Editorial Standards for Kanji.Jepang.org

1. Sources and Foundation

Kanji.Jepang.org is an Indonesia-first interactive kanji dictionary. Source data can include open-source projects, kanji lists, vocabulary assets, and generated derivatives.

Each source has different strengths and limits. We keep attribution and usage boundaries explicit whenever possible.

2. Normalization and Presentation

Raw source data is not always learner-friendly by default. We normalize labels, ordering, and grouping to produce stable reference pages.

This process is intended to improve clarity for real learners, not to hide uncertainty when data is incomplete.

• Keep label naming and section ordering consistent
• Map kanji into stable browse lanes such as JLPT, Jōyō, and radicals
• Prefer lightweight pages with clear information hierarchy
• Review output before publishing updates

3. Indonesian-First Meaning Policy

Kanji.Jepang.org follows an Indonesian-first display policy. Indonesian meanings are primary whenever available and editorially acceptable.

When Indonesian is missing, English serves as fallback support. This prevents empty labels and avoids forced low-quality translation.

• Priority 1: Indonesian meaning (ready and publishable)
• Priority 2: English fallback meaning
• Priority 3: first non-empty locale if relevant
• Priority 4: empty if data is genuinely missing

4. Limits and Interpretation Boundaries

No dictionary can fully remove ambiguity in every context. Some entries have richer data than others, and nuance may vary by usage context.

Use this site as a practical reference aid, not as a legal, medical, or official translation authority.

• Coverage depth can vary by entry
• Short glosses may not capture every nuance
• Vocabulary and sentence coverage may change as data improves
• Large corpus size does not imply uniform field completeness

5. Corrections, Verification, and Updates

Specific correction reports are strongly preferred: include URL, affected section, proposed fix, and supporting references when possible.

Reports are reviewed to identify whether the issue comes from data source, content normalization, or page-level presentation.