A study system built on one honest idea

Learning goes faster when you can see what a goal rests on. ManySpace maps mathematics, physics and AI/ML through their real prerequisite links, then tracks what you can actually do and tells you what to study, review or fix next.

How it decides

• Mastery is computed from evidence (theory, practice, recall, review health, recency), never from a self-rating alone.
• Reviews are scheduled by the open FSRS memory model, aimed at the retention you choose.
• Routes come from the prerequisite graph: the shortest sound path to the goal, with the weakest link named.
• Practice problems are generated with exact answers and aimed at about 75% success, so they stay hard enough to teach.

What is finished, and what is not

Some topics have a full interactive lesson today. Every topic has the study cockpit, review scheduling and, where a problem family exists, generated practice. The curriculum text is in English for every topic and in Turkish for most; untranslated topics are shown in English and are not indexed.

Part of Aporia Systems

ManySpace is a project of Aporia Systems. The main project website is aporia.systems.

Credits

• Milky Way backdrop: NASA/Goddard Space Flight Center Scientific Visualization Studio, Deep Star Maps 2020, from ESA Gaia DR2 data. NASA SVS 4851
• Review scheduling: the FSRS algorithm by the Open Spaced Repetition
• Math rendering by KaTeX; 3D by three.js; charts and tools are plain JavaScript.