What is an uTF-8?
In one sentence: The most common way to pack Unicode numbers into bytes: 1 byte for plain English letters, up to 4 for an emoji.
A little more
UTF-8 is a way to turn Unicode numbers into bytes for saving and sending. Plain English letters take one byte, so old English text is already valid UTF-8. Letters with accents take two, most other scripts three, and emojis four. Almost every web page today uses it.
If a program reads UTF-8 bytes using a different, older code book, each byte turns into its own odd letter: that jumble is called mojibake. And if a system only has room for three bytes per character, four-byte emojis get squashed into question marks, which can't be undone.
It's a bit like… Packing a parcel: small things go in a small box, big things in a bigger one, and the label says which size it is.
See it in action
- BonusWhy do emojis turn into boxes, question marks or gibberish?X-ray any emoji to see the secret numbers and bytes inside it, solve four crime scenes where emojis seem to vanish (empty boxes, ?, mojibake and broken families) and glue your own emoji together.6 min · interactive
- BonusWhy can two passwords look the same but only one works?Drag a magnifying glass over a pretend message to reveal glowing ghost footprints left by invisible Unicode characters, read a plain-words card for what each blank does, paste any text into an X-ray, repair broken messages while keeping ordinary spaces, and solve three cases where a password, filename or file looked identical but wasn't..7 min · interactive
- BonusWho stole all the emojis?Interview four suspects in a Byte Bakery chat heist, X-ray the fake paw prints, and discover how autocorrect and a spam cleaner teamed up to frame the cat.7 min · interactive