Coverage report for the top 1000 concepts per language #21

New issue

Open

opened 2026-09-26 13:58:34 +00:00 by belvedere · 0 comments

belvedere commented

2026-09-26 13:58:34 +00:00

Collaborator

Objective: Know whether "the first 1000 words" is actually achievable for a given language
before promising it in the UI.

Files:

Create: lib/mix/tasks/f1000.coverage.ex
Create: lib/first_thousand_words_web/live/language_coverage_live.ex (admin view)

Steps:

Per language: count words imported, count words with ≥1 concept link, count concepts with
≥1 word in both languages of a pair, and list the worst-covered concepts.
Print a table from a mix task.
Surface the same numbers on an admin page.

Verify:

mix f1000.coverage
# language | words | linked | paired | worst gaps

Notes: this is the gate for enabled? on a language. Do not enable a pair with terrible
coverage; the practice UI will feel broken.

**Objective:** Know whether "the first 1000 words" is actually achievable for a given language before promising it in the UI. **Files:** - Create: `lib/mix/tasks/f1000.coverage.ex` - Create: `lib/first_thousand_words_web/live/language_coverage_live.ex` (admin view) **Steps:** 1. Per language: count words imported, count words with ≥1 concept link, count concepts with ≥1 word in *both* languages of a pair, and list the worst-covered concepts. 2. Print a table from a mix task. 3. Surface the same numbers on an admin page. **Verify:** ``` mix f1000.coverage # language | words | linked | paired | worst gaps ``` **Notes:** this is the gate for `enabled?` on a language. Do not enable a pair with terrible coverage; the practice UI will feel broken.

belvedere added this to the M2 — Lexicon model & data pipeline milestone

2026-09-26 13:58:34 +00:00

belvedere added the

labels

2026-09-26 13:58:34 +00:00

belvedere referenced this issue

2026-09-26 14:02:05 +00:00

Coverage report for the top 1000 concepts per language #73