September 2000

        I have worked on and studied the most complex and standardized systems of information retrieval in the world---library cataloguing---for more than 30 years at the national and international level.  In particular, I have been involved in the development of international standards for library cataloguing and for the exchange and dissemination of computer bibliographic records.  In recent years, I have been particularly interested in the seemingly intractable problem of bringing the same level of standardization and retrieval to worthwhile electronic resources located on the Internet and the Web.  There are two problems in dealing with this question.  One is the manifest shortcomings of free-text keyword searching and retrieval.  The other is the fact that a lot of resources on the Net and the Web are of very local and/or transient interest.  Keyword searching is inexpensive but ineffective (the "14,354 hits" problem), whereas traditional library cataloguing and retrieval is highly effective but extremely expensive.  The latter is due to the fact that it requires highly trained professional intervention.  Thus, those interested in a high level of retrieval effectiveness are caught in a seemingly unresolvable dilemma.  Lee Grant and Susan Capizzi's International Internet Categorization System (IICS) offers a very promising third way that will be inexpensive to apply but effective in retrieving the 90%+ of the Internet and the Web that is of immediate interest, while being an umbrella system that allows articulation between the IICS and fuller cataloguing of the small minority of electronic resources that are of enduring value and need to be archived.

         One of the most innovative aspects of the IICS approach is the way in which it transcends linguistic barriers.  The failures of automatic translation and the difficulties of multi-language thesauri have bedeviled international subject retrieval.  The IICS offers a way out of this maze by prescribing globally recognizable icons with internationally standard meanings.

         One of the major problems confronting content providers and libraries in an electronic world is that of blocking access by minors to pornography.  "Filtering" systems are ineffective for the same reason that search engines are ineffective---they depend on the vagaries of natural language.  The IICS categorization "X" (for sexually explicit sites) combined with its first tier categorization of commercial sites would ensure effective blocking of such sites in appropriate cases.  Moreover, such a categorization would be welcomed by the creators/publishers of such sites as a selling point---the same categorization that makes effective blocking possible would make speedy access by willing adult purchasers possible.

        I believe that, if the IICS is implemented, we have the real possibility, for the very first time, of a sophisticated and effective search and retrieval system for Internet and Web resources that will have major implications for the identification and archiving of valuable human records that happen to be available in electronic form.

Michael Gorman
Dean of Library Services
California State University, Fresno
&
President, American Library Association
&
President, Library and Information Technology Association, 1999-2000
&
Editor, Anglo-American Cataloguing Rules, Second edition

iicsinscte02