The History of Unicode Emoji: From Docomo to Unicode 16.0

Before Emoji: The Emoticon Era

Before emoji existed, we had emoticons — combinations of punctuation marks to represent facial expressions — and the impulse goes back further than most people realize.

On September 19, 1982, Scott Fahlman posted a message on a Carnegie Mellon bulletin board proposing :-) and :-( to distinguish jokes from serious posts. The exact message was recovered from backup tapes in 2002 by CMU researcher Mike Jones. But even Fahlman wasn't first — Abraham Lincoln's 1862 transcript may contain a ;) (though scholars debate whether it's a typo), and *Puck* magazine printed typographic "faces" in 1881.

Japanese users developed their own tradition called kaomoji, read face-on rather than sideways: (^_^) for happy, (T_T) for crying, (╯°□°)╯︵ ┻━┻ for table-flipping frustration. Kaomoji were richer than Western emoticons — they had hundreds of variations by the mid-90s — and they reflected a visual communication culture that would directly give birth to emoji.

The desire to add emotional context to plain text is as old as digital communication itself. Emoji didn't create this need. They answered it.

1999: The Birth of Emoji in Japan

The word "emoji" comes from the Japanese 絵 (*e*, "picture") + 文字 (*moji*, "character"). Despite the coincidence, it has nothing to do with the English word "emotion" — though the overlap is poetically perfect.

Shigetaka Kurita and NTT Docomo

In 1999, Shigetaka Kurita, a 25-year-old artist working on NTT Docomo's i-mode team, created the first set of 176 emoji. Each was a 12×12 pixel grid — 144 pixels total to convey an idea. They were designed for Docomo's i-mode mobile internet platform, which had a 48-character limit for messages. Kurita's insight: if you can't write much, draw instead.

He drew inspiration from manga, where visual symbols convey emotion and action (sweat drops for nervousness, bulging veins for anger), and from weather forecasts and street signs. His original set included weather icons ☀️🌧️, traffic signs, technology symbols, and simple facial expressions.

The original 176 emoji are now in the permanent collection of the Museum of Modern Art (MoMA) in New York, acquired in 2016. If you visit, the contrast between those tiny pixel grids and the polished emoji we use today is striking — like looking at cave paintings next to oil portraits.

Competing Carrier Sets

After Docomo's success, rival carriers SoftBank and au (KDDI) developed their own incompatible emoji sets. SoftBank had around 480 emoji by 2003; KDDI had its own set of similar size. A ❤️ on Docomo might show up as a blank square on SoftBank. Three carriers, three codebooks, zero interoperability.

This carrier-specific mess in Japan was a preview of the cross-platform rendering chaos that would later plague emoji globally. But it also proved something crucial: people *really* wanted to communicate with pictures.

2007-2009: Apple and Google Enter the Picture

Apple's Secret Emoji Keyboard

When Apple launched the iPhone in Japan in November 2008 (iPhone OS 2.2), they included a hidden emoji keyboard to compete with Japanese feature phones that had native emoji support. The keyboard was officially Japan-only, but tech-savvy users worldwide discovered they could enable it by downloading a fake Japanese app or editing a settings plist. Blog posts with titles like "HOW TO GET EMOJI ON YOUR IPHONE" exploded across the early smartphone web.

This quiet leak was a turning point. People were going through absurd workarounds to access a keyboard Apple didn't even want them to use. The demand signal was deafening.

Apple made the emoji keyboard officially available worldwide in iOS 5 (October 2011), and the floodgates opened.

Google's Push for Standardization

Google engineers Mark Davis) (who co-founded the Unicode Consortium in 1991) and Kat Momoi, along with Markus Scherer and Darick Tong, recognized both the potential and the problem. In 2007, they submitted a formal proposal to the Unicode Consortium to encode emoji as part of the Unicode Standard — document L2/07-257.

The proposal argued that emoji had become essential for millions of users and that standardization was necessary for interoperability. It was not an easy sell. Some Consortium members felt that standardizing "cute pictures" was beneath the dignity of a character encoding standard designed for the world's writing systems. The counter-argument was simple: people are using these whether you like it or not. Ignore them and the fragmentation gets worse.

2010: Unicode 6.0 - Emoji Go Global

In October 2010, Unicode 6.0 was released, including 722 emoji characters. This was the moment emoji officially became part of the global character standard used by every major operating system, browser, and digital platform.

The initial Unicode emoji set was largely based on the union of the three Japanese carrier sets — characters that existed across Docomo, SoftBank, and KDDI. This is why the original Unicode emoji set has a distinctly Japanese flavor: 🍣 sushi, 🍱 bento boxes, 🍙 rice balls, cultural items like 🎎 dolls and 🏯 castles, and symbols specific to Japanese life like 🔰 (the *shoshinsha* beginner driver mark).

What Standardization Actually Means

Here is what Unicode standardization does and does not do. Unicode assigns each emoji a unique code point (a number like U+1F600 for 😀). Every platform agrees on *what* the emoji represents. But Unicode does not dictate *how* each emoji looks — that's up to individual vendors.

This is why the same emoji can look quite different on Apple, Google, Samsung, Microsoft, and other platforms. They all agree that U+1F600 is "Grinning Face," but each draws it differently. A 2016 study by GroupLens Research at the University of Minnesota ("Blissfully Happy" or "Ready to Fight") found that people's sentiment interpretations of the same emoji varied by up to 2 points on a 5-point scale across platforms. Samsung's old designs were the worst offenders — their 😬 grimacing face looked almost cheerful.

2014-2015: Diversity and Mainstream Adoption

Skin Tone Modifiers

Unicode 8.0 (June 2015) introduced skin tone modifiers based on the Fitzpatrick scale, a dermatological classification system developed in 1975 by Harvard's Thomas B. Fitzpatrick. For the first time, human emoji could represent a range of skin tones rather than defaulting to Simpsons-yellow.

The five modifier code points (U+1F3FB through U+1F3FF) map to Fitzpatrick types I-II, III, IV, V, and VI. A base emoji like 👋 followed by a modifier character produces the combined result; older platforms show the base emoji plus a colored square. Apple shipped skin tone support in iOS 8.3 (April 2015), ahead of the official Unicode release.

Usage data is interesting: according to Unicode Consortium's own emoji frequency data, the default yellow tone remains the most-used option across all platforms, suggesting many users treat it as neutral rather than picking a skin tone.

Gender Representation

Gender diversity in emoji expanded through the ZWJ (Zero Width Joiner) mechanism. Person + ZWJ + object = compound emoji. Woman + ZWJ + microscope = 👩‍🔬 (woman scientist). This approach was proposed by Google and Apple in 2016 as Emoji 4.0, adding 100+ profession emoji in male and female forms.

Unicode 10.0 (2017) and 11.0 (2018) expanded further, and Emoji 12.1 (2019) introduced gender-neutral options — the "person" base without male/female specification.

The ZWJ Mechanism: How Complex Emojis Work

ZWJ sequences are one of the cleverest hacks in the emoji system. More people should understand how they work, because they're hiding in plain sight on your keyboard.

A ZWJ is an invisible Unicode character (U+200D) — originally from Arabic/Hebrew text shaping — repurposed to tell the rendering system to combine adjacent characters into a single glyph. This mechanism enables:

Family Emojis

👨 + ZWJ + 👩 + ZWJ + 👧 + ZWJ + 👦 = 👨‍👩‍👧‍👦 (family with father, mother, daughter, son)

Professional Emojis

👩 + ZWJ + 🚀 = 👩‍🚀 (woman astronaut)

Flag Sequences

The rainbow flag 🏳️‍🌈 is actually: White flag (🏳️) + ZWJ + Rainbow (🌈)

Compound Expressions

The trans flag 🏳️‍⚧️ uses the same mechanism: White flag + ZWJ + trans symbol

The elegance of ZWJ is backward compatibility. On platforms that don't support a particular combination, the components display separately rather than as a blank square. An old system shows "white flag, rainbow" instead of the pride flag — not ideal but still understandable.

The downside: combinatorial explosion. As of Emoji 16.0, there are over 1,300 ZWJ sequences in the recommended set, and not every platform supports all of them. You can theoretically join *anything* with ZWJ — most combinations just won't render as intended.

The Emoji Proposal Process

Ever wonder how new emoji get added? The process is more formal than most people realize — we cover it in depth in our emoji proposal process article.

Who Decides

The Unicode Consortium's Emoji Subcommittee reviews proposals and recommends additions. The full Unicode Technical Committee (UTC) votes on final inclusion. Full voting membership costs $18,000/year — Apple, Google, Microsoft, Meta, Netflix, and the governments of India and Oman are all members.

The Proposal Requirements

Anyone can submit a proposal (literally — a teenager from Saudi Arabia proposed the 🧕 hijab emoji). But it must meet specific criteria:

Expected usage level: Will people actually use it? Google Trends data required. Image distinctiveness: Can it be recognized at 18×18 pixels? Completeness: Does it fill an obvious gap in an existing category? Not a fad: Once a code point is assigned, it stays forever. No fidget spinners.

Notable Proposals and Debates

Apple famously changed their 🔫 pistol emoji from a realistic revolver to a water gun in iOS 10 (2016). Samsung, Google, and Microsoft followed within two years — a rare case of one vendor's design choice becoming the de facto standard.

The 🧕 hijab emoji was proposed by 15-year-old Rayouf Alhumedhi from Saudi Arabia, who argued that hijab-wearing women deserved representation. Approved in Unicode 10.0 (2017).

The interracial couple emoji required nested ZWJ sequences with individual skin tone modifiers on each person — technically complex, socially important, and approved in Emoji 12.0 (2019).

Unicode Versions and Major Additions

The major milestones, with standout additions:

|---------|------|-----------|------------|

| 6.0 | 2010 | 722 | The original set. Most "classic" emoji. |

| 7.0 | 2014 | ~250 | 🖕 middle finger (controversial), new faces |

| 8.0 | 2015 | ~41 | Skin tone modifiers, sports variations |

| 9.0 | 2016 | 72 | 🤦 facepalm, 🤷 shrug, 🥑 avocado |

| 10.0 | 2017 | 56 | 🧕 hijab, 🦖 T-Rex, 🥟 dumpling |

| 11.0 | 2018 | 157 | 🦸 superhero, redheads, 🥿 flat shoe |

| 12.0 | 2019 | 59 | 🦩 flamingo, 🦥 sloth, accessibility emoji |

| 13.0 | 2020 | 55 | 🥲 smiling tear, 🥷 ninja, 🏳️‍⚧️ trans flag |

| 14.0 | 2021 | 37 | 🫠 melting face, 🫡 salute, 🫦 biting lip |

| 15.0 | 2022 | 31 | 🫨 shaking face, 🩷 pink heart, 🪽 wing |

| 15.1 | 2023 | 118 | Mostly new ZWJ sequences (directional people) |

| 16.0 | 2024 | 8 | 🫩 face with bags under eyes, root vegetable, splatter |

Notice the trend: from 722 in the first release to single digits by 2024. The emoji set is approaching completeness — or at least, the low-hanging fruit is picked.

Technical Challenges and Solutions

Rendering Across Platforms

An emoji sent from an iPhone looks different on a Samsung phone, which looks different in Chrome on Windows. We have a whole article on iPhone vs Android differences if you want the gory details.

Apple's emoji are detailed and three-dimensional. Google evolved from their beloved blob era (2013–2017, RIP) to a cleaner style. Microsoft went full 3D with Fluent Emoji in 2022 — and open-sourced the whole set. Samsung does its own thing, sometimes bafflingly so.

Color Emoji vs. Text Presentation

Some characters exist as both text and emoji. The heart symbol ♥ is a text character; ❤️ is the emoji version. The distinction is controlled by variation selectors - invisible Unicode characters that tell the system which presentation to use.

This dual existence occasionally causes rendering issues where a heart appears as a small monochrome symbol instead of the expected colorful emoji, or vice versa.

Emoji in Non-Display Contexts

Emojis create challenges for search engines, screen readers, data processing, and text analysis. Each emoji has a text description that screen readers use, but processing text that mixes scripts, emojis, and ZWJ sequences is technically complex.

For developers working with emoji-containing text, handling variable-length characters (some emojis are single code points; ZWJ sequences can be many code points displayed as one glyph) requires careful string handling.

The Cultural Impact

Emoji have transcended their origins as cute additions to Japanese phone messages:

Over 5 billion smartphone users have access to emoji. Emojipedia reported that 😂 was the most-used emoji globally every year from 2015 to 2021.
In 2015, Oxford Dictionaries named 😂 their Word of the Year — the first time a pictograph won.
Courts in Israel, Canada, and the US have ruled on cases where emoji usage in texts and contracts created legal ambiguity. A Canadian judge ruled in 2023 that a 👍 thumbs-up emoji constituted acceptance of a contract (South West Terminal Ltd. v. Achter Land).
Emoji are studied in linguistics, psychology, and sociology — over 1,000 peer-reviewed papers reference emoji communication.
MoMA's 2016 acquisition of Kurita's original 176 emoji cemented them as a recognized design artifact.

What Comes Next

The emoji ecosystem continues to evolve:

Personalization

Apple's Memoji and Samsung's AR Emoji point toward personalized avatars rather than universal symbols. This creates tension with the standardization that made emoji successful — if everyone has their own face emoji, we lose the shared vocabulary.

Animated Emoji

Telegram pioneered animated emoji in 2022; Discord and Apple's iMessage followed. These animate the *existing* standard set rather than adding new characters, keeping them compatible with the Unicode system.

Slower Growth

From 722 new emoji in 2010 to just 8 in 2024. The Emoji Subcommittee has acknowledged the set is approaching completeness. Future proposals face a higher bar.

Where This Is Heading

From 176 pixelated images on a Japanese mobile phone to a global standard of over 3,700 characters used by billions — the emoji story is one of the wildest success stories in communication design.

What started as a workaround for a 48-character message limit became a visual language that transcends borders, writing systems, and cultures. The technical infrastructure (Unicode standardization, ZWJ sequences, variation selectors) is elegant in its complexity. The cultural impact — from MoMA to courtrooms to Oxford Dictionaries — is profound.

Every time you tap ❤️ or 😂, you're participating in a communication system that's barely 25 years old. And the story is still in its early chapters.