Home to Document Object Model | Up to Core
Within the Document Object Model, all string data is encoded using UTF-16. Note that the implementation may use a differently named binding where UTF-16 string are already implemented, e.g. both ECMAScript and Java bind DOMString to the String class.
String comparisons within the model are strict binary compares of each 16-bit character position. However, note that the implementation may have folded, for example, all HTML entity names to uppercase as part of creating the DOM structures.