacker News new | ask | show | jobs
by tempaccount1234 945 days ago
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)
3 comments

> 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.

Counterpoint: twee gamification getting in the way of learning is why I stopped using Duolingo after reaching a 2500 day streak — the last straw was the tree turning into a path at the end of last year, but the last time I really enjoyed it was back when I could load each lesson into a separate tab and just spend hours at a time in the zone without getting distracted by their attempts to cheer me on.

Similar direction of changes are now actively annoying me with Brilliant.org

Same, that plus the persistent harassment. Personally, I'm skeptical that Duo in it's current format is designed to teach. The goal seems to be giving the impression you're learning so you but into micro transactions. I've seen people with multi year steaks who struggle with the target language and clearly would have been better served via another approach.

Maybe it's the course design, maybe the lack of supporting material, or some other weakness in the platform itself. But, I left out of frustration and started building Anki decks so I could have more control. As I'm now on my own, I've had to find supporting materials, which highlighted how poor Duo's explanations can be.

That last part comes down to the course design. Hopefully OP's platform doesn't force designing ineffective courses. This isn't my area of expertise, but I think it's important to recognize that adult learners can have different goals, even within something like learning a language (e.g. speaking, writing, accent, etc.)

Just a tangential shower thought on gamification. I found myself skipping Duo on weeknights so I could cram on the weekend, get matched with procrastinators, and ascend the leaderboard easier.

I've heard other reports of people focusing on the game and not their goals. Game designers are aware that humans will "optimize the fun away." In the case of Duo, it seems a group of users will optimize the learning away (regardless of the goals or intention)

I unironically believe that Duolingo is a pacifier that makes you feel good. It's goal is to maximize engagement, not help you learn a language. This is coming from someone who has spent a considerable amount of time on the platform, maxing two different language trees and am currently unable to speak a lick of either of those languages.
Yes and no. It is a great pacifier, but my grandma of nearly 90 years old also legitimately understands any english that comes her way much better after doing that on duolingo for a few months. She enjoys it greatly, I do not understand how with all the advertisements and ten types of currencies/points/lives that I cannot make heads or tails of but she likes it.

Especially the animations, she often remarks how cute they are and wants to show us. It's adorable and I'm super happy for her :) though also disappointed that, when I installed duolingo for her, it wasn't the innocent learning app with good UX that I knew it as from like six years earlier

It being a pacifier and it enabling some (read: non-zero) understanding of a language are not mutually exclusive. You exemplified my thesis.
Your "thesis" rather sounded like it was exclusive since you said you didn't learn one "lick" after maxing out the available content for two languages, but I'm glad we apparently agree after all
I believe the gamification is a patch for the courses themselves not being engaging enough, if you're intrinsically motivated to learn a language the gamification part can be really off putting. Me and most people I met on my journey of learning Chinese have used Duolingo at some point, but most people seem to feel it was not a great investment of their time. Anki on the other hand has zero gamification and looks really boring, but somehow has better long term retention with serious learners. But then it has its own set of problems.

I'm currently trying to get the best of both worlds by building the app I wish I had while starting learning Chinese. Currently actively looking for partners as well. If anyone that groks language learning is interested in building such a thing do send me an email at sasja.ws@gmail.com and lets talk about it.

I actually found I like path better then old tree. The big difference is that with path, I do the next dot on the path or dont. With tree, I was tempted to constantly optimize the xp. Do I want 2x now or later? Should I do revision or new content? Should I continue this node or that node? When was the last time I did the boring node I am avoiding?

With path, there is no obsessing over strategy, pop up app do the lesson and that is it. I also found that I retain more, because revision is build into path instead of me having to manage it.

That worsening trend in consumer products was very well analyzed in this post that appeared somewhat recently on HN:

https://nothinghuman.substack.com/p/the-tyranny-of-the-margi...

I observe the same. I have 700 streak, have completed two full courses.

At the same time - I think gamification and nice packaging are the sole reasons Duo is a success. To be precise the gamification is mostly not about competing with the others, but being held hostage for your own streak.

and $$$$$ on marketing
You’re saying that gamification is bad, and yet you hit a 2500 day streak. I would say, for what it’s worth, the gamification effects worked.
Twee gamification.

At the start of that (2016, and indeed before as the streak didn't begin with me signing up), the gamification wasn't actively getting in the way. At some point in or around late 2019 or early 2020, it became faster to quit the app and restart it than to wait for the end of lesson congratulations screen to finish, I felt insulted by the childish animations (and that the only way to switch them off on iOS was a system-wide accessibility feature which merely reduced them!), and when the transition from tree to path happened I had to motivate myself with the goal of the round number streak length just to get past what had become for me an actively demotivating experience.

I moved to Berlin in late 2018 with the intention of it being permanent, so German skills aren't optional for me.

Very good points, thank you! Yeah, I played a little bit with animations but I didn't manage to make them right. I think I kept one when the lesson is completed but I recognize it's really bad. I think I'm going to remove it.

Totally agree I need to focus more on the course output and overall learner experience.

My thoughts exactly.

I watched the video but all I saw was a lengthy click-process. The key feature is usage, not creation.

(Sidenote: I agree - Most future tools will profit from a having a JSON-like DSL that is amenable to be fed by content coming from LLMs)