Williams v. Arndt

626 F.Supp. 571 (1985)

Larry WILLIAMS, Plaintiff,
v.
George ARNDT and Harvard Investment Service, Inc., Defendants.

Civ. A. No. 83-3397-MA.

United States District Court, D. Massachusetts.

September 23, 1985.

Michael Arthur Walsh, Esq., Choate, Hall & Stewart, Boston, Mass., for plaintiff.

Irving M. Kriegsman, Framingham, Mass., Alvin S. Nathanson, Nathanson & Goldberg, P.C., Boston, Mass., for defendants.

OPINION

MAZZONE, District Judge.

This is an action for copyright infringement under the Copyright Act of 1976, 17 U.S.C. § 101, et seq. The plaintiff is Larry Williams of Kalispell, Montana, an investment adviser and commodities trader. The defendants are George Arndt of Harvard, Massachusetts, a commodities trader and adviser, and Harvard Investment Service, Inc., a Massachusetts corporation. Arndt is the president and sole stockholder of Harvard Investment Service, Inc. Prior to forming Harvard Investment Service, Inc., Arndt did business under the name of Investors Micro Messenger. The complaint consists of 5 counts: Counts I and II charge copyright infringement by Arndt and Harvard Investment of a booklet called "Floor Trader's Manual." Counts III and IV charge copyright infringement by Arndt and Harvard Investment of a second booklet called "Floor Trader's Commodity Method." Count V charges unfair acts and practices and unfair competition in violation of M.G.L. c. 93A. The plaintiff seeks injunctive relief and money damages, including attorneys' fees and costs.

The defendants deny infringement and have filed a counterclaim. They allege the copyrights of both works are invalid and unenforceable and that the applications for certificates of registration were accomplished by fraud and misrepresentation. Further, the defendant, Arndt, alleges he was libelled when the plaintiff publicly accused him of copying the Floor Trader's system in the October, 1983 issue of Futures, a commodities trading magazine.

This Court has jurisdiction pursuant to 28 U.S.C. §§ 1332(a) and 1338(a). Venue is proper pursuant to 28 U.S.C. § 1400(a). The case was tried to the Court without jury. Pursuant to Fed.R.Civ.P. 52(a), I make the following findings and rulings.

I.

The plaintiff, Larry Williams, has been an active commodities trader for 15 years. He is the author of various books, articles and pamphlets on commodities trading. He has conducted various seminars at which his authored works have been distributed. These seminars have been given usually at four locations in the United States. The cost to attendees was $3,000 and they were required to sign a non-disclosure agreement. Similarly, Williams' written material on trading methods sold for about $1,000 and was also subject to a non-disclosure agreement.

In January or February, 1980, Williams began working on a specific, step-by-step method for trading in various commodities. By December, 1980, after spending an average of 6 hours a day on this project, Williams had completed his research and compilation of data and began to rough draft his results and conclusions in a manual form. He sent his rough draft to a printer, Crum's Instant Print, in Kalispell, Montana. Shortly thereafter, he received 10-15 copies from the printer. He kept one for his own file and sent a copy to about 10 of his more preferred customers who had ordered the booklet in response to a promotional flier Williams had limited to certain customers. The price was $2,000.

Williams, in reviewing this first printing, had noted that this first booklet, entitled the "Floor Trader's Method" (FTM) did not have a copyright mark or his name on it. It did not have a title on the cover, and it contained some errors, some typographical and some textual. Making some corrections, and telling the printer to place his name, a copyright mark, and title on the manual, Williams returned one copy to the printer.

An additional copy was sent to a past customer, Jerry L. Snider, a market trader who lived in Oklahoma. Snider had attended a seminar given by Williams in 1980, but had not been successful in trading commodities. He had received the promotional flier from Williams alerting previous customers to the imminent publication of the new method. Snider called Williams and told him he wanted either a refund of the seminar fee or a free copy of the new method. After confirming that Snider had been unsuccessful, although he felt Snider had not followed his method correctly, Williams agreed to send him a free copy on the condition that there would be no refund and Snider would not disclose the method. Accordingly, in late 1981, Williams sent Snider one of the unedited copies that had been received from the printer. As stated above, that copy did not contain any copyright mark. At this point in time, copies of the FTM were in the hands of 10 identified purchasers and Snider had a copy. In all, then, 11 copies were distributed. One copy had been sent to the printers for corrections, and Williams retained a copy.

