Tutorial – Guide to Effective Searching of the Internet
[ Previous | Home | Index | Next ]
This tutorial has spent considerable time on all aspects related to how to search on the Internet and the search services available. What does the current state of Internet searching suggest for the future? And, are there easier ways than needing to learn all of the nuances of various search services?
Topic 49: Ruminations on the Future of Internet Searching
You may click on any of the links above to go directly to that topic.
We do not see the demise or "death" of search engines, as some pundits have argued. Major search engines will continue to be one of the most important first access points to the Internet. The sheer growth and chaos of the Internet assures this. But there will also be twin, divergent forces toward consolidation on the one hand and specialization on the other.
The first-generation of Internet search services are facing an untenable impasse. Sheer document volumes - approaching 1 billion in the near future - doesn't bode well for either traditional search engines or search directories. The largest search engines today approach nearly 150 million documents. Results are overwhelming and force users to study tutorials such as this one in order to query for meaningful results. This problem will only get worse. Directories, however, with their (supposedly) higher-quality content selected by human reviewers are also falling more rapidly behind. For example, perhaps 1.5 million new documents are being added daily to the Web, surely a volume that can not be hand-screened by humans at acceptable cost. And, spot checks of Yahoo! already indicate that 20-25% of existing links are no longer in existence or out of date.
Machine indexing, classification and qualification are the only meaningful techniques whereby the quality content of the Internet can be maintained in a central location. Results presentations also need to be well-organized in a directory structure of much greater richness and depth than what current directories offer. In the words of Clifford Lynch in a recent Scientific American article [45]:
"One sometimes hears the Internet characterized as the world's library for the digital age. This description does not stand up under even casual examination. The Internet - and particularly its collection of multimedia resources known as the World Wide Web - was not designed to support the organized publication and retrieval of information, as libraries are. But if the Internet is to grow and thrive as a new means of communication, something very much like traditional library services will be needed to organize, access and preserve networked information. Even then, the Net will not resemble a traditional library, because its contents are more widely dispersed than a standard collection. Consequently, the librarian's classification and selection skills must be complemented by the computer scientists' ability to automate the task of indexing and storing information. Only a synthesis of the differing perspectives brought forth by both professions will allow this new medium to remain viable."
We see the continued specialization and balkanization of search engines on two levels. The first level, involving the major search services, will see consolidation and specialization at very different ends of a spectrum according to the needs of various user communities.
At the broadest consumer level, one thrust will be to provide more "intelligence" to infer simple query needs. This will enable natural language querying. Services that emphasize this strategy will attempt to become "one-stop" destinations, offering much more than searching, as a means to keep visitors longer. Virtually all of the directory services now fit in this category, with Excite and others moving in this direction as well. One might call this the McDonald's or Pepsi approach to establishing a broad, branded consumer service. Absolute coverage of the Web's content will be less of a driver; listing positions will be based on payments, popularity and advertising support. Query simplicity will be emphasized over user control and elaborate syntax.
At the other end of the major search service spectrum will be full-text engines, with full Boolean support and many filter options, to serve the information-intensive user community. Great emphasis will be placed on expanding the Web's coverage by these services. The revenue model here may be advertising revenues from firms targeting this demographic, or providing demonstrations of advanced technology (Digital's original motivation in establishing the AltaVista service now owned by Compaq). Attempts like Northern Light to provide "special" information on a subscription basis may work well for business users; we have doubts whether this is a sustainable revenue model for educators, students and others with strong information needs.
The second level, an opening created by today's inability for the major services to cover the Web, will likely be the fastest growing category. This level is the specialization of engines by major topic area — law, science, medicine, business, etc. — to serve those specific communities. We see much consolidation occurring in each of these niches, even while the importance of the niche expands. Adding proprietary content, and the possible aggressive entries of traditional search database firms such as Dialog and Lexis-Nexis, should keep specialty topic searches an area of ferment for some time to come.
The bridging "glue" to tie all of these disparate pieces together will be the metasearchers, either Web-based services or dedicated desktop tools. It is quite conceivable, indeed likely, that Web-based metasearchers will partner with the specialty topic services to broaden their current offerings beyond the six or so major search engines that they now cover. This would free the specialty services to focus on the topics they already understand, while giving the consumer more of a single-entry point into the Web's entire content. The role of the metasearchers will be to provide de facto standardization to Internet searches.
[ Previous | Home | Index | Next ]
|
