The U.S. Copyright Office and the Justice Department have reportedly filed briefs supporting Borland against Lotus Development Corporation. Lotus claimed that the option of using a 1-2-3 interface within Borland's spreadsheet was a violation of copyright. The Copyright Office says that simple lists of commands (e.g., menus) are not copyrightable. (The court could disagree, of course.) [SJM, 10/1.]

A new ruling be the Supreme Court holds that Yellow Pages are not copyrightable, other than the subject headings and manner of organization. White-pages copyright protection was denied in a previous ruling. [WSJ, 9/24.] This is likely to segment the information industry into those who compile wholesale information and those who license, augment, format, and distribute it at retail. It opens the way to inexpensive niche-market directories excerpted from comprehensive compilations, with more attention paid to client needs. (There is ample precedent in the private republishing of government data. And if you're looking for a home-based business, consider desktop-publishing an industry directory that you can sell to the people in the directory.) Highly refined data could sell at premium rates, although competition will keep prices down. Data gatherers such as Gale Research and Who's Who may become "knowledge refineries," putting additional effort into data merging, validation, and classification. [Clifford Urr (well!cliffu@apple.com), PACS-L, 9/26.] Knowledge-based database cleaning may become a hot new AI research area.

IEEE reports that 50% of its members move every 18 months. I've heard that management job titles have a half-life of about 9 months, and that lists of corporate officers are 50% incorrect by the time they're published. If you're tracking individuals, how do you predict or validate job changes. What subtle clues help match new data with existing dossiers? (Example: a low- numbered check indicates someone who has moved recently.) Who will be the first to write an expert system named SHERLOCK?

I mentioned last week that TRW is having a terrible time keeping its credit reports accurate. Did anyone see that as an opportunity for knowledge-based expert systems? Every problem is an opportunity.