Plan to write a paper for FSRS

[L-M-Sherlock]

FSRS has been released since one and half years ago. Many things happened and my understanding about memory model is also improved. I find out many incorrect practices in spaced repetition algorithm researches (including the first paper by myself). So I want to write a new paper about FSRS and other findings, such as the metrics, evaluation, simulation and so on.

The paper will help researchers learn about FSRS, provide proofs for softwares using FSRS and promote the progress in that field.

Here I'd like to call for co-authors, reviewers and supporters. Any suggestion and contribution is welcome.

For the dataset, is it permissible to use the revlogs collected from Anki users for my research? @dae

For the co-authors, @Expertium @user1823, you help me a lot in improving FSRS, so I want to add your names in the acknowledgement section. It's nicer if you want to contribute to the paper directly (and to be co-authors).

[dae]

is it permissible to use the revlogs collected from Anki users for my research?

Yes, but if you need to provide the dataset to others when sharing the research, please make it clear that people will need to ask permission before using the dataset for other things.

[L-M-Sherlock]

Is it possible to publish the dataset in Dataverse? For example, the dataset of MaiMemo is released on https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/VAGUL0. We can write License/Data Use Agreement for the dataset.

[dae]

If we can list a custom license, then it's ok. There's some ambiguity about "non-commercial" in the CC license. I'd like to make the data available for university academic work / papers, but not available for use by company-backed research/development.

[L-M-Sherlock]

It's nice if you could write the license or upload the dataset because English is not my native language and I'm not familiar with law terms.

[dae]

I'm not a lawyer, but how about this:

This data was provided to the FSRS project for use in memory research. University students and individuals may also use it for their own research. Please do not redistribute the data publicly (other than linking here). If you wish to use the data in other contexts, please request permission on https://anki.tenderapp.com.

[L-M-Sherlock]

I hosted the dataset at huggingface.co: https://huggingface.co/datasets/open-spaced-repetition/FSRS-Anki-20k

[Expertium]

I'd love to contribute to the paper if I can, of course.

[user1823]

I would also love to contribute.

[jiweiyuan]

I want to be reviewer of this work

[joshdavham]

I'm also down to review if you're looking for more help. My native language is English and I've been meaning to study the internals of FSRS for a long time.

[RickCarlino]

Not sure how I can help, but please reach out if there's anything I can do to support the project.

• Software developer by day
• Implemented a typescript FSRS implementation based on the previous paper.
• Native English speaker.
• I use FSRS every day in a custom application I built.
• My contact information is very easy to find 😉

[Expertium]

So did you put off the idea of writing a paper?

[L-M-Sherlock]

Yes. I feel tired.