C. Wayne Ratliff 13 years ago simply wanted to write a personal computer software program that would help him win the office football pool. Instead, the then-young computer programmer came up with one of the best-selling software programs of all time--but now also the center of a brewing industry controversy.
By day back then, Ratliff and his colleagues at the Jet Propulsion Laboratory in Pasadena kept track of the comings and goings of the solar system for NASA's unmanned space program on a huge Univac mainframe computer.
By night, in the cluttered den of his Tujunga home, Ratliff tried to turn the same principles that tracked, sorted and assessed information on the stars and planets into a PC program that would analyze National Football League statistics and spit out the likely winners of each week's matchups.
By 1978, Ratliff had written such a program. Called Vulcan, after the home planet of "Star Trek's" Mr. Spock, the software was built around a set of 58 computer commands--33 of which Ratliff said he "borrowed" from the JPL software. It could sort, edit, display, copy and count vast amounts of data--more information than had ever been tracked on a personal computer.
Ratliff tried to make a business success of his work, but it took two aggressive Los Angeles marketing men to exploit its full potential. George Tate and Hal Lashlee renamed the program dBase, because it sounded "techno-trendy," and built an entire company--Ashton-Tate--to bring it to the masses.
The plan worked. Virtually overnight, dBase made Torrance-based Ashton-Tate what it remains today: the nation's fourth-largest software publisher.
However, Ashton-Tate's continued success, and the status of its star product, were thrown into jeopardy Thursday when a federal judge stripped the company of its copyright protection for dBase.
U.S. District Judge Terry J. Hatter in Los Angeles ruled that the company had repeatedly failed to disclose, on a series of copyright applications filed with the federal government from 1984 through mid-1990, that Ratliff's program was derived from the JPL program. The judge said the company's actions were undertaken "knowingly and with an intent to deceive."
If the ruling is upheld, it would be a serious blow to Ashton-Tate because it would give rivals a legal basis to freely produce dBase clones.
Ashton-Tate executives, who say they have nothing to hide, have vowed to fight the ruling.
Ashton-Tate's current senior executives, none with the company when the first copyright application was filed in 1984, said Friday that while the company has repeatedly acknowledged the connection between Ratliff and JPL, it does not consider dBase to be a derivation of the laboratory's software.
"DBase is an original work inspired by the ideas in the JPL program," said William P. Lyons, Ashton-Tate's chief executive. "We weren't trying to hide anything. Everyone in the entire industry knew where Wayne had worked and gotten some ideas."
Lyons said the only reason the connection was not disclosed in the copyright filings is that the lawyer completing the first application was unaware of it. Subsequent filings, he said, were copied from the first.
Ratliff, now 44 and a computer consultant living in San Luis Obispo, said Friday that he is angry that his work and its derivation and inspiration have been called into question.
"The JPL program is the model I used. I have always acknowledged that, and so has the company," said Ratliff. "I borrowed here and there . . . I didn't copy it . . . I added something to it."
The birth of a computer program may be a marvelous thing, but tracing its lineage is often complicated and murky. A software writer can contend that he has been "inspired" by the work of another, but others might judge it outright mimicry, or even copying.
How do software writers tell the difference? With some difficulty.
"There is not exactly a bright line between what is or is not a derivative work," said Stuart Lubitz, a Los Angeles attorney specializing in intellectual property.
Ratliff agreed. "My mother always said that there was nothing new under the sun, and she was right. I challenge anyone to find work without elements of stuff that has preceded it."
Ratliff said that while 33 of the 58 commands in the original program were taken from the JPL program, the two sets of software bear little additional resemblance.
The JPL program, he said, was written in the Fortran computer language, while his was written in Assembly, and later translated into the "C" computer language.