Hi, my suggestion is to link cards and manually mark them as siblings to take advantage of the automatic burying feature.
The “problem” I have is that I have a couple of cards that are slightly different (and not on the same note) but talk about the same topic, so when I review one I don’t want the other(s) to show up as it would help me.
Linking them as siblings would solve this because of the built-in burying feature (and the “disperse siblings” feature of the FSRS4 helper addon, even more effective).
My idea to implement this is to add a “mark as sibling with” button in the browse window and then we could select other cards to link.
I hope it’s understandable as English is not my first language, hope to hear back from the community to know what you guys think!
This explanation presents poor arguments and I do not agree with it.
would have to define all the relations yourself
No one is forced to. Most cards would not need this anyway, but it’s good to have an option.
you’d have to search through the rest of the deck and assign relationships between the old and new material
Not if I create all cards within the group at the same time.
Adding constraints to card display that cause cards to display earlier or later than they were supposed to will make the spaced repetition system less effective
If “Cloze” and “basic and reverse” are allowed to make this, what makes manual linking different? Double standard.
More context:
Sometimes I don’t really see a way to organize some closely related meanings / alternative word definitions, into one note.
Apparently right now only cards from the same single note can become siblings.
I would really like some way to manually (or with some other magic way) mark some technically-unrelated cards as siblings, to enforce that there will never be a review of more than one card from the same sibling group on the same day.
(the topic of the deck is Dutch vocabulary)
There is a way to do what you ask, but it is much more cumbersome than what you want.
You can make a new note type that has space for all the cards that you want to link. Shige fixed the add on that allows you to copy the scheduling data from one card to another.
So if I wanted to do this for a basic and reversed card, I would have fields something like {front1}, {back1}, {front2}, {back2} etc. Then have cards for each field from the same note.
Not if I create all cards within the group at the same time.
I agree with you to some extent.
I believe the idea of siblings can even transcend note types. A specific field (like the is for tags) irrespective of the note type could contain a unique string that binds all cards of a note with that same unique string together as siblings.
I also really wanted to take advantage of the FSRS4-helper-add-on disperse siblings feature but, if this could only have effect on card types of the same note with static field content (aka siblings), then I think it usefulness is way limited.
If there was a way to link any arbitrary cards, then theoretically, when you review card A, FSRS could also adjust the interval of related card B. This could significantly reduce the number of reviews for highly interconnected knowledge. But this would require changing the way Anki’s database works and would require too much effort to implement. So I doubt that this will ever happen.
Not really sure the extent of interconnection you imply when you say “highly interconnected knowledge” but, @guffaws said;
There is a way to do what you ask…
However unconventional that workaround is, its still achievable. In what @guffaws describes, one can have more than two (several, actually) card types out of the same note (aka what binds siblings currently). I suppose Anki’s database will work fine either way, when one manages to pull it off.
Even if it has to be just two card types (which is what you get with Basic-and-Reversed-card) why should it be limited to using the same note, as a binding agent; when some level of flexibility could be allowed as I discussed initially.
Perhaps there could even be a limit to the number of siblings one can define (say five) just to moderate the effect.
Here’s an example. Suppose there are two linked cards, A and B. Normally, when FSRS increases the memory stability of a card after a review, it does not affect another card. But imagine that it does. And for the sake of simplicity, if the memory stability of card A increases by x, the memory stability of card B also increases by x, and vice versa. And again, for the sake of simplicity, assume that without linking them, memory stability increases at the same rate for both of them.
With linking, it will increase at double the rate. Every time card A is reviewed, the memory stability of both cards increases. Every time card B is reviewed, the memory stability of both cards increases. Since intervals are proportional to memory stability, intervals will also grow at double the normal (without linking) rate. So just by having one link you doubled the interval growth rate of both cards.
What about three linked cards? You can tripple the interval growth rate of all three cards.
Four linked cards? Quadruple the interval growth rate of all four.
Of course, in reality, there would need to be an optimizable parameter that controls how much of a “bonus” linked cards get. So if card A’s stability increases by x, the stability of linked cards only increases by, maybe, 0.5x or 0.2x. Still, that would be beneficial, and could noticeably reduce the workload. The issue is that in Anki, FSRS (or any other algorithm) cannot access other cards when you are reviewing a specific card.
Then you will need to take my earlier recommendation. Make more complicated note types that have all the cards that you want to have siblings. There are add ons that allow you to transfer scheduling data from one card to another, so nothing will be lost using those.
Alternatively, you can create an add on that does what you want. You can use the one referenced as a starting point. I’m not good enough in python programming to do it myself.