#Feature Examples and Integration Scenarios

This guide is intentionally practical and step-by-step.

It has two sections:

1. Feature examples: how to use each module directly
2. Integration scenarios: how modules work together in real workflows

---

#Part 1: Feature Examples

#1. Phonology: Build a usable sound system in 10 minutes

Steps:

1. Open Phonology -> Inventory.
2. Load a preset close to your target style.
3. Add/remove 3-5 phonemes to make it unique.
4. Open Romanization and define key mappings.
5. Open Phonotactics and set (C)V(C).

Result: lexicon entries can generate IPA reliably.

#2. Lexicon: Build a 30-word core vocabulary

Steps:

1. Open Lexicon.
2. Use the ⚡ quick-entry popup for fast input.
3. For each entry, fill spelling + POS + gloss.
4. Tag core entries as core.
5. Run QC and fix all error-level findings.

Result: enough lexical coverage for auto-gloss and sandbox tests.

#3. Word Generator: Expand vocabulary safely

Steps:

1. Open Word Generator.
2. Set count to 80 and syllable range to 1~3.
3. Generate candidates and remove noisy forms.
4. Assign POS to selected rows.
5. Import selected candidates to lexicon.

Result: fast and controlled vocabulary growth.

#4. Grammar: Create a minimal inflection system

Steps:

1. Open Grammar -> Dimensions.
2. Add Number (SG/PL), Tense (PRES/PAST).
3. Open Grammar -> Inflection.
4. Add PL -> -en, PAST -> -ka.
5. Validate with Inflection Test.

Result: morphology is operational for sentence-level use.

#5. Sandbox: Validate rule behavior quickly

Steps:

1. Open Translation Sandbox.
2. Enter star-PL shine-PAST in gloss mode.
3. Run conversion.
4. Verify conlang form, gloss line, IPA line.
5. If wrong, fix lexicon or grammar and re-run.

Result: tight edit-test loop.

#6. SCA: Apply historical changes with control

Steps:

1. Open Sound Changes.
2. Create ruleset Stage A.
3. Add one rule and test on single words.
4. Switch to batch preview and inspect all pages.
5. Apply when results are acceptable.

Advanced step:

6. Enable Allow SCA edits on selected grammar rules.
7. Re-run batch preview and apply scoped grammar changes.

Result: controlled evolution of lexical and selected grammar forms.

#7. Corpus: Auto-gloss and evolve text via diff

Steps:

1. Open Corpus and create a text.
2. Fill original text + free translation.
3. Click Auto-gloss.
4. Review report (total/auto/pending/unresolved).
5. Resolve pending suggestions.
6. Click Preview & Apply SCA to Corpus.
7. Deselect unwanted changes in diff table.
8. Apply selected changes.

Result: corpus updates are reviewable and traceable.

#8. Family Tree: Derive and maintain daughter languages

Steps:

1. Open Family Tree.
2. Select parent and click Derive.
3. Name the daughter language.
4. In daughter language, configure SCA.
5. Use Pull Sync to import new parent words with evolution.

Result: stable genealogical workflow.

#9. Export: Publish and collaborate

Steps:

1. Export PDF for readers.
2. Export Excel for collaborators.
3. Export LLM Prompt for AI-based testing.
4. Keep CSV snapshots for backup/import.

Result: complete sharing package.

---

#Part 2: Integration Scenarios

#Scenario A: Zero-to-usable language loop

Flow:

1. Phonology setup.
2. Word generation + lexicon import.
3. Grammar dimensions/rules.
4. Sandbox validation.
5. Corpus auto-gloss verification.

Why it works: each stage validates the previous one.

#Scenario B: Parent-to-daughter evolution pipeline

Flow:

1. Stabilize parent lexicon/grammar.
2. Derive daughter language.
3. Define daughter SCA rules.
4. Pull sync new parent words.
5. Optionally apply SCA to selected grammar rules.

Why it works: clear lineage with controlled divergence.

#Scenario C: Corpus-driven correction loop

Flow:

1. Run corpus auto-gloss.
2. Check pending/unresolved counts.
3. Fix lexicon and grammar weak points.
4. Re-run auto-gloss.
5. Compare report changes.

Why it works: corpus quality metrics reveal real system gaps.

#Scenario D: Safe corpus evolution with diff

Flow:

1. Validate SCA on single/batch preview.
2. Open corpus SCA diff preview.
3. Keep only intended changes selected.
4. Apply selected changes.
5. Let auto-gloss rerun and review output.

Why it works: review-before-apply prevents destructive bulk edits.

#Scenario E: Export-feedback-iterate cycle

Flow:

1. Export PDF/Excel/LLM Prompt.
2. Collect reader/collaborator/AI feedback.
3. Update lexicon/grammar/corpus accordingly.
4. Re-export next iteration.

Why it works: closes the loop between design and real usage.