An element content type consisting of parsed characters (i.e., entity references) but no elements. Entity references included in this data will be resolved. See also CDATA, parsed-character data.

parsed-character data

Any character data that should be checked by the XML processor for entity references. These references are resolved and their replacement text parsed recursively to replace all entity references. See also PCDATA.


A software program that reads XML, validates it, and passes it on for further processing. If a document is not well-formed (i.e., there's an error in the markup), the parser catches it and reports the problem. See also markup, XML processor.

physical structure

The physical organization of data in a file, as opposed to its document or logical structure. An object-oriented database, for example, is a physical structure that doesn't align with its logical structure. See also document, logical structure.


The appearance of a document that has been formatted for human consumption. See also stylesheet.

processing instruction

A markup object that conveys information to a specific XML processor. It has the form:

<?target data?>

where target is a keyword to alert a particular processor, and data is a text string containing the special information. Any processor that doesn't recognize the processing instruction ignores it.

properties declaration

The part of a stylesheet rule that sets the formatting for the selected element. See also CSS, rule, stylesheet.

proposed recommendation

A specification that has been evaluated by the public and is deemed by a standards body to warrant a full recommendation.


A CSS selector that specifies a certain instance of an element instead of all occurrences of the element. For example, the first paragraph of each section can be grouped as a pseudo-class. See also CSS, rule, selector.

PUBLIC identifier

The identifier in the document prolog that gives the name of a DTD or external entity.