Jump to content

Gopher

From EDM2
IBM Gopher

A hypertext document system and associated protocols for displaying, serving and searching such document on the Internet. Although it was introduced into the wild a little bit later than the WWW system, Gopher was from 1992 and well into 94 much more popular than the web, it was more advanced at the time since it had built in support for structured documents and searching giving rise to tools such as the VERONICA automatic Gopher search system.

Interest in Gopher waned dramatically after the University of Minnesota–Twin Cities announced in late 1993 that they would demand license fees for any commercial use of the technology and that combined with an almost total lack of development of the system by the original authors meant that by mid 1995 Gopher had all but disappeared.

OS/2 Gopher software

Links & publications

Please note that while the article is fairly well written and is a good overview it contains a number of factually incorrect and misleading statements.

Standards

Background and history

Initially developed by Farhad Anklesaria as simplification of Campus-Wide Information System (CWIS), a proposed university wide structured information system for the University of Minnesota–Twin Cities. The board of the university had agreed to fund a small team of developers that would build a client/server system that would connect the students PC and Mac based systems with the resources available on the university's mainframe computers.

While hardware and funding had been in place for a considerable time no actual work had been done on the system, and when a meeting was called by the governors of the university to discuss progress on the system, Anklesaria in a panic created a simple client/server system on an Apple Macintosh that implemented a small subset of the proposed CWIS functions so that he would have something to show for the meeting. The name Gopher was chosen as a play the word gofer, but the Gopher (an animal in the rodent family native to Minnesota) is also the mascot of the UMTC.

While the meeting turned out to be something of a disaster, with the board outraged that the system eschewed the use of the available server capacity and offered no interaction with the existing information systems available on campus. Nonetheless Anklesaria got a permission to create a team to develop the Gopher system further and he choose to team up with Mark McCahill, whom he had created the POPMail email client a few years earlier and they together built a team of around six people to develop and support Gopher.

While initially seeing only a limited success in 1991, things started to happen in 1992 as the original Mac client and server programs improved, a client version was released for Microsoft Windows and UNIX alongside a server version for UNIX systems, and on OS/2 the excellent GoServe program saw the light of day alongside a simple Gopher client Howard Gilbert had written in REXX. At the same time the WWW client/server was only available for the NeXT workstation which limited its popularity a great deal and the subsequent release of NASCA Mosaic and Cello web browsers for Windows and UNIX system did little to improve its popularity as it lacked much of the features Gopher offered.

The real drivers of Gophers popularity however was the built in support for telnet and the two indexing services, the "Very Easy Rodent Oriented Network Index of Computer Archives" (VERONICA) that indexed Gopher sites and Archie that indexed FTP sites. The Telnet support meant that you could telnet to a Gopher server from a terminal even if your computer lacked a Gopher client, this opened up the technology to people that had only terminal access to the internet and not SLIP or PPP connections and also allowed BBS's act as gateways to the gopher world without offering full ISP services.

The search engine VERONICA took advantage of the indexing features built into Gopher servers, rather than search every Gopher page it simply lifted the indices from each server and combined them locally, allowing you to search the entire Gopherspace, a similar search engine did not exist for the Web until the advent of Altavista. The file indexing engine Archie however traversed publicly available FTP sites and indexed the files it found on them.

Despite the obvious advantages Gopher had vis a vis the Web it started to expand at a slower pace than the latter technology in late 1993 and by 1994 the difference was quite startling, by 1996 we have actually started to see the Gopherspere contracting and by the late 90's it barely existed. Shortly after the turn of the century the last Gopher server at the University of Minnesota–Twin Cities was unceremoniously turned off.

While the blame for Gophers demise is usually placed primarily on the University's decision to start charging license fees for commercial use of the system, the situation was actually more complex than that. The team's decision to implement the servers on a Mac rather than the university's existing server hardware came back to bite them as soon as the system gained any popularity, the Mac's single use unprotected operating system was totally unsuited for use as multitasking server, while just serving gopher pages it performed adequately, when people started to use the indexing and telnet features the system ground to a halt. During the Gopher system's heyday the University had a dozen Macintosh computers running as server which altogether cost more than a minicomputer or server system would have cost them which would easily have performed the tasks of the Mac's and more, and in addition the mainframes that the system had originally been intended for the back end of the system had excess capacity and could have run the Gopher back end without any extra investment.

Authors

  • Mark McCahill
  • Farhad Anklesaria
  • David Johnson
  • Bob Alberti
  • Paul Lindner
Retrieved from "https://www.edm2.com/index.php?title=Gopher&oldid=32987"