The World Wide Web has quickly become an essential tool for librarians to communicate with their clientele, provide services, and build collections. Well over a thousand libraries around the world now have Web servers (according to Libweb, a directory of library-based Web servers at http://sunsite.berkeley.edu/Libweb/), and many have public terminals available for accessing the Web. Such a rapid deployment of this technology required rapid learning by the professionals who implemented it. Part of that learning took place on an electronic discussion called Web4Lib (http://sunsite.berkeley.edu/Web4Lib/).
Web4Lib is an electronic discussion for library Web managers. Although the primary audience is library staff, there are many others who find the list helpful who do not work in libraries. Topics appropriate to the Web4Lib discussion include anything relating to Web systems and libraries. Some examples include:
- Web resource selection in relation to existing acquisition and collection development procedures
- cataloging and metadata issues regarding Web information
- in-house patron access to Web servers (e.g., Netscape on patron-accessible computers)
- training staff or users to use the Web or to create Web resources
Off-topic postings may lead to the removal of the person posting the message, and commercial advertisements are not allowed.
Web4Lib has been blessed with a cadre of very knowledgeable Web managers that often supply informative and helpful answers to queries from those of us who are less experienced. Although occasionally Web4Lib receives an off-topic message or sometimes a discussion is pursued to a ridiculous length, in general Web4Lib retains a fairly high overall signal-to-noise ratio. This is reflected in comments such as:
"I thank you, for running a list that is head and shoulders above most. I really appreciate the professional level of conduct and content."
"I subscribe to this list and also to comp.infosystems.www.providers and sometimes, ironically, I find this group more helpful when it comes to providing, or webbing in general, so that I don't even miss www.users and www.misc."
"From my corporate corner, I agree that Web4Lib at Berkeley is a wonderful service. I too am grateful."
The Web4Lib discussion averages about ten messages a day (approximately 300 per month) on all kinds of topics related to Web servers and libraries. Current membership stands at nearly 2,800 individuals from more than 40 different countries. There are over 100 subscribers from the United Kingdom (using June 1996 statistics). The top five countries with the most subscribers are the U.S., Canada, the United Kingdom, Australia, and Germany. More detailed statistics by country are available at http://sunsite.berkeley.edu/Web4Lib/subscribers.html.
Web4Lib messages are available in a searchable and browseable archive (http://sunsite.berkeley.edu/Web4Lib/archive.html ) from April 1995 to the present. The archive is updated automatically as messages are posted to the discussion.
The Library Web Manager's Reference Center
The Library Web Manager's Reference Center (http://sunsite.berkeley.edu/Web4Lib/faq.html) began as an outgrowth of the Web4Lib discussion, first to provide easy access to some of the most informative Web4Lib postings, then later to also point to key resources for library Web managers. Major sections of the page include the Best of Web4Lib, CGI Programs, Directories, Guides and Tutorials, Launching CD-ROM or Other Applications From a Web Browser, and Web Policies.
In the spring of 1994, Information Systems Instruction & Support (ISIS) staff (a unit of the Library Systems Office) at the University of California, Berkeley Library were preparing to build a prototype Web server. By that time it had become clear that such a server would be an inevitable expansion of network-based services that The Library had begun offering with its library catalog and Gopher server. As knowledgeable as we may have been about the Web at that time (we had investigated the Web as early as 1992), we still had much to learn about mounting a Web server of the size that would be sufficient to serve a campus of 30,000 students and a library staff of 400.
In late April 1994 Bill Drew from the State University of New York posted the following message on the NEW-LIST electronic discussion:
"I have been looking high and low for a list aimed at librarians involved in developing resources for the World Wide Web (WWW). I have searched the List of Lists, print directories and other sources. There already is a list for librarians developing gopher resources but there is no interest there to expand it to include WWW efforts."
Bill Drew's message acted as a catalyst that spurred us to pursue the possibility of hosting such a list at UC Berkeley. We received permission from Library administration to host it, at which point we began work on the technical details. I replied privately to Bill Drew that we were working on putting up such a list, and on May 5 I announced our intent to begin such a discussion on the GO4LIB-L list (a list for library-based Gopher managers which served as a conceptual model for Web4Lib). On May 12, 1994 I announced the existence of the discusion Web4Lib on GO4LIB-L, NEW-LIST, and PACS-L (a broad-based library discussion). In less than 24 hours we had 360 subscribers. We hit 1,000 subscribers within two weeks. By a measure of numbers alone we were, much to our delight and dismay (see below), a smashing success.
It quickly became clear that we were ill-prepared to host such a large discussion. The software we were using (The Unix Listserver Program or TULP) was a bare-bones package that lacked some of the more sophisticated commands and features demanded by users familiar with electronic discussions. TULP was slow, and it was frequently backed up with a long queue of messages waiting to be delivered to subscribers. Dealing with returned mail was a nightmare. The systems technician assigned to TULP dropped everything and spent most of her time for the first couple weeks dealing with problems from the list.
But despite a rocky start, the list was immediately viewed as helpful by its subscribers. One early response we received read in part "I have read this list one day and have already found it useful...maybe if I have 900 messages over the next three days I will change my mind but it is certainly augmenting my R&D efforts. So, thanks for starting it." As cheering as that message was, it was also prophetic, as at one point a subscriber from Australia was causing an endless loop. This meant that everyone on Web4Lib was getting multiple copies of the same message, leading to dozens of duplicate messages piling up in subscriber's accounts. After tracing the problem to an institution and removing everyone on the list from that location (as well as corresponding with individuals there), we discovered the problem remained unsolved. At one point, in our frustration and our belief that the message was being propagated by other servers, we were forced to remove the entire continent of Australia from the list and, for a time, to suspend the list entirely.
This incident also forced us to moderate the list, which meant checking every message manually before forwarding it to subscribers. Web4Lib was moderated in this fashion for several months, until we felt comfortable that the bounced mail problem had ceased. In September 1995 we moved the list from the TULP software to List Processor 6.0 (often called ListProc for short), which afforded much better electronic discussion support.
The following are a few of the basic commands for interacting with the list. A more complete list of commands is available at http://sunsite.berkeley.edu/Web4Lib/.
Send the message "subscribe Web4Lib your name" to email@example.com
Send the message "unsubscribe Web4Lib" to firstname.lastname@example.org
To Contribute to the Discussion:
Send your message to email@example.com. For more information on appropriate discussion topics, please see the information below.
To Receive a List of Commands:
Send the message "help" to firstname.lastname@example.org
The list is managed by Roy Tennant, with technical assistance provided by Janet Garey, Eileen Pinto, and Tricia McCarthy, all staff at the UC Berkeley Library. Other staff who have assisted, both current and past, include David Rez, John Ober, and Lisa Rowlison.