Mojibake

Mojibake

Incorrectly interpreted character encoding

• 
  Handwritten mojibake (originally Russian: Россия, Москва,) on a postal envelope from Paris, 2002.
• 
  This example was made in Russia, 2011.
• 
  Сyrillic “моя страна” in mojibake. A Latin-1-only Sony device plays a Bulgarian song. 2012.
• 
  A thin-client browser, Windows-1252 only, on a page with a text in Shift JIS. Some Japanese and the © character are present. c. 2001.
• 
  This is neither Japanese nor Chinese. Only a byte-oriented stream viewed as UTF-16 due to limitations of Microsoft’s IsTextUnicode().
• 
  Cool Mozilla coders know how to interpret Internet Media Types better than IETF; the “text” is in the left-bottom corner. The bug was introduced c. 2012.
• 
  Centericq, running in a Latin-1 terminal, shows a UTF-8-encoded nick.

Failed encoding/decoding

• 
  The font engine of a Kodak printer failed.
• 
  Javanese replacement question marks.

Missing characters or inappropriate glyphs

• 
  A standard Universal Character Set replacement box with a code point in hexadecimal.
• 
  Incompetent vector graphics production. Labels converted to <path> using a font lacking comma-bearing Romanian letters in Constanța and Ploiești. Rectangles are exported in place of those.
• 
  Japanese language. The font has a short supply of Kanji characters.
• 
  Not mojibake, strictly speaking, but a typeface inappropriate for Russian language. Made in Serbia