After making the required corrections and editing, the printer delivered about 75 copies of the revised edition of the method to Williams. This edition was called "Floor Trader's Commodity Method" (FTCM) and contained the copyright mark on the front cover with Williams' name. Williams sent out a second advance notice of this method to his entire mailing list in June, 1981. It was generally the same notice that Snider and the other 10 customers had received. The distribution of these copies was limited to about 75 customers and these copies were distributed in November or December, 1981. The price for this edition was $3,000.

About this time, and after he had completed the floor trading manual, Williams began to develop a computer program for commodities trading which he called "Striker." He hired an experienced programmer to help him, James Stack, of Kalispell, Montana. Stack owned and operated his own investment research firm, Invest-Tech, and also had considerable experience in the commodities market.

Sometime in December, 1981, Snider, the trader from Oklahoma, called Williams. Snider told Williams he had made money using the FTM, but that he was losing his eyesight and would not be able to use the method much longer. Snider asked Williams for permission to computer program the method. Williams gave him permission on the condition he not disclose the method.

After receiving Williams' permission, Snider contacted the defendant, George Arndt, in December, 1981. He asked Arndt if he could write a program for Williams' method for an Apple II computer. Arndt told Snider he had heard of Williams but was not aware of his method. Snider assured Arndt the system worked. He sent a copy of FTM and some notes to Arndt with a check for $250. Arndt reviewed the manual and called Snider, telling him it looked like a good system and should be done right. He needed more time and more money for the task. Snider sent Arndt another $500. A few weeks later, Arndt again called Snider and said he was still having trouble finishing the program. He told Snider he had spent many hours on the program and requested another $750. Sometime before March 22, 1982 Snider received a floppy disk from Arndt. Snider then sent another $750 to Arndt on March 22, 1982.

Snider tried the program he received from Arndt, but it did not generate the same results as the FTM. He complained to Arndt that the program did not function properly. Arndt asked for more money to complete the job, but Snider refused, asking for his money back. Arndt declined to refund any money.

Sometime after he sent Arndt his last check on March 22, 1982, Snider learned that Arndt was selling the FTM computer program. Snider, annoyed because he had not received a usable program, called Arndt. He demanded to know what right Arndt had to sell the program which did not belong to him. Arndt told him he had checked with his attorney who had researched it and had advised him that Williams had failed to copyright the FTM.

Snider then called Williams and asked if the FTM was copyrighted. He told Williams that Arndt had told him that it was not copyrighted. This was not the only notice Williams received that the FTM was available in a computer program. At this time, Williams was working on the "Striker" program with James Stack. Stack had also taken on the task of responding to phone calls from customers who would call Williams on the "hot line" service Williams offered together with the FTM. Stack received a call from an irate purchaser of the FTM who complained that the floor traders method was being offered on a computer disk without any restriction on its distribution. Williams was engaged in a political campaign at this time and was away from the office a great deal of time during this period.

Stack called Arndt directly about the sale of the program. Arndt said he did not realize he was doing anything wrong because he believed the method was not copyrighted. Arndt asked Stack to talk to Williams about some arrangement which would allow Arndt to sell the program. Stack told Arndt he doubted Williams would do that. Arndt told Stack he would not sell the program without permission.

A short time later, with Stack present, Williams called Arndt. Using a speaker phone, and without identifying himself, Williams asked if he could buy the Larry Williams floor traders method. Arndt answered that it was not the Larry Williams floor traders method, but produced the same results. He offered to send further information.

