About a year ago a message appeared on the LIBREF-L listserv hyperbolizing Washington and Lee's 'lawlib' as "the greatest Internet site". Articles in the Vancouver Sun and in several legal publications publicized the login instructions, and so many Internet explorers connected that it became necessary to limit the number of logins from outside computers. 'Lawlib' (or Netlink, as it was more properly known) is no more, but its functions now operate as a subpart of Washington and Lee's gopher server and its indexing and connection features can thus reach a larger audience. We want to describe the evolution of W&L's suite of electronic information tools from both the technical/operational and the intellectual/pedagogical perspectives, in the full awareness that the picture we present looked very different six months ago, and will be different again, in ways we foresee imperfectly, by next fall.
Changes occur so rapidly that evolution of an electronic information system from the germ of an idea to a realized organism would be difficult to study, and there are no rule books or manuals that lay out procedures for the development process--or at least none that remain current for more than a few months. A site develops as new possibilities occur to its keepers, or appear as items of discussion in the semi-formal media of listservs and usenet groups. The players in the evolutionary process share information and procedures in a remarkably cooperative and open way, based very much in adventurous ad- hockery. It is easy to lose sight of how recently the tools we take for granted today were just ideas in somebody else's mind, and how recently these essential tools changed what we perceive as the information access reality of the moment. It is correspondingly hard to reconstruct one's own sequences of realization of possibilities and acquisitions of skill.
Three years ago lists of sites which could be reached by telnet and FTP were the main resources for exploration, the community of users was much smaller, and the level of sophistication about computing was probably higher overall than is now the case. Three years ago most users didn't venture far from their own specialties --a biologist might know about BITNET listservs in his or her field, but most had no reason to be concerned with other subject areas and no easy means to explore unknown territory. While the average level of technical computer knowledge among users is probably lower now, sophistication about the uses that can be made of resources is much greater, largely due to the development of menu interfaces like Netlink, gopher and WWW, and effective search and connection tools like Archie and Veronica. The tasks of those who teach the use of Internet resources are less entangled in technical maneuvering, and their efforts can therefore be directed more to the _content_ of resources that are much easier to locate and retrieve.
The menu presentation of Internet resources known variously as 'lawlib', 'Liberty', and 'Netlink' first appeared in late 1991 on a Prime Computer, and in early 1992 migrated to an HP Unix box. It began its life primarily as a text repository for law librarians. It was not long before word of Netlink's contents spread via the Internet bush telegraph. In mid-1992 the site offered, in addition to a broad range of law-oriented text files, access to a number of the large Internet Guide textfiles of the time (Noonan's, Barron's, Yanoff's, St. George's, the Hytelnet list), and the SysOp was experimenting with direct connections (via scripted logins) to telnet sites, primarily library OPACs. Archie, Gopher, WAIS and WWW links were on the menu, but the user's main means of discovery was serendipitous browsing.
By the beginning of 1993 the menu was much changed, and Netlink included classification of its contents by subject area, means to search and limit retrieved sets of resources, and utilities to make the user's visit more profitable. In its maturity Netlink provided indexing and connection functions for more than 2500 Internet resources, and was the easiest way to find out what was new on the Internet. Even with restricted logins the external use continued to grow, and through 1993 was averaging about 600 interactive logins per day.
(John Doyle will discuss the technical evolution of Netlink)
The hierarchical menus of the gopher interface make it easy for beginners and occasional users to explore Internet resources with a minimum of frustration and disorientation. The simple fact that a gopher burrow leads to burrows everywhere is easily understood, and the transparency of connection (no arcane protocols, no confusing directory structures to navigate) lets the user succeed with retrieval tasks _and_ facilitates the serendipitous attitude that gladdens a librarian's heart.
A gopher ringmaster has to choose a means and method of organization of the material contained in gopher menus, and the main guidance is the structure of other gophers. There are some rules of thumb (keep the initial menu short, maximise ease for the user), but gophers seem to grow according to their keepers' whims rather than in response to the rules of any internal genetic code. Some gophers begin as a means to distribute texts (and/or images, etc.), while others are mostly composed of links to a selection of other people's resources. A few are consciously modeled as Electronic Libraries (NCSU's Library Without Walls and UCSC's Infoslug are admirable examples), while others have chosen to structure themselves to serve a particular local or disciplinary clientele. But most gophers are hybrids, serving several functions and in any case permitting easy connection to other gophers with different characteristics and content.
(Hugh Blackmer will discuss the gopher as a librarian's tool, designed to support local instructional objectives and provide appropriate and effective guidance to student and faculty users)