<hypertext> (Or "span", "region", "button", "extent") An area
	within the content of a {hypertext} {node} (e.g. a {web page})
	which is the source or destination of a {link}.  A source
	anchor may be a word, phrase, image, or possibly the whole
	node.  A destination anchor may be a whole node or some
	position within the node.

	Typically, clicking with the {mouse} on a source anchor causes
	the link to be followed and the anchor at the opposite end of
	the link to be displayed.  Anchors are highlighted in some way
	(either always, or when the mouse is over them), or they may
	be marked by a special symbol.

	In {HTML} anchors are created with the <A..>..</A> construct.
	The opening A tag of a source anchor has an HREF (hypertext
	reference) attribute giving the destination in the form of a
	{URL} - usually a whole node or "page".  E.g.

		<A HREF="http://www.foldoc.org/">
		Free On-line Dictionary of Computing</A>

	Destination anchors are only used in HTML to name a position
	within a page using a NAME attribute.  E.g.

		<A NAME="chapter3">

	The name or "fragment identifier" is appended to the URL of
	the page with a "#":

		http://www.fairystory.com/goldilocks.html#chapter3

	(Though it is generally better to break pages into smaller
	units than to have large pages with named sections).

	(1997-11-15)

<language, text> An {SGML}, {HTML}, or {XML} {token}
	representing the beginning (start tag: "<p ...>") or end (end
	tag: "</p>") of an {element}.  In normal SGML {syntax} (and
	always in {XML}), a tag starts with a "<" and ends with an
	">".

	In {HTML} jargon, the term "tag" is often used for an
	"{element}".

	(2001-01-31)

1. <networking> A {bang path} or explicitly routed {Internet
	address}; a node-by-node specification of a link between two
	machines.

	2. <file system> {pathname}.

	3. <operating system>  The list of directories the kernel
	(under {Unix}) or the command interpreter (under {MS-DOS})
	searches for {executables}.  It is stored as part of the
	{environment} in both operating systems.

	Other, similar constructs abound under Unix; the {C}
	{preprocessor}, for example, uses such a search path to locate
	"#include" files.

	[{Jargon File}]

	(1996-11-21)

<networking> 1. A lasting connection between a user (or user
	agent) and a {peer}, typically a {server}, usually involving
	the exchange of many packets between the user's computer and
	the server.  A session is typically implemented as a layer in
	a network {protocol} (e.g. {telnet}, {FTP}).

	In the case of protocols where there is no concept of a
	session layer (e.g. {UDP}) or where sessions at the {session
	layer} are generally very short-lived (e.g. {HTTP}), {virtual}
	sessions are implemented by having each exchange between the
	user and the remote host include some form of {cookie} which
	stores state (e.g. a unique session ID, information about the
	user's preferences or authorisation level, etc.).

	See also {login}.

	2. A lasting connection using the {session layer} of a
	networking protocol.

	(1997-08-03)