In May, 1982, Arndt sent out a notice of his computer program. This notice was entitled "Floor Trader's System" and was underscored with "SOLD RECENTLY FOR $2,500 ON PAPER." The notice went on to state that Arndt had "PROGRAMMED the FLOOR TRADER'S System" and while it only took 8 hours to get the system running, it took over 150 hours to get it to "today's point." The notice said further "I offer the system to help you make $. The FLOOR TRADER WAS NOT COPYRIGHTED. No AUTHOR'S name on the original. We do have something HOT here." On the reverse side of the notice were examples of the program's actual results. The program was offered at $1,000. Williams and Stack both received copies of this notice. Arndt also inserted an advertisement for "FLOOR TRADER" in Commodities, a magazine of futures trading, in June, 1982. That advertisement stated "Profits over $100,000 last 6 months of 1981. Plus over $30,000 first 2 months." It went on to say "Many paid $2,500 for this System written down, (but it was never copyrighted) now you can trade it perfectly for only $1,000."

Arndt had no basis of his own for representing profits of $100,000 in late 1981, and $30,000 for January and February, 1982. He had received that information from Edwin Walter, a commodity broker from Chicago. Walter had ordered the FTM from Williams when it first came out in 1981. He studied the method, practiced it, and used it to trade for a few clients. He found it had performed extremely well and published the results for January and February, 1982. Later, he received a copy of the same results from Arndt on which Arndt had written "These are published results. The Floor Trader sold for $2,500 on paper. Now computer accuracy." He also received a document from George Arndt called "Floor Trader" Results for January & February, 1982. That document contained a handwritten addition that "the Floor Trader sold for $2,500 on paper. Now computer accuracy." Customers had called Walter to ask his advice on whether to buy the Williams FTM or the Arndt Floor Trader program. He called Williams, believing that Williams had made some marketing arrangement with Arndt. Williams told Walter he would look into it.

On May 14, 1982, James Vidal, attorney for Williams, wrote to Arndt to demand he discontinue selling the Floor Trader computer program. A similar letter was sent on June 10, 1982. Thereafter, there were letters on August 23, 1982 and March 24, 1983 from Vidal to Arndt's attorney, Irving Kriegsman, warning of legal action if Arndt persisted in his sale of the program.

In June, 1982, Williams called Arndt and wrote to Arndt directly and told him to stop selling the program because it was his property. Williams refused to enter into any arrangements under which Arndt could sell the method. Stack also called after receiving the promotion and warned Arndt that Williams would bring suit unless Arndt stopped selling the program. Arndt continued to sell the program, which he called "Floor Trader" from April to November, 1982. He sold 46 copies at a price of $1,000 each. After November, 1982, Arndt offered the same program under the name "Trend Counter Trend." He sold 98 copies of this disk at $2,500 for a total of $245,000.

By this time, both Williams and Arndt had attracted the attention of other commodities traders. One publication, Club 3000 News by B.A. Thunman, described Arndt as being "of Floor Trader fame" in its publication of November 20, 1983. The Commodities issue of August, 1983 (Exhibit # 12) contained praise for the Larry Williams Floor Trader system and went on to say specifically:

".... but Floor Trader is no longer for sale. However, Thunman says he hears the Trend Counter Trend system offered by George Arndt of Investor's Micro Software generates exactly the same result."

On August 10, 1983, Williams responded to the Commodities article, thanking them for their kind words, but pointing out that Arndt's program was an exact duplicate and Arndt was selling it on an unauthorized basis. Williams advised that legal proceedings had begun and readers should be cautious. This letter, which forms the basis of the libel counterclaim, was published by Futures in its October, 1983 issue. Futures was the successor to the Commodities magazine.

