Gopher and Gopherspace

Several attempts have been made to unite many of the information services offered on the Internet. While the World Wide Web offers a world-wide hypertext, Gopher, developed in 1990, takes a purely menu-driven approach. The information available through Gopher is called the Gopherspace. It consists of several thousands of Gopher servers, each offering many menu items, which can be of about 10 different types (including menus, text, sound, pictures, binary data and ftp and telnet sessions).

Browsing through Gopherspace is done by selecting menu items. The information nodes (text, sound, pictures, etc.) cannot contain anchors leading to other nodes. There is a strict separation between menus (only containing anchors, called menu items) and information nodes (only containing information, not anchors).

The Gopherspace can be searched efficiently for regular expressions using the Veronica query language. The queries are actually evaluated on a remote server, and are only based on the titles of the nodes, not on their content. The remote server(s) scan the entire Gopherspace once every few weeks (one or two), retrieving all menus and recreating their database.

Gopherspace is often not considered an example of a real (world-wide) hyperdocument, because it is not possible to add (outgoing) links to information nodes. All links are concentrated in the menus. Only by combining gopher with information nodes written in HTML and by accessing Gopherspace through a WWW-browser instead of a specific Gopher interface can links be added to nodes and accessed through the interface.