What 3 means
Switch a Japanese input method to full-width alphanumeric mode.
On Windows, Microsoft Japanese IME, the Japanese text input tool, offers full-width alphanumeric mode; F9 converts selected text to full-width alphanumeric, and F10 to half-width. On Mac, Japanese input settings include “Full-width numeral characters”; type the number in that mode. As of 2026, these are input-mode choices, not font styling.
Why is 3 wider than 3?
It is the same number in a wider compatibility form.
In older mixed-width East Asian text systems, characters in double-byte encodings (two bytes per character) occupied two display cells, while ASCII (the basic Latin character set used in computing) occupied one. A wide digit could line up with Japanese kana and kanji. Unicode preserves both forms so older text can retain the characters it used. The Japanese terms are zenkaku (全角, full-width) and hankaku (半角, half-width); the same width distinction also appears in Chinese and Korean computing. The Japanese standardization trail reaches JIS C 6226, a Japanese character-encoding standard, in January 1978; it was renamed JIS X 0208 in 1987. The form developed through encoding standards; it was not a new numeral.
Will 3 work in a form that accepts 3?
Not always: it has the same value, but it is a different character.
Unicode compatibility normalization, a text-cleanup step that folds some alternate forms, can convert 3 to ordinary 3. A strict identifier, database field, or search can still distinguish them until that cleanup happens.
Neutral and technical: 3 marks a character-width or input-mode choice; its numeric value stays three.
An ASCII-only form can reject 3; use ordinary 3 when the field requires ASCII.
Don't confuse it with
1Questions people ask
1Does 3 always take exactly two character spaces on screen?
No. Unicode’s East Asian Width property gives a default width relationship; the font and layout engine determine how much space the glyph actually occupies.
How to type 3
3\FF13Codes and encoding1 code points
A visible symbol can contain several code points. JS .length counts UTF-16 units; UTF-8 counts bytes.
