I use Anki to study vocabulary. For each word I add at least 2 cards: target language → native language and vice versa. Sometimes it’s even more cards, with example sentence, or synonyms. I have sibling burying enabled for all cards, so it can take a while until all of the cards to a specific note have been introduced.
I sometimes bulk-add vocabulary, so I often have quite a large backlog. But I want to limit the new cards per day to say 20 cards. I would like Anki to choose the new words randomly, BUT I want to prioritize those cards whose siblings have been studied already. So I want Anki to first check whether any of the already introduced notes have siblings that are still unlearned and not buried right now. In this case I want to study those cards first. Only if the number of those cards is <20 I want Anki to introduce completely new notes. So in essence: I want to minimize the days I need to introduce all of the cards associated with a specific note. I want to avoid studying the target-language → native language side for 100 words before starting to see the native → target cards. But for each of the notes, I want to see the cards in order, so first target → native, then native ->target, then example sentences etc.
My best guess for achieving this was:
New Cards.Insertion Order = Random
Display order.New Card Gather Order = Ascending Position
Display order.New Card Sort Order = Random note, than card type # this is to ensure that for a specific note card 2 is not introduced before card 1
But this is not working well for several reasons:
1: If I started studying a number of cards and then add more notes, - the new card positions could turn out to be higher, than the old ones. so my method only works if I have a batch of words and study all of them before adding more.
2: I am confused of the way Anki is generating the random new card positions. Recently I added cards and they all got positions between 1 and 90. now I added more yesterday and they all got positions in the range between 90000 and 1000000.
What is even more confusing and also breaks my setup is that if I want to manually reposition a card in ankidroid than the highest position I can enter is 99999. So if I wanted to move some cards to a position after the cards I created yesterday, than this is not possible. (other than doing it the other way round and giving the cards from yesterday smaller numbers)
Why is that? And is there any way of achieving what I want in Anki?
I am using AnkiDroid and Anki for Linux. The above described issues have occurred on AnkiDroid.
Please let me know if you need more information! Thank you so much in advance for your help!
This has to do with the Insertion order you’ve set. It’s recommended that you don’t use it and as far as I’m aware, it’s mostly there for compatibility reasons. I can help you with what you want (though I’m not sure the random+not-so-random order is doable) but I’m not sure you completely understand the options you’re trying to use. Do you mind reading through the docs once? Deck Options - Anki Manual.
Hi @sorata,
thanks for your quick reply! Could you give me a hint on where specifically you think I misunderstand? I think I read every centimeter of this manual already, maybe I still missed something? I do understand that the insertion order “random” leads to the number that card position to differ, I mean that’s what random is about. What I do not understand is: Why do I add ten cards on Tuesday, that all get a position between 2 and 70, and then on Thursday I add 20 cards, and they all get positions between 90000 and 100000. With the “random” option, I expected the numbers for each batch of cards I add to show a somewhat similar distribution.
I also saw that using this option is not recommended anymore. Can you tell me, why?
I haven’t found any reason for that warning message – and I’ve looked! My only theories are that using v3’s ability is newer and therefore deemed “better,” or maybe there’s a slight performance difference when creating new notes?
Random insertion order works just fine, so I don’t think it’s anything you need to worry about. I’ve been using it myself for a couple years without any issues! It is actually the only way I’ve found to have a New card queue with a “front” that I can advance cards to, and still have a shuffled introduction of notes. [ I have been meaning to ask about this though, so I’ll do that in a separate post …]
But since you use sibling burying, you don’t need to worry about this possibility, right? You’ll never have multiple cards from the same note in the same day. Any of the sort orders are fine for you – and it only impacts the cards after they have been gathered.
Yes, with these settings, newly created notes/cards can be placed at the front of your New-queue (which are the lower numbers, based on your gather order), and delay the introduction of those leftover New siblings – but not for very long. Since I’m already introducing 2-5 cards per note, and it can take a couple weeks all of them to be introduced while they wait for their siblings to clear out – I find that potential for delay is not really noticeable.
I can think of 2 possible reasons –
“Random” has a tendency to mean “random,” not “evenly distributed,” so it’s no more of a surprise to get queue numbers close together than it is to get queue numbers far apart.
While it doesn’t hurt anything for if multiple cards to have the same New-queue number, it’s possible that Anki avoids generating cards on a queue number that is already being used by another note. If your queue had a few openings between 2 and 70, and then you filled those up, Anki would turn to other available queue numbers.
So, how do those sky-high New-queue numbers happen? Mostly, they come when you import shared decks. Once your queue has been expanded to have a “queue-bottom” of 100,000 – it will stay there unless Anki has a reason to change it.
You can fix it! And I’d encourage you to, because unless you have 100,000+ notes, those queue numbers are just an annoyance. I recommend doing this in Anki desktop, because you’ll be able to see better what is happening.
Get all of your New cards in view in the Browse window – search is:new. [You can do this deck-by-deck if you want to, but since queue numbers only matter relative to each other, there’s no reason not to so your entire collection at once.]
Sort them by the “Due” column, with the lowest numbers at the top, and select all.
Cards > Reposition – [make a note of what the current queue-bottom is] – start 0, step 2 [or more, if you like, since the purpose is to leave some “room” in your queue for when you add new notes], leave random and shift unchecked.
You should be able to see immediately that your New queue numbers are from 0 to some-reasonable-number. If you’re not happy with how it looks, Edit > Undo, and you can try something different.
Once that is done, and all of the queue numbers from some-reasonable-number to 100,000 are empty – Tools > Check Database will fix your queue-bottom. You can check that it worked by opening the Reposition dialog box again.