hckrnws

This is so cool and something I’ve wanted for a long time, but it isn’t quite right yet (for what I personally want in an app like this.)

I am your target market, and I’d buy the lifetime /annual sub in a second if it had these features:

I want control of the SR sequence or, I want to know what SR algo you are using and know it is best practice model. The landing page says 4 sends, but that isn’t true SR.

The next thing, I’d want to see all of my “cards” or information pieces when I am logged in, so I can see and edit and delete and keep the database clean with a total view of content. The next thing I’d need (maybe you have this?) is for the email to effectively be a flash card, where the email content is the front and a link the email takes me to the “back” of the card so I can’t use cloze delete and other techniques. The last thing is bulk upload of content via a csv so I can bulk import mochi /anki / llm generated content.

I wish you luck with this and would (selfishly) encourage you to not ship so many different things, and instead encourage you to pick one and make it best in class for niche users like me who would spend and spend on premium solutions, but won’t spend on superficial implementations.

Can't thank you enough for the feedback. These are all great ideas which I'll look into.

I was wondering what is your experience with Anki? Are there reasons you are looking into alternatives or do you just like the idea of getting stuff by email? Thanks again!

On the topic of SR specifically, what does it mean to be best in class? I am just a layperson who uses a SR tool or two and has a passing interest in the topic, but the Wikipedia article on SR gives the impression that there are a variety of algorithms out there and none are established by research to be definitively better than the others, in fact the Criticism section mentions a study which found that absolute spacing was just as good. https://en.wikipedia.org/wiki/Spaced_repetition#Research_and...

The OP said he's using Ebbinghaus' "forgetting curve" which is not exactly a SR algorithm but something similar, there is an actual formula associated with it - https://en.wikipedia.org/wiki/Forgetting_curve

Take a look at FSRS.

The paper: https://dl.acm.org/doi/10.1145/3534678.3539081?cid=996605471...

Various implementations: https://github.com/open-spaced-repetition/awesome-fsrs

Some benchmarks of various srs algorithms: https://expertium.github.io/Benchmark.html

Hey not to distract from OPs post, but I wanted to get into SR for a while, and your "what is your algo" question aligns with my line of thinking when approaching new methodology.

Could you please share your SR tech stack? Are there good apps etc., that can make the process a bit easier? The pen-and-paper approach I used in uni for learning kanji / new languages has scarred me somewhat, so I am eager to try something tech-heavy.

Hey, happy to help here.

During my research, I found this Reddit post which was in lots of help, especially for beginners.

https://www.reddit.com/r/Anki/comments/17u01ge/spaced_repeti...

They expand their work on their GitHub as well.

https://github.com/open-spaced-repetition/fsrs4anki/wiki/Spa...

Congrats on the launch! I second the request for lifetime pricing option.

Also, the jump between $4/mo to $59/mo could use more explanation to justify the price gap.

It's worth considering listing GinkoNotes on AppSumo if you decide to offer a lifetime deal. I think it would do extremely well.

Hey, there is a lifetime option and it's just the higher number you mentioned.

I suppose it's not immediately obvious that it's a one-off purchase, so I'll be working on the design aspect of this.

Thanks for your interest and suggestions!

Why email at this point and not just Anki?

Not the person you asked - but for me it is that I check emails compulsively any other app I have to remember/tend to forget. Like Todoist is great but I can go weeks without opening it and I skip tasks because of it unless I set email notifications.

Hmm. SRS is a daily commitment though. If you skip it the reviews just pile up, which tends to be painful enough that you quickly learn it's a daily commitment. I couldn't imagine getting 150 emails a day to review stuff, or that being a workable UX.

I think 10-20 flashcards a day is better than 0. But on the other hand it might not be rate to learn something quickly enough.

With the original Anki algorithm the rule of thumb is your daily review load will be 5x your rate of new cards per day. So 10 to 20 reviews per day is between 2 to 4 new cards per day. Definitely not enough for language learning, which is my use-case, but may be ok for certain other things.

> 2 to 4 new cards per day. Definitely not enough for language learning

What is your expected rate of acquisition for a new language?

40*365 ~ 14 000 new cards a year - that's a ten thousand word vocabulary with four thousand cards to spare for grammar, idioms etc?

When I learned Japanese around 15 years ago I did 10k in one year. Was a great pace. Still speak it to this day, so it worked like a charm.

I try to target 35 a day, and expect to hit that at least 80% of the time.

[dead]

This is an interesting approach. Congrats on shipping!

I worry that it won't be effective. Systems like Supermemo and Anki work because of:

1) Spaced repetition (showing you the thing at the right time).

2) Retrieval practice (having your brain practice retrieving the thing you're about to forget).

3) Feedback and automation (using your self-rating to schedule the next review).

You are doing #1 and #3.

But you totally skip #2, because you show all the info in the email. So, unlike Anki, Supermemo or Quantum Country (which someone mentioned in another comment), there's no front/back or cloze deletion, and no retrieval practice happening.

Perhaps putting the question in the email subject and the answer in the email body would work?

Thanks for the feedback!

I agree, and this is something I thought about a lot during the design process. In my experience, just looking at the note (#1 in your example) helps a lot more than no repetitions (which is obvious of course), but it's still a huge improvement compared to my previous flow.

As for repetition, I was thinking of replying to the email with what you think is the answer and letting an LLM decide if you remember correctly. Is that something that sounds effective to you?

Just create a “show answer” link.

The link can literally encode the answer in the URL, which you can just hide the raw url it in the emails’s HTML. (Fox examples, put it as base64 chars in query parameters)

Then you can host a (static) webpage that renders the text. This lets you host any users text w/o an interactive site. No live database, no ops burden, etc.

If you wanted to get fancy (which users of such a product probably would probably want) throw in “success/failure” links so your users can report the results and get changed frequency of spaced repetitions based on their success rate.

For Pete sake, don’t use an LLM.

Just use CSS to hide/show the answers, or a little bit of JavaScript for just that. Or scrolling, if it is text only. Or links to the answer an http server.

  I was thinking of replying to the email with what you think is the answer and letting an LLM decide if you remember correctly. Is that something that sounds effective to you?

Email seems like a no-go here. It would feel like spam. If you wanted a something more conversational, I'd consider doing this over text messages.

You can practice retrieving something without entering and checking it.

More information on #2 (colloquially known as 'active recall'): https://en.wikipedia.org/wiki/Testing_effect

Very interesting concept! I was about to dive right in when a voice in my head reminded me that my inbox discipline is terrible and I don’t need even more automated emails that will serve only to increase my unread count (which currently stands at 16878 in my Personal inbox).

Have you considered expanding this to notify via other mediums such as iMessage (if possible, this would be my preference) or the presumably easier WhatsApp/Telegram (the Telegram Bot API is pretty great, I’d imagine would be very easy)?

I’d also echo the free trial sentiment expressed elsewhere in the comments, take the mythical drug dealer approach and get ‘em hooked on freebies!

Yes, that's definitely a major drawback for people who are not so active via email (or got their inboxes spammed by marketing haha).

Will explore some options with the messaging apps, thank you! Do you think at this point it's not easier to open a specific app for that or there's a reason you prefer it to be via a messaging app?

Getting messaged is different from opening Anki, which I often avoid for months at a time. I imagine I'd be much more likely to respond to a notification just because of how strong of a habit that is.

I feel like writing a specific app is not only more effort on your part but easier to ignore on mine (and more friction, a lot of people are hesitant to install new apps).

Personally I'm a big iMessage/Telegram user and as such unlikely to leave things unread on those mediums (they're one of the few apps that I allow to push notifications and display an "unread" badge); my reading of your post was that a selling point of GinkgoNotes is that it appears in already established workflows.

