Some shortcomings of an otherwise excellent app

Post a reply

:D :) ;) :( :o :shock: :? 8-) :lol: :x :P :oops: :cry: :evil: :twisted: :roll: :!: :?: :idea: :arrow: :| :mrgreen: :geek: :ugeek:
BBCode is ON
[img] is ON
[flash] is OFF
[url] is ON
Smilies are ON
Topic review

Expand view Topic review: Some shortcomings of an otherwise excellent app

Re: Some shortcomings of an otherwise excellent app

Post by Ernie » Fri Aug 23, 2019 3:46 am

I responded to your email. I'll include my response here also.
You made a suggestion for my "spaced repetition" mode, but I didn't fully understand, and I'm hoping that the information below will help you solve your issue.

I think you should use spaced repetition. That will generally work better for larger decks.

If you haven’t already, please read the below details on how my spaced repetition works:

The main difference between my app and others is that have a “new card” mode that is not based on timed interval. The app will pick 10 cards (by default) and *repeat* these 10 cards until you get a card correct 3 times in a row, at which point it gets an initial interval of about 1 day (so the card will become “due” again in 1 day). A new card will be taken and added to your 10-card “new” card set.

The purpose of the new cards round is to get some initial familiarity of the card before they start being scheduled as an “active” card at timed intervals.

I don’t know exactly how you want your scheduling to go, but below are a few comments:

Increasing “max new cards” is a way to increase duration of new cards. But, the app still show these same new cards again the next round if they are still “new”.
You might want to change the “new to active after” from “3 correct” to “1 correct”. This means that when you want a card to become active, swipe up or swipe left. Swipe down means you want to repeat in the next “new card” round
When a card moves from new to active, it gets an initial interval of either 20 hours, or 3 days (swipe left or up). You can change these
So you go through 200 cards in one day, the next day they will all become DUE, so you are now having to go through these due cards, plus work on new cards. It seems your goal is to get your intervals up as quickly as possible. You’ll likely want to increase initial intervals. Max initial interval for swipe left is 200 hours = ~8 days. This might be an option for you for cards you just kind of know. If you know them pretty well, swipe up and create a larger initial interval
I don’t know how you want to repeat cards, but you can auto-exclude cards once they’ve been answered a couple times (Cards to Study -> Streak) or after their interval is over a certain level
When a card moves from new to active, the statistics are cleared (starts fresh), so you’ll need to answer at least one time more in “due card” mode before a card can be filtered out in “cards to study”.

Hopefully the above will help. If not, or if you have more questions, let me know.

Some shortcomings of an otherwise excellent app

Post by saquiwej » Tue Aug 20, 2019 5:42 pm


I am writing to you with a request. Your Flashcards is the best flashcard type iOS application I could find after two days’ search on App Store. It has some shortcomings though, which could be eliminated in order to make it an even better program. One of the biggest drawbacks is the deplorable lack of information as to the role of the many, many variables governing its operation. I have spent more than a week trying to decipher what is what, but I am still in a kind of semi-darkness concerning some values.
My problem briefly is the following: I have a list of 4400 French nouns that I have gone through once with your app. I used spaced repetition, and it took me about 6 months to complete. Now, a year later, I would like to have a second pass and I find that spaced repetition is not the best method to adopt for this second go. The biggest problem with spaced rep is that there is no way to defer the second, third... appearance of the NEW cards. If your ROUND is small, you can get confronted with the same cards several times a day, which is rather counterproductive, especially so if you return to an older list already mastered to a certain degree. So, to counteract this I opted for the Leitner method, only to find its limitations as well. To maximize the ROUND to only 30 cards makes this method practically useless in my view. My request is that you increase the ROUND to at least 200 cards in Leitner, just as it is in spaced rep. In general I think, there should be a sleeping cycle (but ideally more than one) between the reappearance of any card (whether dismissed as CORRECT or WRONG) in a query. I have tried all kinds of ticks to arrive to such an effect in spaced rep but nothing worked. When I increased the ROUND to 200 cards in the hope it would suffice for 24 hours, at each break in study the program reset the counter and started from 1 again, mostly with the same cards it gave me a couple of hours before. Before, when I studied with the Leitner method some other decks, it continued the ROUND always where it had left it. It was a good thing, but it seems spaced rep doesn’t have this useful ability. So that I am completely at a loss. Then I tried to put the status of all the cards to ACTIVE, thinking I would have more influence on the cards INTERVAL, but it resulted in the app running out of DUE cards. So that I see but two roundabouts to remedy the shortcomings described above without rewriting the whole app:
1) Maximizing the ROUND in Leitner to 200 at least. This would suit people learning easy cards. Take for example a list of 2000 French nouns, where you only have to decide the gender (masculine or feminine) of the nouns. To decide between these two options you don’t really need more than 5 seconds, so that 200 cards a day is quite feasible. Studying such items, you won’t get very far with ROUNDs of 30.
2) Implementing the function SAVE SESSION in spaced rep, so that the ROUND of 200 should really consist of 200 different cards and not just reiterate the recent ones when you take a break. (By the way, I think it should be the behaviour by default, just as it is with Leitner.)
I hope you will have the patience to read this long post with attention. I was something of a programmer amateur myself. I have written an Excel macro for learning lists of words. It isn’t bad, but it’s much less sophisticated than Flashcards, and it works only on a PC, mobile Excel apps still being unable to run macros.

Best regards, and thanks for the nice app: