Editorial method

How this Korean name reference works

The database is designed around evidence boundaries: pronunciation is not meaning, legal character status is not popularity, and a ranked-name dataset is not a complete population census.

1. Hangul is the primary name form

Given-name entities are organized by Hangul spelling. Romanization is a navigation aid, not a pronunciation recording and not a claim that one Latin spelling is uniquely correct.

2. Hanja meaning stays at character level

Where Hanja choices are available, the site displays the meaning context of each character. It does not combine two glosses into a confident story about what a person’s parents intended.

3. Usage figures keep their date and scope

Name popularity comes from Korean Supreme Court top-name statistics. National Top-20 rows are kept separate from regional Top-20 attestation rows; regional counts are never relabelled as nationwide totals, and absence from a ranked slice is not treated as zero use.

4. Possible is not the same as attested

The name finder can analyse any 1–5-syllable Hangul candidate against the mapped legal-Hanja reading inventory. That on-demand analysis does not create a public name entity, prove naturalness, prove actual use or guarantee registration acceptance. Standalone pages remain evidence-gated.

5. Indexing is quality-gated

Standalone pages are promoted only when they meet the current evidence and usefulness threshold; thinner or purely theoretical candidates remain discoverable through internal tools without becoming search landing pages.

6. Unicode enriches; the Court determines legal status

Unihan dictionary glosses, Korean reference properties and stroke data are supplemental character metadata. They never override the Court-derived legal-character and reading layer.

7. Historical surname evidence remains historical

Surname and bon-gwan figures keep the census year visible. Older population counts are not silently relabelled as current estimates.