He predicted that Microsoft's Windows 98 would usher in an explosion of network connectivity. Internet-focused Inktomi Corporation claims that there will be a billion Web sites by 2001. The number is now estimated at 200 million-plus. The Internet will play a central role in commerce and in society as a whole, initiating some startling changes. For instance, Arnold reported, we can expect shifts in population density to areas that provide good connectivity and bandwidth. Just as the building of the canals and then the railroads determined the location of population centers in the 1800s, so the availability of good and affordable access will attract larger populations.

A supranet is already emerging, which includes the Internet, all intranets and extranets, and all tools. This will revolutionize the way manufacturers take and place orders. Mass customization of products will be the result. We see this already in the custom manufacturing Levi's jeans or in computers made to specification and delivered within a week.

Beware the Datasphere

Predicted changes in the workplace seem a bit ominous. Arnold expects a Adatasphere@ to emerge: a 7-day-a-week, 24-hour-a-day work environment. To interject my own comment, most AIIP members are well aware of how this kind of schedule can become a habit and a shackle. Since technology enables us to work from home at any time, personal and work lives become inextricably entwined. Phones ring at night, on weekends, and holidays until the idea of time off for real life becomes a dim memory. Is this progress?

On a more optimistic note, Arnold also suggested that young people are crossing the boundaries between companies and organizations, using technology to collaborate among disciplines and between competitors. We see explosions of innovation as a result.

The increase in the Web population will create new opportunities. We will use new tools to automate what we have done in the past, and use our skills instead at a higher intellectual level. Some of the opportunities Arnold pointed out include the following:

• New opportunities for reintermediation. New information technologies accelerate the distance between the technically proficient and the ordinary consumer. Our clients will need us in new roles, as they discover that new technologies can be a drain on their staff when they all try to learn and keep up with technical changes.

• New demand for services from small businesses. Small businesses are only beginning to join the Internet. They will require the same services that the larger companies have had, but will seek them on an as-needed basis, since their demands will be sporadic.

• Information infrastructure assessment. Evaluate current information systems= infrastructure, then develop, integrate, and re-engineer systems to improve information access and delivery.

• Developing access tools for heterogeneous sources of multimedia information. This will stretch indexing, cataloging, classification, and version control skills.

• Knowledge assessment and mapping to create merged knowledge bases from existing separate ones.

Watch Your Wallet

Concurrent presentations prevented me from attending a session on "Getting Your First Five Clients," with Mary Ellen Bates, as well as a session on accounting procedures, and one on public records information with Carol Lane, Michelle Ayers, Alex Kramer, and Richard McEachin. Curious to know what threats existed to my wallet other than potential emptiness, I chose "Watch Your Wallet," with T. R. Halvorson instead. (Several years ago, Halvorson, a lawyer as well as an AIIP member, presented an excellent talk on liability for information professionals, and it has since been turned into a white paper available from Burwell Enterprises.)

This session explored the problems the Internet and the Web posed to being an information intermediary. In particular, Halvorson raised the alarm concerning UCC Article 2B, which has appeared in the most recent draft (March 10, 1998) of the proposed Article 2 of the Uniform Commercial Code. This covers copyrightable materials and states that information professionals are responsible for the accuracy of the information they provide. This is a frightening prospect, since we can only cite our sources as being generally reputable. Except, of course, in the case of the Web, which is certainly not a respected, peer-reviewed journal.

Most information professionals have expertise in how to find information, not in the subject for which they are searching. They are usually not researchers. Being expected to vouch for the accuracy of retrieved information leaves us open to suit for errors over which we have no control. Halvorson suggested that we disclaim this responsibility at the time of contracting for information services with each client. One wonders where this leaves information professionals in public or academic libraries, who are accustomed to finding information without first signing a contract with the user.

Another potential pitfall Halvorson noted was the question of where a case will be tried if a contract goes awry in cyberspace. Venue is difficult to predict, making a long-distance defense a problem. Further, differing intellectual property agreements for each database on an online information service can make legally copied information in one database illegally copied in another.

When using Web search engines and sources, Halvorson said, be aware that it is difficult to determine the actual date of the information found on a Web page. Also note that recent studies show that only about 34 percent of the indexable sites on the Web are actually indexed, to say nothing of non-indexable sites such as databases that require a CGI script or that use a robots-exclusion feature to prevent indexing.

For Webmasters, creating a Web site that frames someone else's content and represents it, perhaps unintentionally, as their own has been ruled illegal. Similarly, linking or deep linking to a page that bypasses a home page on a site may be illegal if it impinges on the revenue that the site might have earned had the user gone through the home page first.

Halvorson's session was accompanied by a detailed document that explains these and many more potential legal problems concerning information delivery. By extension, it also raises questions of responsible professional practice, since it discusses quality of information and the obligation of the searcher and the Webmaster to supply excellent information. Titled "Internet Opportunities and Liabilities for Information Professionals," the document will be available shortly from Burwell Enterprises.