Neat idea! I might consider pushing a free trial more since this feels like the type of product someone could begin to rely on and then be more likely to upgrade. A few positive testimonials wouldn't hurt.

A couple minor link fixes on your homepage, see here: https://triplechecker.com/s/230357/ginkgonotes.com.

1) Blog on footer is broken link 2) Affiliates on footer doesn't take you anywhere

Yes, I agree with both statements. Although I'm just starting and don't have any testimonials yet, I was thinking of sending an email to users requesting testimonials in exchange for the premium plan.

Just fixed the footer, nice catch! Shipping fast comes with some faulty copy pastes I guess...

That could work. Happy to help!

Storing facts is a poor proxy for acquired knowledge or skill. When you are good at something, you know a lot of facts. Trying to memorize facts to get good at something is inverting the order of factors. You memorize a lot of facts about a domain because you spent a lot of time and attention in it. But finding shortcuts to memorize facts won't in fact give you the actual knowledge and skills, just a way to mimic them convincingly.

I recently started learning Japanese and "storing facts" using spaced repetition, and it's great for learning Hiragana / Katakana.

It's also very useful for things like passphrases.

Sometimes, you just need to memorize things.

I have acquired several languages, and I can tell you that naive memorization is a poor proxy for language acquisition. The best way to acquire a language is to use it a lot, consume a lot and produce a lot of language. Memorizing words and grammar may feel like learning a language, but it's very inefficient compared to just consuming and producing language in a natural setting.

Research disagrees with you. Memorizing vocabulary list is one of the most efficient use of time for learning a language. Sure it's not sufficient alone, and not fun but it works. It's especially useful at the start when learning even a few hundred words makes a huge difference.

Which research? I know this is mainly related to Duolingo, and it’s in Danish, but outside of confidence, grammar and vocabulary there doesn’t seem to much benefit in memorising vocabulary lists in terms of actually learning a language.

http://www.carstenbuus.com/da/00007-jagten-paa-det-graeske-s...

(not the parent poster)

https://www.academypublication.com/issues2/tpls/vol05/04/25....

> Abstract — There are disputes over the role of memorization in language learning. Memorized language, a mainstay of education for almost all of recorded history, was widely repudiated for suppressing creativity, understanding and enjoyment in learners. This paper aims at highlighting the fact that, despite these criticisms, memorization is a helpful strategy which can be employed by the learners and teachers in their process of language learning and teaching. It is discussed that memorization: 1) provides the learner with linguistic data; 2) is the first step to understanding; 3) enhances association in memory; 4) causes cognitive development as a learning strategy; 5) helps noticing; 6) provides rehearsal; 7) is especially helpful in early stages of learning. It is also pointed out that all these become possible when memorization is accompanied and complemented by other strategies and techniques.

I don't know why you're getting downvoted. Acquisition of a second language is a very narrow field of research and to claim that there is a clear consensus on what works best is showing very little insight.

> but outside of confidence, grammar and vocabulary

Just those, huh?

s/Romans/Duolingo/g

But for the task of learning the letters, this does wonders.

This is just a tiny part of learning the language, but it's a great tool for quickly learning letters and new words.

This is a common fallacy around science students who use anki for the first time. Smashing a deck without accompanying learning ressources won't lead to connecting the dots like you would in a traditional learning setting. So given an exam you would probably fail as you can't connect the dots because you just memorized facts in a random order. The exception is of course language learning for which anki and spaced repetition was made originally because random order doesn't matter and connecting facts neither.

> Smashing a deck without accompanying learning ressources

why does everybody keep pretending this is a dichotomy?

This is such a smart answer. i am adding this comment so i can remember where to look when I am thinking of re-reading your comment about memory;

For me, more than memorising things, is creating paths in order to find that piece of information when need it; and yes, you are right, we memorise when we need that information to apply it in a specific moment;

I finally admitted to myself that a robust PKM can store and recall more information more easily than I ever could. It's quite a bit of work to process things I read or listen to into my PKM, but whenever I decide to revisit an idea, I can immediately pull up and review everything I've previously thought it would be useful to remember about it.

