Maybe you don’t know what many people (like me) use Memrise for. Or maybe I’m in the minority, though judging from the birdsong and geography courses that people do (I don’t), this minority is probably pretty active, not to mention more likely to pay than an average kid boasting about a three-day streak on Duolingo. You want to teach us languages in the best way you can think of. This is absolutely not what I want from Memrise.
There are tons of language apps out there. Duolingo, Rosetta Stone, Pimsleur, Glossika, etc., are huge and by sheer size difficult to compete with regardless of how brilliant your original content. There’s also a huge selection of apps and learning programs for specific languages users can choose from. For Chinese, I’ve done Duolingo, HSK Standard books, the Heisig books for Hanzi, and half a dozen of different apps, big and small. But here’s the point: Every single one of those courses I combined with Memrise for learning the relevant words.
My approach that I keep recommending right and left has been, learn the language on any app you like, but learn the words for that app on Memrise. (Check out my post on how to study Chinese with Duolingo; I talk about Memrise about as much as about Duolingo.) Whatever the language app, it’s never very good for learning the vocabulary, because its goals are different. Memrise works as a great flash-card app, for two reasons. First, there are user-generated courses for almost every occasion. I’ve learned a bunch of HSK word lists created by one BenWhately, but I’ve also used the lists for HSK Standard books and the Heisig books, and word lists for the different Duolingo courses I’ve been doing (struggling when there wasn’t a good list on Memrise), and created my own private lists for the apps and missing words that I needed. There’s absolutely no way sourcing the words from any course you might design the way you’re planning to would match my needs.
Second, the mems. For courses with tons of user-generated mems, there are often great ones there, and there would have been many more had you guys not been so unfriendly to them. Some mems help bind the parts together and some provide background information, and some I make for myself because I know what helps me learn best. No matter how smart you are in creating generic mems, different people need different things to learn, and there’s no way you can make them better than what the collective wisdom can do given proper tools. The Heisig books suggest stories to bind character components together, and there’s a free site that collects user-generated stories to choose from, which work like text-only mems. But I’ve learned the Heisig lists on Memrise because I learn better with images instead of stories. And by the way, I go through the list first with a click-only course, then with a typing course, and then with a typing-in-Chinese course. (Goodbye typing, hello clicking galore!)
In short, I don’t think you can (or should want to) compete with language-learning platforms by offering presumably superior content; the field is too crowded already and different people have different needs. But Memrise has been my tool of choice for learning the words from other courses - every single one of them. Outcompeting Anki and Quizlet isn’t nearly as tough, particularly given your ideas about rich content for every work drawn from a central database. What I don’t like is creating the word lists based on your course content. How would you ever hope to replace the way I learn with your own (no doubt brilliant) language study tool?
People who want to study a language have a lot of choices, and most of those people don’t stay learning for very long. I would suggest targeting a more mature audience with a superior flash-card app, so that instead of saying “combine this with Anki” people would recommend “definitely combine this with Memrise”. I know this doesn’t look as flashy or ambitious, but there’s a real need for a product like that, and my guess is I’m not alone in that I’ve been using Memrise to fill this need. Fix the problems with what you have. If after you’ve done that you still have resources, implement your ideas (which do sound promising) about how to make the flashcards better. Don’t replace what make Memrise unique with something for which the demand is questionable.
The very least the team could do is make a mass export tool. That’s an extra slap to the face to the loyal user base. I expected a CSV export or something. With only a month of notice, there’s no time for anyone to try and make their own script to do this, either. Some people have hundreds of mems, do you expect them to save them all manually? And those of other users?
