Transparent by design
Learning methodology
Turkish Quest is the first course inside Language Hub. It is designed to make starting easy without pretending that recognition games alone teach a language. Its three modes collect different kinds of learning evidence and reuse the same material in progressively harder contexts.
Study
Construct and interpret Turkish inside practical roleplaying scenes.
Lightwork
Maintain memory through low-pressure recognition and active recall.
Fun
Passive Turkish exposure through readable turn-based strategy and ambient dialogue.
What mastery means
Each vocabulary item stores exposures, correct and incorrect answers, average response time, a 0–100 mastery value, and the next review time. Evidence is separated into recognition, recall, speed and construction. Selecting an obvious answer once therefore cannot prove that a learner can produce the word in a sentence.
| Evidence | Example | What it demonstrates |
|---|---|---|
| Recognition | Turkish → English multiple choice | The form is familiar in context. |
| Recall | English cue → Turkish answer | The learner can retrieve the form. |
| Speed | Correct response in under two seconds | Recognition is becoming fluent. |
| Construction | Build a sentence or ordered command | The learner can use forms together. |
Review scheduling
Weak items return quickly. Learning items return after roughly one day, strong items after several days, and mastered items after about a week. A normal vocabulary round aims for about 60% due or weak material, 25% familiar material and 15% unseen material, backfilling when a bucket is too small.
Scoring is not proficiency
Scores reward correct and sometimes fast play. They are motivational game feedback, not CEFR certification, a diagnostic exam, or evidence of conversational fluency. “A1” labels the intended content range, not a credential earned by completing a mission.
Controlled variation, not free-form invention
Command Sprint runs are assembled from hand-authored definitions. A seed chooses one validated variation per mechanic—including levelled arithmetic—and shuffles visible choices; it does not invent translations or suffix chains. Wrong answers pause on the exact command, correction and explanation until the learner continues. Lingo League uses a fixed, validated pool of bilingual move names, creatures and effects; its seven arenas are cosmetic, while exact numbers, AP costs and rival intent remain visible so strategy never depends on translation. Midnight Courier reshuffles hand-authored encounters, risks and upgrades. Its visible street signal gives the player something fair to read before committing. In both Fun games, Turkish remains passive exposure rather than becoming a quiz. All four Study missions are hand-authored.
Scene missions vary only reviewed slots—such as a selected drink or amount—and explicitly supported morphology. Free-form sentence generation is excluded because grammatical possibility is not the same as natural Turkish.
Content status and limitations
- • The current build contains 138 starter entries, a supplied 1,104-entry development list, a 1,100-word unlock path and 54 phrase records; it does not claim a verified frequency ordering.
- • Structural validation checks IDs, required fields, game answer integrity, mechanic coverage and route buildability.
- • Automated checks cannot certify idiomatic naturalness, register, translation nuance or pronunciation.
- • Public educational marketing still requires qualified Turkish editorial review and confirmed reusable rights for any corpus-derived list.