In addition to the challenges listed here, IMO there are rapidly diminishing returns for the type of recall learning that spaced repetition enables. As you progress further in your career, there is much less emphasis on what you know, and more emphasis on how you apply it, how you communicate, and how your knowledge ends up helping others around you. I suspect most professionals decide at some point that they need to start "paging out" specific knowledge to make room for broader experience, retrieving it from the bookshelf (swap partition) when needed.
I'm also curious on the fixation with creating a startup in the VC-funded sense. Why choose able-to-find-VC-funding to be your metric of success?
But for languages, SRS is great.
And I'm also glad I memorised a whole bunch of math formulas way back. E.g. Boolean algebra I keep using an identity that I couldn't find on identity sheets by web search.
With Duolingo it kind of encourage small bursts rather than the hour (minimum) per day you need with Anki for better or worse.
I understand the educational value of what you made. You clearly overcame many technical hurdles to combine the platform with solid educational principles. Awesome stuff!
But another major step needed was to know that a great product, marketed poorly, is likely doomed. Ironically, a lesser product marketed well, will likely succeed- it can be improved as it gains attention (and revenue).
Sorry it didn’t work out. Maybe one day your personal runway will expand for it and you can explore how to put it in front of an audience.
Once you get an audience, and you know how to build that audience, you’ll get user feedback, adoption traction and can explore new audiences to share it with.
We’re in an age where they won’t come to you. You must go to them (with marketing, brand, building an audience).
Mentioned Duolingo has spaced repetition. It is not a flashcards app.
For me as an adult who did learned stuff, both the supposed frictions and necessary components of learning as described in the article does not ring true. It is not really consistent with how I perceived anki/duolingo nor how I understand the reasons for "create own cards" advice. It is not even consistent with where my learning successes and failures came from.
Maybe I am simply not the target of this startup then. But also, I do not think I am the only snowflake adult for whom these points did not ring true.
With traditional Q/A-style spaced repetition, I feel like accumulating a long list of isolated facts sometimes (I know, you can remedy this a bit by also quizzing connections, context, but I feel like the general tendency still remains).
The part of seeing "You still need to decide what’s worth remembering" as friction I strongly disagree with.
That is a very important part of the learning process and IMO should not be automated. It is difficult because learning is difficult, but if you kill that you also kill the spirit of self-learning!
And not mentioned in the article, but I think the most important factor is that everyone in the SRS niche knows Anki, and despite Anki faults, everyone has put the effort to learn how to use it, got used to it, found a good-enough workflow, and have zero incentive to move to a more expensive alternative even if it is better.
> I find the common advice of “write your own cards” misplaced: you pay the opportunity cost of not reviewing all the cards you don’t make.
This perspective of cost is a nice way of putting it. The other side to me is that writing your own cards can, but does not necessarily, improve retention on its own. Much of card creation is still mostly manual formatting, copy-pasting, etc.
My own solution to this has been to build an 80-20 version of what was demo'd in the "AI canvas" part of the video. I drop in a source (pdf, epub, etc), and an LLM prompt creates cards. I just use Anki Connect to directly add these cards to my collection.
Would be interested in seeing the Rember system prompt for card creation, and hearing what LLM is being used. I've seen wildly different results from changing my own prompt (and from using different models).
rickcarlino•4mo ago
In college, spaced repetition helped me get ahead, but it felt nearly impossible to convince my peers of this. While I was able to effortlessly remember thousands of vocabulary words in my language program, my peers struggled and failed to heed the simple advice of "just put it into Anki". Spaced repetition is valuable, but much like having a good diet or exercising regularly, it is really really hard to convince people of the benefits. A spaced repetition SaaS can easily become akin to an unused (and eventually cancelled) gym membership.
It seems that most of the apps that make it in this space ultimately need to compromise on their fundamentals in order to get traction. I don't need to name any names here, there are plenty of examples of SRS apps that started as solid tools for learning but evolved down a path of gamification to a point of diminished usefulness.
Anyway, sorry again to hear about this one and best of luck to you on whatever you decide to build next. Like you, I am continuing work on my (noncommercial) SRS and I always enjoy hearing your perspectives. Feel free to reach out if you ever want to bounce ideas around.
LittleFishyChan•4mo ago