Errors in USCO database

I would like to bring to the attention of this list the problem of the possibility of errors in the U.S. Copyright Office computerized database of copyright registrations.

A few weeks ago, while searching the database, I stumbled on an entry involving a book I had written and published (under a pseudonym) in 1992. The entry was not for the original copyright TX registration, but for a set of six audio cassettes made with that book title and my name as author, the entry giving the audio publisher and an SR (sound recording) registration number. I immediately contacted the print publisher of my book, castigating them because I had never been told they had sold audio rights, etc., etc. They said they had no idea what I was talking about and said they would investigate. Since then, after thinking about the matter, I have decided the idea that anyone would make a sound recording on six cassettes of that particular book is ridiculous, as ridiculous as making a musical of Feist v. Rural. (Although if someone did produce a musical based on Feist v. Rural, I would certainly buy a front row seat.)

So my present view is that the computerized CO database is in error on this entry, the error originating either in the database itself or in the software that searches the database. (The error is replicable and definitely in the CO data or software). The entry that came up looked perfectly fine: Book title, my name as author, and then the usual record information -- except in this case the record information is probably from another record. If errors such as this truly exist in the CO database (or bugs in the search engine), they will probably go unrecognized by everyone except the author and publisher. So any third party searching the database for any purpose, including litigation, may be confronted with an erroneous record and not know it. Since authors and publishers (the people who would recognize the errors) only rarely search the database, existing errors will not easily be located and corrected. I think the existing errors are a lost cause -- there is really no reasonable way for third parties to locate them systematically. But as far as future entries are concerned, I suggest the copyright community urge the CO to consider the introduction of a software error code for each record, so that if any record does become corrupted it will at least be flagged when it is brought up. There are many possible alogrithms for error codes, and I suggest one ought to be used to protect future entries in the database.

Dan Agin
Spectrum Press
specpress[_at_]earthlink.net Received on Mon Oct 28 1996 - 01:39:44 GMT