Paul> I do not believe that there is any way ot implementat a legal XML
Paul> parser without keeping around all of the information required to
Paul> implement short end tags. Checking that an end-tag matches its
Paul> start-tag (the current situation) is no easier than not checking.
But there are plenty of (non-parsing) applications that benefit from
XML standard end-tags. An obvious one is selection of an element from
a document; a regexp search for the start-tag, and then just match
start and end tags *for that element type*, keeping track of depth
*for that element type* (we don't even need to do that if the element
type is known not to be nestable in itself). That application need
not even notice tags for other element types.
--