[krychu]

This is fascinating. I’ve come across SuperMemo and use spaced repetition for self-created decks. However, incremental reading is something new to me, like a missing piece. Seems like it’s a process of turning any linear material into a deck. Have you or anyone else use it? And could comment on results / issues?

[LittleFishyChan]

Hey how cool to see others still finding out about SuperMemo! I’ve been using it for 15 years, it’s super great, incremental reading is an awesome game changer, basically it’s like a book emulator with lots of extra features. I started making YouTube videos about it, I just uploaded three today about incremental reading: https://youtube.com/channel/UCpkWsnAPl-rhGwUOqdi2Vow

[alessivs]

It is the process you broadly describe, and more. I use it and it has turned addictive. The very registration of texts into a priority queue (SuperMemo implementation) relieves from stress inherent to keeping reviews of high volumes of possibly disparate, disconnected, badly formulated, texts, prior to properly formulating them in active-recall form (usually question and answer pairs), to commit them to memory.

Because these reviews are distributed (it is a separate algorithm to the one governing core Q/A repetitions) and also interleaved with other material, you are encouraged to operate on portions of texts on each review, giving your brain a chance to consider different successive improvements, such as:

- transformation into smaller portions (which are themselves registered and subject to distributed review),

- introduction of supplementary material (e.g. other articles, multimedia),

- better wording,

- contrasting with other sources,

- personalization.

It beats the single exposure to material and upfront "ankification" (formulation into Q/A cards), in which biases—such as single-perspective bias, or recency bias—are prone to manifest more strongly in the formulation of your material. Furthermore, because the sources of information tend to remain unreviewed after this (i.e. kept off-system), there could also be a tendency to accept their premises or claims without competition from other sources.

Issues I can find with it usually arise from the desire to meet tight deadlines: it is counter-productive to distributed reviews, crystallization of knowledge, and memory consolidation, to rush the process. The SuperMemo implementation also has a vast toolset, which may be hard to grasp in the beginning. This article[1] provides a coherent explanation of the process from a user perspective.

Edit: SuperMemo 15 and above are incremental-reading-enabled; there are freeware downloads and trials should you be inclined to try it[2].

[1]: https://www.masterhowtolearn.com/2019-08-06-supermemos-incre...

[2]: https://super-memory.org/english/down.htm

[xkfm]

I've used it for about a decade or so. It's really good, but the entire program is tightly coupled with old MS APIs/IE, so importing material is a pain.

Roughly the process goes: Import article -> read section (create extract/readpoint (a bookmark of where you were in the article) -> review extracts/articles -> create cloze deletion/card (if necessary).

I can't use any other app for this process because you can't really schedule material like that in Anki/Polar Bookshelf/RemNote. Supermemo is unique in that the material you want to make cards of eventually gets scheduled itself, so you don't have to remember to read something. I would use Polar Bookshelf, but every time I've checked in on it, they haven't added that feature. In Supermemo can import dozens of articles at a time, assign them priorities, and know I'll eventually be shown them to read.

I have a couple thousand articles imported at this point, and I never need to remember to read them. This is the most important feature for me, and why I've continued use Supermemo for so many years rather than switch to Anki or Polar Bookshelf.

Not needing to remember to read something you found interesting, rather than just save it as a bookmark somewhere (every bookmarking app) is a killer feature for me, but apparently is not very useful to other people. I've tried explaining it to several people (including the developer of Polar Bookshelf), and they just don't really care.

The longer I've used Supermemo, the fewer cards I end up making from material (most things aren't really worth memorizing/remembering or even reading once you step away from it for a few days/weeks). But, I still import a lot of articles/comment threads from HN etc I eventually want to read.

Often times I'll create extracts, and not actually make cards of them, because I just like rereading that particular paragraph.

I have tasklists of business/programming/etc ideas, that also get shown during my daily reviews. I can add to them, or generally get reminded that I was thinking about X at some point. Very underrated feature.

If you don't want to see something again, you can dismiss it from the review process so it doesn't show up in your reviews. However, the information is still in your collection, so you can still search for it.

Most people take notes and then never look at them again (this comes up in every note taking thread on HN). Supermemo continuously shows me my notes over days/months/years. It also shows me interesting articles I want to read on a daily basis. It also helps me remember whatever I want. All I have to do is open the app once a day and hit next repetition.