XML and Document Type Standards

A second keynote address, by well-known industry speaker and author John December, covered the potential of XML for presenting information in a predictable format. December noted that the technologies for the Internet are stabilizing, and he called for the development of open, agile, and lucid systems that support the delivery of high-quality information. He urged attendees to participate in the current project to develop standards for document type definitions that define the structure of documents by category—play, novel, company home page, technical paper, etc. There was some discussion, prompted by a question from the audience, of how this project compared to the extensive work done under the aegis of the Text Encoding Initiative, which has already published document type definitions. And one audience member mused that the hand coding of documents by field might be seen as a step backward, since much of that information can be automatically determined by a good NLP search engine.

Integrated Knowledge Discovery

December's presentation contrasted nicely with the next session, "Beyond Document Retrieval," with Elizabeth Liddy of TextWise. After an introduction to the technology behind natural language processing, she explored the potential of natural language processing-based tools for analyzing the contents of databases, not merely retrieving them separately. Liddy calls this Integrated Knowledge Discovery, "the computational process of identifying, selecting, and extracting useful information from massive volumes of digital data."

Tools are being developed that can detect meaningful patterns within the data, she said. These products will create opportunities for a different kind of information delivery. Demonstrating KNOW-IT, which visually portrays such relationships as causation, she said such tools will enable discovery of competitors, trends analysis, visual exploration, knowledge management, patent portfolio analysis, and summarization. Attendees expressed excitement at the potential this offers for analysis and presentation of research results to clients and users. KNOW-IT, which is currently in beta test, should be on the market by late 1998.

Looking Back to Look Ahead

The last session of the conference was an extraordinary opportunity to hear the founders of the online world reminisce about why and how they began, what their challenges were, and if they would do it all again. Roger Summit, who created DIALOG; Eugene Garfield, the founder of the Institute for Scientific Information (ISI); and Richard Harris, who is president of Responsive Database Services, recounted the early days of inventing the online information services.

What drove them to invent and initiate? Did they have a path mapped from the beginning, or did that path only appear clearly in retrospect? From their stories, it appears that the ultimate path was never clear, but they had the ability to see opportunities and needed services at crucial points in the development of the industry. Garfield said it had been his clear intention to return to graduate school, but he was offered a chance to earn a living instead, first at Smith Kline Pharmaceuticals, and then at Miles Laboratories. The combination of his background in both library science and in chemistry, plus his recognition of a need for and a niche for quick delivery of contents pages of journals, drove him to establish his own business, because no one else would. Current Contents was the first automatically generated current awareness service. Others have been refining the idea ever since.

Summit wanted to develop cognitive applications for computers. In 1960, as an intern at Lockheed, he was asked to look into the use of computers for information retrieval. His vision of the need for interaction with the computer coincided with the arrival of third-generation computing technology with its massive random access storage, remote applications and multitasking, and a large database of materials from NASA, as well as a supporting research environment. The result was DIALOG. Things we now take for granted were invented at this time: recursive and cumulative sets, so that the searcher could combine and recombine strategies; index terms that were displayed as a simple thesaurus, in order to guide the searcher to the most-used terms; remote access; multiple simultaneous users. Coincidentally, the NASA RECON project needed an information retrieval system, and so did the AEC and ESRO. Summit set up systems for each of them. However, DIALOG became a service bureau and not a systems development business because ERIC didn't want to maintain its own system. Rather, they were looking for someone else to offer access to their files. There is a reason why ERIC on DIALOG is File 1. The rest is history.

Richard Harris is a longtime player in the online arena, with his background first at ISI and then at Predicasts. However, he became an entrepreneur in 1994, after Information Access Company bought Predicasts, and he "needed a job." RDS presents an interesting contrast to the stories of DIALOG and ISI, for it began 30 years later and consequently had to spring full blown into the online world. DIALOG and ISI developed slowly, inventing as they went along. Harris knew what was needed: indexing systems, editorial systems, licenses for full text, a product that would compete against those already established. Would he do it again? Absolutely, but "sometimes you have to be pushed."

Although this session was about the past, it was also a prelude to the future. These visionaries are still shaping the online industry, with optimism and clear-sighted determination.

The AIIP Conference Awards

This year's IAC Authorship Award went to Wenda Webster Fischer for her article "Push Technology: An Opportunity for the Information Professional," which appeared in Connections, Winter 1998.

The Myra T. Grenier Scholarship Award, presented to an associate member to enable him or her to attend the conference, was won by Glenna Rhodes.

Next year's AIIP conference will be held April 20-27 in Berkeley, California. AIIP can be reached at 609/730-8759, or on the Web at http://www.aiip.org.