I have quite a bit of decks, each broken down hierarchically, and at any given time I have quite a bit of cards to review. I don’t utilize Anki quite as the doctor recommends, but it works for me.
On any given period I’m particularly interested in reviewing a subset of my decks, strewn across various deck hierarchy trees. It would be nice if it were possible to aggregate these together in a single location. I’m thinking here of something like a symlink. Currently it’s very easy to move around decks, but I’d have liked to be able to keep the original membership of the decks, so that they would still be reviewed when reviewing their parent deck, they would be included in the statistics of the original deck tree &c.
I’d like then to have a, let’s say, digest symlinking deck, under which other decks could be directly accessible as if they were one hierarchical level below it:
Perhaps I should motivate my request more concretely.
I’ve helped myself learn and maintain several languages with Anki. Each of these have thousands of cards and at any given time some of these decks would be a bit neglected.
I subdeck each language like so: I’d have a subdeck with forward cards, another with backward cards, another for other kinds of notes. I like practicing these separately if only because for me reviewing forward cards is easier, and in some way it makes sense to be on top of that —recognizing words— before being on top of producing words.
For me having the “forward cards” subdecks of all my languages under a single symbolic deck would make ploughing through them easier. Sure, I could jump between them over the deck tree, but having them all in one place would make not only their access simpler, but in a banal way, it would make the decision of reviewing them all tad easier.
For this reason tags would not be on help here. First, I’m not aware that there’s a way to do a custom study based on tags? (I’ve never used the custom study functionality beside incrementing today’s new cards number so I’m not certain how it works) But also tags are properties of notes, while here the subdecks are sequestering cards.
I reckon that it would still not do the trick for this here case, because such a filtered deck would mix together all the subdecks I’m interested in into one pile, so that I’d have to review i.e. German cards intermixed with French cards.
Does anybody have an insight about how challenging it might be to implement such a functionality as an addon? That is, does the way that the Decks screen is rendered lend itself to showing “duplicates” or is there anything that would obviously make it a problem?
I can code Python and did make an addon for my own use once, though it was quite a few years ago now and I’d have to dig again in the documentations..
You’re not limited to only one aspect of the card. So if you used all of those “food::…” deck-names as tags instead, and you wanted the French red fruits “forward” cards, you could think about how you would search for those in Browse –
deck:French card:1 tag:*red* tag:*fruits*
– and that’s your search filter for your Filtered deck [Edit > Create Filtered Deck]. You’re calling it a “symlink” – but it’s really just a way to bring together an assortment of cards, based on your own criteria (and perhaps without regard for whether or not they are due). But that’s what a Filtered deck is.
[As far as the suggested add-on – I’ll note that making actual duplicates of your decks/cards seems like the opposite of what you want here. Generally it’s a terrible idea. Studying multiple copies of a card splits your review history between them, and makes it impossible for Anki to schedule them correctly.]
Indeed, I do not want to duplicate any cards.
I do not want a filtered deck either, or I don’t think so: that would put all the legible cards into a single pile, if I understand correctly, while I only want certain subdecks to be next to each other (or rather on top of one another) without removing them from their appropriate place.
I’ve been using this (my second) Anki account for 2000 days, and accumulated much over the years. I’ve truncated my collection a couple of times, so to speak, but I still have quite a bit of subbed decks.
I understand that this is contrary to the doctor’s orders. I’m not at all on top of my reviews, but I’m ok with that. It has worked for me. Sometimes I’d concentrate on one deck and review depth first, as it were, other times I’d go for breadth first, reviewing a subset of subdecks across the decks. Being able to move them together without moving them (therefore the “symlink”) would have made doing this breadth first review easier.
As a simpler solution, I’ve had the idea of being able to create a queue: something like a playlist of (sub)decks that one can define, which persists across days, where one reviews each item (subdeck) at a time before moving to the next one. Like a filtered deck but retaining the groupings of the subdecks. If that makes sense?
You say “a filtered deck” like it’s a singular thing – but Anki can put your cards in any number of “piles.” It’s up to you to make a separate Filtered deck for each “pile” you want. But you’d need to do the same thing for each separate “symlink” you wanted.
So – you want to temporarily gather together these cards so you can study them, with them always remembering what deck they belong to? I hate to break it to you, but that still sounds exactly like a Filtered deck! When you’re done studying the cards, they’ll all be happily back in their home decks with their updated scheduling.
That sounds like just clicking directly on the subdeck you want to study. Is there a reason why that won’t work?
Thank you Danika Dakika! I don’t know why you hated to break it to me.
I’ve had some wrong assumptions about filtered decks, and I’ve always been wary to experiment with them because I was not sure how reviewing them affected the normal algorithm, let’s say.
Taking a courageous step into checking them out, I see that you are right. In fact, it’s even better than the solution I envisioned. Having filtered decks under my belt, being aware of how they function, I could simplify the structure of my subdeck hierarchies, too.