← Return to the learning studio

ASK SHAI · A SACRED HOMES LEARNING EXPERIENCE

Learning sources & software

What the connected skill paths contain, where to read further and which open-source components make the tools work.

Original teaching material

The 12 connected concepts, their 84 starting, practice, transfer and recall questions, and their worked examples were authored for ASK SHAI. The subjects are fractions, fraction operations, ratios, percentages, linear equations, equation exceptions, algebraic domains, square roots, statistics, data claims, reading evidence and writing revision.

Automated consistency checks cover the supported answer forms and authored answer keys. These lessons have not been independently reviewed by an educator, and no validated learning-gain, curriculum-completeness or credential claim is made. Written reasoning receives a visible review guide, rather than an automatic grade.

Hints, worked solutions, retries, new independent applications and delayed recall are tracked separately. A successful check is evidence about that particular question. Suggestions can be skipped. Imported history is labelled as imported and does not establish that the earlier work was independent.

Optional further reading

These links do not imply that their authors reviewed or endorse ASK SHAI. They open external sites only when selected.

Open-source components in the learning paths

ComponentPurposeIncluded licence
MiniSearch 7.2.0Search the authored notes locally, including small spelling mistakes.MIT
ts-fsrs 5.4.2Suggest review dates from minimal practice evidence. Reviews can be paused and capped.MIT
MathLive 0.111.1Optional visual maths entry. Plain-text entry remains available.MIT; KaTeX fonts: MIT
JSXGraph 1.14.0Authored fraction, slope and data explorations, with native controls and text/table alternatives.MIT option; bundle notice

Exact upstream package versions, package integrity values and individual file hashes are recorded with the project. The distributed files and fonts are served by this site. The visual keyboard and graph code load when their tool is selected. A separate accessibility check uses axe-core 4.14.0 (MPL 2.0) during development; it is not part of the public learning bundle.

Your work and device

Optional Python practice lab

The Python practice lab adds an optional download of approximately 18.2 MB for Python and symbolic mathematics. It uses CodeMirror 6 for the editor and Pyodide 314.0.7 with Python 3.14.2, SymPy 1.14.0 and mpmath 1.4.1 for calculations. See the included editor licences and Python runtime and package licences.

The lab is separate from the saved lesson pack. Each run starts fresh, and the browser isolates learner code from the site's page and stored data. Runtime networking is blocked. A five-second execution deadline and output limits help keep small examples responsive; they are not hard memory or processor quotas. The lab checks its hosting configuration before downloading the runtime and stays unavailable when that isolation cannot be established. Its examples are learning explorations and do not award grades or independent-learning evidence.

Learning progress and read aloud

Learning-path answers remain in the open page. Saving is off by default. When enabled, only minimal item evidence and review preferences are stored in this browser. Typed answers and conversations are excluded. Anyone using the same browser profile may see saved progress. Restore and import keep saving off until you choose it again. Valid review dates are kept as unverified reminders, while earlier independent and delayed-recall claims are cleared. Continuing an imported reminder can reschedule practice without proving earlier independence.

Read aloud is optional. It uses only voices the browser reports as local to the device, starts only after a click, and stops when its tool closes. If no local voice is available, text remains available and no online voice is selected automatically. Available voices and audio quality depend on the browser and device.

The original studio has its own clearly labelled progress controls. Resetting the whole studio also clears these learning paths; resetting only these paths preserves other site data. Downloaded model and offline lesson packs have their own removal controls.

Optional local AI

These learning paths work without a model download. The existing optional local AI setup remains separate. Model wording is not allowed to award correctness, independent evidence or progress. See the model notice for the installed model’s status and licence.

Quick answers

Where are the learning-source and software notices?

This page describes the teaching content, further reading references and open-source components used by the learning tools.

Have the lessons been independently validated for learning gains?

No. The page discloses that independent educator review and validated learning-gain evidence are not claimed.