Child sequences

Navigating by the numbers

IDs are intrusive Sometimes you need to establish a quick link, but don't want to add IDs to your document (or someone else's), or can't count on the environment to support IDs. Child sequences let you 'walk a tree' to reach a target destination.
This number child, then that number child... A child sequence starts with a /, representing the root of the document. The number following that identifies which child element to move to next - typically 1, for the root element. (External parsed entities may have multiple top-level elements.) If you need to navigate beyond the root element, you add a slash followed by an integer value identifying which child element you want - 1 is first, 2 is second, etc. You produce an identifier like /1/4/5, for the fifth child element of the fourth child element of the root element.
Combining with bare names If you wanted to reference the third child of the fourth child of the element whose ID was 'neato', you could write: neato/4/3.
Integrating with URIs Child sequences and bare name/child seqence combinations can appear directly after the pound sign (#)in the URI: http://www.example.com/mydoc.xml#neato/4/3