This page turns atlas tools into short classroom exercises. It is meant for
Sanskrit students who can read a headword and need practice deciding what a
dictionary record proves.
- Evidence: atlas lookup/dossier pages, dictionary chooser guidance, learner
reading layer, and linked
csl-guides lessons.
- Limitations: teaching route only; examples are prompts, not final translations
or philological decisions.
- Validation: linked atlas pages are checked by
npm run build; exercises use
existing public tools and do not introduce generated data.
- Owner repo:
csl-atlas.
- Next use: assign one lesson, require a cited dictionary record, and ask the
student to name the evidence level.
| Lesson |
Question |
Tool path |
Student output |
| 1. First dictionary |
Which dictionary should I open first for this word? |
Which dictionary? -> Reader Lookup |
One first-stop dictionary and one reason. |
| 2. Coverage is not meaning |
How many dictionaries know this lemma, and what does that not prove? |
Reader Lookup -> Lemma dossier |
Coverage count plus one limitation. |
| 3. Source trail |
Does the gloss depend on a named source or only dictionary tradition? |
Citation apparatus -> dictionary page |
One cited source link or a note that no source is explicit. |
| 4. Dictionary disagreement |
Where do two dictionaries differ in grammar, sense, or convention? |
Lemma dossier -> Gender conflicts |
Two records and a neutral description of the disagreement. |
| 5. Corpus caution |
Does dictionary evidence equal corpus attestation? |
Learner layer -> Lemma dossier |
One sentence separating dictionary record from corpus signal. |
| 6. Do not correct too quickly |
Is an odd spelling likely an error, variant, or apparatus note? |
Researcher Dashboard -> SanskritSpellCheck |
A cautious classification and the evidence needed before filing. |
| Set |
Words |
Why this set works |
| Basic concepts |
agni, dharma, yoga, rAma |
Familiar enough that students can focus on evidence instead of vocabulary. |
| Grammar bridge |
gam, BU, dA, vid |
Shows why roots, classes, and dictionary lemmas are not the same surface. |
| Source-trail practice |
mokza, medas, mfga, Siva |
Encourages source links and dictionary-to-dictionary comparison. |
| Caution set |
divaraTa, akalkala, cApaqa |
Useful for discussing variant, correction, and source-check evidence. |
| Rule |
Classroom use |
| Cite records, not summaries |
The atlas guides the path; the dictionary record carries the claim. |
| Keep both sides of disagreement |
Do not force MW/AP/PWG/VCP/SKD into one artificial answer. |
| Say what kind of evidence it is |
Use observed, derived, inferred, or reviewed in every note. |
| Stop before correction |
A suspicious form needs source checking before it becomes a correction. |