SpacedBlue is a playground for experimenting with text and spaced-repetition interfaces, workflows and ideas.
The objective is to explore the latent space of possibilities to interact with text, and by utilizing interfaces that map across both text and memory, using (or not) different spaced repetition algorithms (SM-2, FSRS and even simpler/novel ones).
Wrong, strange, impossible, idiosyncratic, novel, different ideas are welcome and encouraged. Open a discussion at any time regarding anything you’ve been mulling over for a while, or any concept emerging from an unclear space. All contributions are welcome and are filtered based on the project's scope.
The current domain spaces are:
-
Generate cards directly in the note's body: after creating and opening a deck, a text note can be added, and pressing the shortcut for the Q&A, the cloze with empty body or the cloze from nearby text, inside of the note adds the text template from which a card is automatically created and instantly reviewable.
-
Split articles in paragraphs that are visible in review: after creating an article with a title and a body, the content is parsed and splitted in paragraphs. Pressing A while a paragraph is focused opens the dialog to add a Q&A question under it. During the review, the related paragraph is shown, and optionally the previous and the next.
-
Spaced repetition algorithms: the implemented ones are FSRS by the Open Spaced Repetition community, SM-2 and Leitner system. I feel there's space for different types of spaced repetition algos (framed for mid-term memory goals, auto-updating based on cards added and retention), specifically ones that can predict and achieve reliable result in the short-medium term, without relying on cramming.
-
Text content extraction and serendipity: lines in a note that starts with #[word] are grouped and filterable by tag. Article's paragraphs order is randomizable. Everything is readable text, no binary blobs.
- Interconnected cards: cards linked, dependent on others e.g. jpdb already implements this for learning kanji using radicals.
- Recursive cards: cards about cards
- Definitions learning: interface designed to learn word meanings in context
- Each algorithm its own: develop an algorithm for organized fast learning (not cramming) for interests sprints
- implicit memories: ease the discovery and integration of implicit memories
- Modes l vim: make the tool invisible, discover patterns to ease the creation of meaningful meta-content (cards)
- External addon to generate cards from the browser
- Cards/Evergreen notes as autonomous cells: Dave Ackley is working on an indefinitly scalable system based on T2 Tiles. Michael Levin is also using the cell metaphor for exploring the latent space in digital computing, for example in simple sorting algorithms. What if evergreen notes (for their great atomical structure, much similar to the sorting array example), acting as cells and using simple algorithm (behavioral science?) to move in a 2d (3d?) space of notes, could connect and act in unprecedented ways, forming clusters of meanings, topics unsual but with a valid extractable signal/pattern? Something like bottom up self-organizing living notes.
Clone the project and open it with Godot (4.7).
MIT License. See LICENSE for details.
Researcher in the field of learning, specifically UIs.
The project Quantum Country, developed in collaboration with Michael Nielsen, unifies text with spaced repetition cards, opening up many avenues for testing.
The concept of evergreen notes revolutionized my usage of the text file metaphor and informs lots of my workflow.
The creator, with Laurel Hulley, of a psychotherapy approach named Coherence Therapy.
The paper A Proposal for the Unification of Psychotherapeutic Action Understood as Memory Modification Processes made me realize how much memory is at the core of the human experience, and even though the spaced repetition process is related only to the explicit aspects of memory, I think there's space to research the implicit ones.
Biologist, professor and lab's director at the Tufts University.
His idea that there exists a latent space composed of organized patterns discoverable through experiments, and the goal oriented nature of systems, is what makes the interface’s creation process interesting to me (and not only).
The father of the first spaced repetition algorithm.
I spent long evenings reading his blog, which shaped my interest in how memory works, and the learning process (specifically the concept of pleasure of learning and all its corollary ideas).
Philosopher and psychotherapist that developed the Focusing method and co-started the experiental shift in the psychotherapy field.
He also developed the Thinking At The Edge method, to find fresh new language and methaphors to move forward a research idea blocked by a felt sense i.e. how to connect and symbolize a latent space of patterns. The explicitation interview stems from the work of Eugene Gendlin as another avenue for symbolizing the felt sense.
FSRS algorithm: https://github.com/open-spaced-repetition
Font: font license
Icons: https://feathericons.com/
Script tabs addon: https://github.com/Raven8088/godot-script-tabs
Godot: https://godotengine.org/
I'm always willing to discuss ideas.
Email: plucafz@proton.me