Contribute a record
Contribute to the database
This database grows the way a shared scholarly reference grows: people who know a color tradition, or who can read the empirical literature closely, add and revise records. It is infrastructure others build on, not a finished file. A contribution is not a blog post or an opinion; it is a structured, sourced record that meets the same standard of evidence as every record already here. This page states exactly what a record needs and how a submission is reviewed and accepted.
What you can contribute
Record typesThree kinds of record, the same three kinds the collection already holds.
- A color record (identifier
GCMD-NNN): a core color concept, read across at least three cultures. This is the larger, synthesizing object. A proposal to add a seventh color (white, purple, orange, and brown are the obvious next candidates) is a color record. - A case record (identifier
GCMD-CNN): one distinct, separately citable cross-cultural finding, tied to a single study or scholarly source. Most contributions start here, because a case record is smaller, sharper, and easier to source cleanly. A strong case record can later be folded into a color record's variation apparatus. - An interview record (identifier
CTP-<code>): one consented, recorded interview run under the project's version 2 protocol, with its rating card. This is the contribution a trained interviewer makes, and it is the only kind that adds primary data. See Contribute an interview below.
What a new entry must contain
Submission standardA complete submission carries the following. A record missing any of these is returned for revision, not rejected; the fields exist so that prose and evidence cannot drift apart.
| Field | What it must state |
|---|---|
| The color | Which color the record concerns, with a hex value for the plate (for a color record), or which existing color the finding relates to (for a case record). |
| Culture, era, context | The specific culture or cultures, the period (antiquity, a named century, the present), and the setting (ritual, dress, language, market). "The West" or "Asia" alone is not specific enough; name the community and, where it matters, the date. |
| The proposed meaning | The claim itself, in one or two plain sentences: what the color is documented to mean in that context. State it as a documented reading, not as what the color "really" means. |
| The evidence | What kind of evidence stands behind the claim, graded against the methodology page: a cross-national experimental study, a single-culture study, historical scholarship, or a reference source for a well-established cultural fact. A number must come from a study, never from a reference work. |
| Primary or secondary sources | At least one resolvable source per claim: a DOI for a peer-reviewed article or a stable URL or full pullable reference for a book, chapter, or reference work. A quantitative claim must cite the study that measured it, not a summary of it. |
| Cross-cultural variation | For a color record, at least three cultures, each with its own cited claim, so the record shows how the meaning shifts rather than asserting one reading as universal. |
| What is not yet known | Any limitation: a contested finding, a thin sample, a claim that rests on a single source. Flagging the gap is part of the record, not a weakness in it. |
No fabricated survey numbers, no invented adoption or citation figures, and no copyrighted images. Ratings and quotes enter the collection only through a consented interview run under the protocol and reviewed by the curator; a contributed color or case record never carries fieldwork data of its own.
Contribute an interview
Wave 2, primary dataFrom wave 2 the fieldwork layer accepts sessions run by trained interviewers, usually students, under the published version 2 protocol: a 12 to 15 minute recorded interview with the consent script on the recording, followed by the unchanged rating card. This is the one contribution that adds primary data, and it is governed by the strictest review on the site.
| Item | Requirement |
|---|---|
| Consent on the recording | The consent script read word for word before anything else, with the contributor's attribution choice (first name and initial, initials, or code) stated on the recording. A session without it is not published. |
| The protocol | Version 2 followed in order; reserve questions only after the wrap. |
| The card | The canonical rating card, completed after the conversation and photographed before the contributor leaves. |
| The submission | The recording, the card photo, and one log row (code, background, languages, date, attribution choice, photos promised), sent to the curator within 24 hours, with files named by code and never by name. |
| Review | The curator checks the recording against the protocol, assigns the identifier, transcribes and translates the quotes, enters the card with any cleaning flags, and publishes the record in the public tier by code. The recording and transcript stay in the private tier. |
| Credit | The interviewer appears on the contributors list by first name and initial with the role Investigation (the CRediT taxonomy). The contributor's own attribution choice governs how they appear, and they may withdraw the record from the public tier at any time. |
Why open this at all: interviews are the one part of the collection that requires a person in a room with another person, and one curator's network reaches one set of backgrounds. The review step is what lets the archive grow past that network without lowering its standard. The standards the review enforces are on the standards page.
How to submit
FormatSubmit a record as plain structured text: the fields above, each filled, with every source given as a full reference plus its DOI or stable URL. Use the form below; plain text is transcribed into the structured data on acceptance. An interviewer submitting a wave 2 session uses the same form to open the conversation, and the curator replies with where to send the recording and the card photo.
Submissions go to the curator only. Your name and email are used to reply to you and are not published.
How a submission is reviewed
Editorial processEvery submission moves through the same four-stage review before it is published. The aim is not to gatekeep but to keep the standard of evidence uniform across the whole collection.
- Intake and scope. The curator checks that the submission is a color or case record in scope (the cross-cultural meaning of color), that all required fields are present, and that it does not duplicate an existing record. Incomplete or out-of-scope submissions are returned with a note on what is missing.
- Source verification. Each cited source is checked to resolve and to actually support the claim attached to it. A DOI must lead to the named study; a reference work must genuinely state the cultural fact. A claim whose source does not support it is sent back for a better source or a narrower claim.
- Evidence grading. The claim is graded against the methodology page (A for cross-national experimental work, B for single-culture or historical scholarship, C for reference-class background). Quantitative claims must reach Grade A or B; a Grade C source may support only a well-established cultural fact, never a number. The grade is recorded on the record so a reader sees how strong the evidence is.
- Editorial read and decision. The record is read for accuracy, for the visible separation of evidence from interpretation, and for house style. It is then accepted, accepted with revisions, or returned. On acceptance it is assigned a permanent identifier (the next free
GCMD-NNNorGCMD-CNN), added to the structured data, and rebuilt into the site. The identifier is never reused or reassigned.
Acceptance criteria
What gets inA record is accepted when all of the following hold:
- every claim resolves to a source that genuinely supports it;
- the evidence is graded and the grade fits the claim, with no number resting on a reference-class source;
- a color record documents variation across at least three named cultures;
- evidence and interpretation are kept visibly separate, and any limitation is stated plainly;
- the writing carries no fabricated figures and follows house style.
Once accepted, a record is citable immediately: it ships with its permanent identifier and a copy-ready "Cite this entry" block, exactly like every record already in the collection. See how to cite for the citation form, and the methodology for the full evidence grading.