Hyperlinks are the heart of HTML. They allow the linking together
of remote documents with the convenience of a click or other activation
mechanism. Hyperlinks are used to establish a relationship between two documents:
a source (usually the current document) and the destination. The
destination of a hyperlink jump can be either an entire document or a
specific location within a document. In addition, the hyperlinking element (A)
can play the role of destination for a URL (using the NAME
or ID attribute), the activation point to another URL (using the
HREF attribute), or both at the same time.
The A element has many other attributes as well (such as TARGET)
which serve to further communicate information about the source and destination documents.
Absolute Vs. Relative URLs
A hyperlink destination is defined as a Uniform Resource Locator
(URL) using this general format:
[protocol]://[network
location ID]/[path on machine]/[file
name]?[query string]#[internal
document location] This is only a general format for specifying a URL - There are many
variations within this framework (such as a "mailto" URL), and
portions of the format can be omitted (such as the [internal
document location].) Please see RFC1738
for more details on URLs.
The path to a document can ALSO be specified relative to the
originating document. In such cases, the [protocol]
and [network location ID] portions of the above
syntax are omitted and [path on machine] is given
relative to the current file or to a complete URL referenced in the BASE element.
URL Encoding
URLs only allow a subset of allowable ISO-8859-1
(ISO-Latin) characters to be used. Any other characters must be encoded
using a special escape syntax that uses combinations of the allowable subset.
A separate page in the Topics section covers URL
encoding in much greater detail, and even provides a little utility to
automatically create URL encodings of characters.