Glossary

The terms used across this archive, and what they mean in Unicode.

Codepoint: A single number identifying a character in Unicode, written as U+1F600. Emoji are often more than one codepoint joined together — 👩‍💻 is three. Every record here is keyed on its codepoints rather than its name, because Unicode renames characters but never renumbers them.
RGI: Recommended for General Interchange — Unicode's list of the emoji and sequences vendors are expected to support, published as emoji-test.txt. Characters outside it may still display somewhere, but no vendor is obliged to draw them.
ZWJ sequence: A zero-width joiner (U+200D) is an invisible character that glues emoji into one glyph. 👩 + ZWJ + 💻 renders as 👩‍💻 on systems that support it, and as two separate emoji on those that do not — which is why an unsupported sequence appears to “split”.
Variation selector: U+FE0F requests the colourful emoji form of a character, and U+FE0E the plain text form. They change presentation, not identity, so this archive strips them when keying a record — ❤️ and ❤ are one emoji, not two.
Skin tone modifier: Five characters (U+1F3FB–U+1F3FF) based on the Fitzpatrick dermatological scale. Appended to an emoji that supports them, they change its skin tone. They are components rather than emoji: on their own they render as colour swatches.
Regional indicator: Twenty-six letter symbols (🇦–🇿) that form flags in pairs: 🇯 + 🇵 gives 🇯🇵. This is why flags are two codepoints and why an unsupported one shows as two letters instead.
Tag sequence: Subdivision flags — England, Scotland, Wales — are built from a waving black flag followed by invisible tag characters spelling the region code. They are the only emoji that encode text inside themselves.
Keycap sequence: A digit, # or *, followed by U+FE0F and the combining enclosing keycap U+20E3, rendering as a key: 7️⃣. The bare digit is a component, not an emoji.
Qualification: Whether a sequence carries every variation selector Unicode expects. Fully-qualified is the canonical form; minimally-qualified and unqualified are shorter spellings that usually still render. Component marks a building block rather than a standalone emoji.
Emoji version vs Unicode version: Two separate numbering schemes that only converged at 11.0. Emoji 5.0 shipped inside Unicode 10.0; Emoji 0.6 and 0.7 label characters that predate the emoji specification entirely. Both are recorded here per emoji.
CLDR: The Common Locale Data Repository, Unicode's localisation project. It supplies each emoji's short name and search keywords in many languages — the source of the translated names on every page here.
Shortcode: A :colon_syntax: name for typing an emoji in chat. They are conventions of individual platforms, not Unicode, so they differ between GitHub, Slack and Discord — see shortcodes.
Private use area: Codepoint ranges Unicode leaves for anyone to assign privately. Japanese carriers put their original emoji here before Unicode encoded them, which is why some historic characters render only on the hardware they came from.
Vendor artwork: Unicode defines what an emoji means, never how it looks. Each vendor draws its own set, which is why the same codepoint differs across Apple, Google and Samsung, and why the comparison grid on every page here exists.