My PKM of choice (Logseq) does have a built-in SRS, though, so maybe I should consider having putting things I really, really, really want to remember into it so they're stored in my brain as well.

I think you're making fun of me, which is fine.

But memorizing in stead of learning is an anxiety driven behavior that leads to inefficient use of time. You're welcome to do it but I am trying to contribute my own hard earned lesson that you don't need to fret about retaining facts. Just spend time doing the things you want to learn about and the useful facts will be retained.

I developed an open source app that tries to solve this problem. Active recall at the topic level with spaced repetition email reminders using FSRS algorithm. https://github.com/Gerosullivan/Learntime

Readwise also has this feature. I get a daily email with a random assortment of highlights that have been pulled in from multiple sources (Reader, Notion, Kindle, etc.)

The product benefit in their case is that it's kind of like Zapier, but for notes.

https://readwise.io/

Gingko notes have been a thing maybe 10 years ago? I'm not sure if I remember correctly, but wasn't graph-based note taking? Was really cool!

The idea of spaced repetition via email reminds me of readwise as well.

You're thinking of this [1]. Super duper cool idea, ahead of its time. Author seems to have transitioned to a ludicrous pricing model, but I suspect it's because the people who like it really _love_ it. I used to be among them, but then switched to org-mode .

[1] https://gingkowriter.com/

I’m curious what you mean. $9/mo for a tool targeting professional writers doesn’t sound that ludicrous to me.

It's all a matter of point of view I guess. Given that I replaced it with something costing zero dollars may explain my attitude. Also, the fact that it started out as basically free, and is now a recurrent expense over a hundred bucks a year.

That said, the author seems like a cool guy. Hope he's finding success.

What did you replace it with?

Emacs org-mode -- it's less visually obvious that it's exactly equivalent from a structural / hierarchical point of view. But the editing experience is infinitely better, because Emacs.

Yes, Readwise was sort of an inspiration for me too, but looked like wasted potential as they don't track or use any algorithms for you to remember, and lack support for external links/sources (I may be wrong here).

Well done shipping your project! I like the low-friction demo on the page! I assume I'll receive an email in a few days from now? Are you using those emails for marketing/a mailing list?

Thank you! You're both right—you'll get your repetitions, and I'll add you to the user mailing list (unless you opt-out).

With the number of cards I have in Anki, something like this is going to fill up my inbox. Recent Anki versions ship a much more advanced algorithm called FSRS which performs much better (fewer reviews for the same recall). If you haven't tried it in some time, I recommend taking another look. Also, many note taking programs have ways to export to anki, which is probably more pleasant than adding notes directly.

Cool idea, can it just be an app that sends me notifications instead of the email? Anything that goes to my email is basically a black box and I won't see it. I am an email hoarder and I don't clean them up... Once every other year or so I'll try.

The only thing that helped was routing "updates" and "promotions" straight to my trash, but I miss things sometimes that way.

Seems like the root cause is not email.

I am curious how “another app” would be better in your situation than just setting up an email rule (3 min task at most) for this mailer if you truly care about it.

nice one, what differentiates it from many other tools like https://x.com/yapthread or https://x.com/SupermemoryAI is the scheduled delivery part. i wish to remain on the same track , add some one capabilities like RAG for QnA with the notes / bookmarks created and show maybe reference sources.

What the --

This is such a great idea! I've had this same problem for as long as I can remember, and I feel like most intellectually-oriented folks have this same problem, as well.

My work firewall blocked the site, but I will check this out later :)

Thank you! Glad you like the concept. Don't hesitate to shoot me an email if you have any suggestions/features you'd like to see :)

I use mochi.cards and am very happy. It has a really good app and it's very privacy friendly. I recommend it to anyone who is studying or preparing for any event that requires efficient memorizing.

