[TODO] switch of enable/disable short-term schedule

[L-M-Sherlock]

The implementation of ts-fsrs:

• Feat/long-term schduler ts-fsrs#100

[joshdavham]

Can I get some more context on this? I have a lot of trouble reading typescript.

[L-M-Sherlock]

I also implemented the Golang version:

• Feat/long-term scheduler go-fsrs#20

[joshdavham]

Code aside, at a high level, what is this new long / short term scheduler thing? Is it described anywhere else other than in source code?

[L-M-Sherlock]

@ishiko732 , do we have the diagram?

[ishiko732]

do we have the diagram?

Currently, there isn’t one available; it might take some time to create a diagram.

[ishiko732]

I have updated the diagram, and you can download the original file from the following link: https://drive.google.com/file/d/1oigP0bheX9cXOO9baqwDTzAZ9IWvpCio/view?usp=sharing

[ishiko732]

Long-term Scheduler WorkFlow:

Enable_short_term:

[ishiko732]

Short-term Scheduler:

Long-term Scheduler:

[domino14]

And what does it mean? I thought the interval times were inherent to the algorithm. Do the different schedulers have different constants?

[joshdavham]

@domino14 py-fsrs currently only has one scheduler. You can take a look at ALGORITHM.md for details.

As for the new long term scheduler option, which isn't implemented yet, there will be no learning/relearning state for the cards and furthermore, rating a card again, may not necessarily cause it to be due again in the same day, as the interval may be longer (e.g., 2 days, 6 days). The long term scheduler will be sorta be a more 'natural' version of fsrs with less rules.

[domino14]

Thanks for the response; I’m actually using the Go version but liked looking at this code since it is better documented. The long term scheduler for the Go version has no learning / relearning states?

…

On Mon, Sep 23, 2024 at 8:26 PM Joshua Hamilton ***@***.***> wrote: @domino14 <https://github.com/domino14> py-fsrs currently only has one scheduler. You can take a look at ALGORITHM.md <https://github.com/open-spaced-repetition/py-fsrs/blob/main/ALGORITHM.md> for details. As for the new long term scheduler option, which isn't implemented yet, there will be no learning/relearning state for the cards and furthermore, rating a card again, may not necessarily cause it to be due again in the same day, as the interval may be longer (e.g., 2 days, 6 days). The long term scheduler will be sorta be a more 'natural' version of fsrs with less rules. — Reply to this email directly, view it on GitHub <#56 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAEO4ZWD3K6BGISLPRROANDZYCWSNAVCNFSM6AAAAABNFHYF52VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRZHA2DMNJQGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[joshdavham]

@domino14 I can't say for sure. It'd be best to ask that question in the go-fsrs repo.