Next-Generation Spaced Repetition Software
Donation protected
Spaced repetition software (SRS) is used by tens of thousands of people all over the world to study foreign languages, the natural sciences, gruelling medical-school curricula, computer science, math, history, general trivia. Study after study has shown that it's one of the most effective ways to study per hour spent, period, and some of its results have been downright extraordinary: this Reddit user was able to get his five-year-old daughter up to a middle-school reading level using the SRS platform Anki, which was also a preferred tool of the top Jeopardy! contestant Roger Craig. I've used it to great success myself with subjects ranging from chemistry to ancient Greek.
For all its effectiveness, however, SRS usage is mostly confined to a handful of autodidacts (and medical students), and hasn't broken out into wider educational usage. After many years of personal SRS usage and a couple of years teaching middle-school Latin (a subject to which SRS is, or should be, very strongly suited), I think I've figured out why:
- Existing SRS platforms such as Anki and Mnemosyne are old (often approaching two decades in age), and come in the form of downloadable, desktop-side software—a poor fit for the age of Chromebooks.
- Anki, the most powerful platform, is highly customizable—if you're a power user who's not afraid of HTML and Javascript. Custom or conditional formatting is out of reach for busy teachers and students who want things to Just Work. Even then, there is a great deal of low-hanging fruit in the user experience that can be improved. It should be possible to color-code for gender or tone at a button click, or have Latin-alphabet input automatically convert to languages such as Greek or Russian, instead of wasting time wrangling keyboard layouts or cutting-and-pasting from online keyboards. Diagrams must be made in third-party software such as Figma and imported manually.
- The mind processes information in a hierarchy of chunks and prerequisites: we can memorize something new by rote, but need to be able to link it to other concepts to make any use of it. It's not memorization that's the enemy, but memorizing orphaned knowledge with no existing connections. No SRS platform on the market today allows for any but the most primitive card relationships, even though they're not hard to implement and make for a much closer fit to how the mind actually processes knowledge. A next-generation SRS platform should make sure the user knows what fliegen means before testing them on its past tense flog, or what a carboxylic acid group looks like and does before testing them on the full structure of an amino acid. But no major SRS platforms run these checks, and the result is friction and frustration that detracts from learning rather than adding to it.
- There is a consensus among SRS power users that good card- and template-design is a sine qua non for effective review, but existing platforms make these processes out of reach for casual users. A forty-five-year-old Spanish teacher should be able to create good card types with a few button clicks rather than fiddling with Javascript, and save time by manufacturing lots of cards at once within the the platform rather than playing around with CSV imports from Excel.
- Existing SRS platforms are built for autodidacts, not schoolchildren. Teachers can't easily create cards for students, check to see whether or not they've done reviews, implement a "quiz mode" to make sure they're doing their reviews honestly, or see what their students are having trouble with. Online, classroom-based flashcard platforms such as Quizlet are built for cramming a unit's content knowledge before the text, not long-term review. While a few classroom-oriented platforms have implemented spaced repetition, it is generally an afterthought, and the power and flexibility of real SRS software is lacking.
All of these deficiencies are surmountable, sometimes very easily so (a what-you-see-is-what-you-get card designer, for example, takes a few hours to code up). But nobody's yet thought to put the pieces together into a single platform, until now.
My name is Campbell Nilsen—nom de plume Nephew Jonathan—and I'm raising funds to build a browser-based SRS platform that's equally suited to the power-user autodidact and the high-school Spanish classroom, based on my experience as a Latin teacher and heavy SRS user. After writing an in-depth whitepaper on the nitty-gritty of the platform and talking with a few VCs, it became clear to me that this was a project better-suited to crowdfunding (at least early on), for a few reasons:
- The real prize is finally bringing SRS into classrooms across the country—but there are tens of thousands of autodidacts who'll be interested in using a more convenient platform as well.
- I need time to work on this, which requires living expenses (before working on this project I was working two jobs and commuting two and a half hours a day), but the needed amount in question is far too small to interest VCs. I'll also want to hire a cybersecurity specialist and graphic designer before launch.
- I believe a next-generation SRS platform is something users and schools are willing to pay for.
What's this platform going to be called?
That's a very good question! It doesn't have a name yet, though its working title is 'Mneme' (Greek for memory). As that's a bit of a mouthful, I'm soliciting a new name. Email me or DM me on Twitter if you have a good idea.
When do you anticipate launching?
I am hoping to have a viable, usable prototype by the end of this coming summer, though it probably won't have all the planned features.
Are you looking for cofounders?
Yes! Email or DM me on Twitter if you're an SRS nut who's got experience in webdev, graphic design or sales and are willing to commit for the long haul.
How far along are you right now?
I'm currently working on the template-generator. At the moment it's a bit sparse and needs some aesthetic touching-up, but I've already got automatic peer-relationships and dependency/prerequisite relationships, a WYSIWYG card editor and markdown formatting.
But to get this off the ground quickly, I need to not be working two jobs--that's why I'm here.
How much of a walled garden will this platform be?
I don't want it to be a walled garden at all--I've lost decks before and it's awful. If you want to import cards from a CSV file from Anki or Mnemosyne--feel free.
Do you have a more in-depth overview of what this should look like and why it's necessary?
Yes! See the white paper.
I have an idea for a feature.
Please get in touch!
Disclaimer
No raffles, sweepstakes, giveaways, or returns on investment are offered in exchange for any donations made to this GoFundMe.
Organizer
Campbell Nilsen
Organizer
Ashburn, VA