This is very cool. Reminds me of Quantum Country (https://news.ycombinator.com/item?id=30467585) but for everything else in life.

Email is the only thing I have conditioned myself to reliably check so this system probably has the best shot for me. Nice job, will try it out. Ive been meaning to make a real attempt at learning Japanese.

Thank you! Have fun learning Japanese!

I have 3 email in-boxes. One ancient rarely used since the early 2000, my personal daily email and my work email. Between then there are some 100k unread emails.

I am not the target audience for this service.

I sometimes go most of a week without checking any email address. I’ve gone a month without looking at my personal one. It’s basically just for transactional mail, I don’t get things from people there.

I do an annual amnesty, every email goes into archive on the first of January.

I think there are a lot of people like you, so I would definitely look into other ways of communicating. Thanks for the feedback!

If it was over WhatsApp, for example, would you be interested?

This is really impressive! Would you mind sharing some analytics like subscriber count or DAU? I'm curious to learn more about how your platform is performing. Thanks!"

Thank you! Sadly, it's just at the start of the project so I don't have anything to share yet.

I'll give an update after some time of collecting metrics. You can also follow me on X to be in touch with the project.

I really like how you can create a note from the landing page. 1) it helped me understand how the app worked better than any screenshots or videos while 2) nudging me to sign up.

I made the exact same thing except with notifications instead of email.

I called my app "Mnemonist".

My last as just hasn't gotten around to shipping it yet...

I can't find them who would use this over Anki or AnkiDroid.

I have hundreds of items to review. I wouldn't want that in my inbox.

I get you. In cases where you have that many cards, that's definitely not for you.

My case is rather 2-3 quotes from a book I just read of 5ish articles per week which I find interesting and don't want to forget (so it will email them again to me after some time so I can reread). Thanks for the feedback!

Anki scales down to a few questions, and is a local-only app requiring no infrastructure for mail delivery.

A book you read will eventually turn to twenty books you read, and then you have 40 to 60 quotes and so it goes.

Looks awesome - would you mind expanding on the tech stack you used?

It seems you've made a digital mnemometer!

I wish there's a away I could upload my screenshots then they get e-mailed at intervals

For now, I have found a workaround which is to just paste the screenshot URL if you are using one of those apps that automatically uploads them.

I'll let you know as soon as the image upload is live!

Interesting. I couldn't see myself using a tool like this for bulk remembering of things. And I don't think it would be practical for that. But for specifically important things like certain words my brain like to forget from time to time, numbers I need to remember or a very important hospital appointment, this could be very useful.

The idea is good but:

1) Paid model, $10 for an email on timer? Copilot, Spotify and Windsurf charge the same. Won't pay on principle THIS amount of money for a service which is done self-hosted in 30 minutes with ChatGPT, neither will a normie buy paid calendar, no offense. $1 is realistic, still you're gonna be profitable and much more scalable imo. Trial period of 2 weeks is mandatory

2) I almost never read mails, plus email notifications are inconsistent if you're not inside an ecosystem, e.g. Apple/Google. For me it is not a reliable way of communication

I'm a big fan of SuperMemo16, if you could add incremental reading that would be great.

This is a super cool idea!

Also, which spaced repetition scheduler are you using for this? FSRS?

Yes, it's FSRS based on this article

https://github.com/open-spaced-repetition/fsrs4anki/wiki/Spa...

Huge thanks to the dude.

The comments here remind me of the HN comments when Dropbox launched!

Cool idea! What were your other 3 projects this year?

This looks like it might have been one of them? https://ginkgovest.com/.

*edit* seems different - google search led me astray. I didn't realize ginkgo is a tree with a distinctive leave

I kicked off the year with a web app that summarises foreign books using AI.

2nd is a desktop emulator of an iPod (macOS only)

3rd is a Stardew Valley tool

Let me know if you need a URL for any of them :D

> you can remember anything

Except to check your emails.

Could be a good feature for AnkiWeb.net

Wish this existed when I was in school

Where is the code?