Methodology

Data Sources

The Whispering Lore dataset draws from a wide range of sources across folklore studies, anthropology, and comparative mythology:

• Published folklore collections and academic anthologies (e.g., Asbjørnsen & Moe, the Brothers Grimm, Y.T. Kawamura)
• Regional mythographic surveys and encyclopedias of world mythology
• Peer-reviewed journal articles in folklore studies and cultural anthropology
• Primary oral tradition transcripts where available in published form
• Indigenous cultural heritage materials published with community permission

Each entry includes a source field naming the specific reference used. Entries marked source_quality: "verified" have been cross-checked against at least two independent scholarly sources. Entries marked "unknown" are pending verification.

Inclusion Criteria

A creature or story is included if it meets at least one of:

• Documented in a published folklore or mythology source
• Part of an identifiable cultural tradition with oral or literary transmission
• Referenced in multiple independent sources (for cross-cultural entries)

Entries derived from modern fiction, video games, or film-only creations are excluded unless they have roots in traditional folklore. Contemporary legends and urban folklore are included where well-documented.

Data Structure

Each creature entry contains 25+ fields organized into categories:

• Identity: name, slug, alternative names, type classification
• Geography: country, region, cultural tradition
• Description: appearance, behavior, habitat, cultural significance
• Relations: related creatures, parent species, associated stories
• Metadata: source quality, last updated, verification status

Editorial Standards

• Entries are written in a neutral, descriptive tone that respects the source culture
• Alternative names and transliterations are preserved alongside primary names
• Regional and cultural variants are noted where documented
• Ambiguous or contested folklore is flagged rather than flattened into a single narrative
• AI-assisted text generation has been used for some summary expansions and is clearly marked in the metadata; such entries are explicitly tagged and are being progressively replaced with verified content

Verification Process

Verification is an ongoing process. Entries pass through these stages:

• Stage 1 — Initial compilation: Data collected from published sources, formatted into the standard schema
• Stage 2 — Cross-reference: Creature-to-story and story-to-creature references checked for accuracy
• Stage 3 — Source verification: Named sources checked against library catalogs and academic databases
• Stage 4 — Expert review: Subject-matter folklorists review entries for cultural accuracy

As of the current dataset version, 99.0% of entries (5,331 of 5,387) have reached Stage 3 or above. The remaining 1.0% are at Stage 1 or Stage 2.

Current source quality distribution (creatures):

• Expert: 608 entries — reviewed by subject-matter specialists
• Verified: 734 entries — independently confirmed against primary sources
• Researched: 2,292 entries — traced to named academic or archival sources
• Well-documented: 28 entries — supported by multiple independent sources
• Primary: 4 entries — sourced from first-hand ethnographic fieldwork
• Excellent: 2 entries — fully cited with peer-reviewed references
• All 3,668 creatures are at Stage 3 or above.

Current source quality distribution (stories):

• Expert: 22 entries — reviewed by subject-matter specialists
• Researched: 1,647 entries — traced to named academic or archival sources
• Good: 2 entries — from general references
• Fair: 48 entries — from general references or secondary sources
• 1,669 of 1,719 stories (97.1%) are at Stage 3 or above.

Versioning

The dataset is versioned with semantic versioning (MAJOR.MINOR.PATCH). Each version's date and change summary are recorded in the dataset manifest. The current version is displayed in entry footers. Major data updates trigger a new version release.

How to Contribute

If you have corrections, additions, or source materials to share:

• Email: mafolsson@tutamail.com
• Include the specific creature or story name, your source, and the correction
• For academic contributors: citations in MLA or Chicago style preferred

Limitations

• This is a living archive — completeness varies by region based on available source materials
• Some cultures are underrepresented due to limited published resources in English and Nordic languages
• The dataset is primarily text-based; illustrations are planned but not yet implemented
• Oral traditions are represented through published transcripts, which may not fully capture performance context

Last updated: 2026-07-20