On October 20, 1983, Williams applied for a Certificate of Registration of Copyright for the FTCM, # TX-1-197-066. This lawsuit was filed on November 1, 1983, charging copyright infringement of the FTCM only. In the course of pre-trial discovery, Williams was shown some documents as a result of which he recalled that he had sent out the unedited, 11 copies of the first FTM which did not contain any copyright mark to Snider and other preferred customers. He immediately prepared a letter to all those customers and enclosed a copyright sticker to be attached to the manual. Those letters were sent to all those customers, including Walter, in March, 1984. On September 17, 1984, Williams filed two applications for Certificates of Registration of Copyright, one for the first, unrevised FTM sent to 11 customers, # TX-1-413-121, and a second, # TX-1-462-942, supplementing the earlier certificate for the FTCM. This latter supplementary registration corrected the text errors, explained that existence of this material (the 11 FTM copies) was unknown when the October, 1983, application was filed and described the work as a derivative work. The complaint was then amended to conform to these factual developments.

Finally many of the trading systems that are available to market traders focus on market trends. The FTM was a unique system because, while it also attempted to predict market trends from historical data, it alone contained 7 separate and unique features. No other system contained these features except the programs sold by Arndt, Floor Trader and Trend Counter Trend. When the results of the FTM calculated by hand, and the results of Arndt's computer programs were compared, the vast majority of signals were identical. While there were some inconsistencies in the original FTM manual and the edited FTCM manual, those errors would become obvious to an experienced investor who would then take the corrective action necessary to resolve the conflict and understand the result. The output of Williams' manuals and Arndt's programs was substantially the same, even though one was written in English and one was written in a computer language or source code. The program, Trend Counter Trend, contained an additional feature called prioritization. This prioritizing resolves conflicts in the program. It does not alter the result in any way, but simply makes the choices for the user that the user would have to make when presented with the alternatives. While some of the steps of both programs may come at different times in the program, both programs contain essentially the same steps. They call for the entry of historical data to generate signals. These signals are filtered through certain checks into a sorting process, resulting in a signal to be used by the investor to make his trading decisions and entered with the broker the following morning. It is the presence of the 7 special features in both which is significant. Both systems have 4 distinct entry signals and both systems have qualifiers and variations which provide a signal at which a trade can be made or when it should be made. Both systems generate a signal for a possible entry during the day and a second possible entry at the close. In overall structure, the two competing systems are substantially similar.

II.

While the dispositive issues posed by the plaintiff's action involve the validity of the copyright and whether it was infringed, the defendants raised at the beginning a preliminary and novel issue as well. They claim the source code devised by the defendant, Arndt, for the computer program was a new and different expression of the idea of a market trading system. O'Neill v. Dell Publishing Co., Inc., 630 F.2d 685 (1st Cir.1980) (copyright protection extends only to expression of idea and does not protect the idea itself).

I address these issues seriatim.

III.

(1) The Source Code As a New Expression

To understand this issue, some elementary background is needed. A computer program is a set of detailed instructions by which the computer performs certain functions or delivers a desired result. While the program usually involves very complicated problems and procedures, these same problems and procedures could be solved with a pencil and paper. The difference, of course, is the very long time it would require to perform the same functions manually that the computer, properly programmed, would perform in a matter of seconds.

Also, unlike the human mind, the computer can only perform one function at a time. It performs in sequence according to the step-by-step pattern that has been determined by the programmer. The computer can follow the same set of instructions over and over again or the program can be enlarged or modified to accomplish other objectives or reach other results. The programmer must know, then, what information must be put into the computer, what information is desired from the computer, and the steps the computer must follow to solve the problem at hand. Not only can the computer do simple arithmetic, it is also capable of making decisions if the programmer tells the computer what decision is desired. The computer "decides" by making comparisons, such as the lowest price, or the largest amount, or the best odds. The computer can also "decide" by printing an order once the programmed conditions have been met.

The system at issue here is designed to make predictions in the commodities market by analyzing prices and trends over a period of time. The question is not whether it is a sensible or accurate system of predicting market trends. Obviously, there is no fool-proof system of market trading. But it is possible to store an enormous amount of information in the computer, recall it at the touch of a button, analyze it in terms of trends and cycles, and make the calculations and comparisons to be used by the trader in arriving at his trading decision.

Those steps must be written in language the computer understands and can follow. The programmer writes a program in the selected computer language, using what is called a source code. A program written in a source code is a source program that the computer will translate into machine language using its compiling and assembling facilities. While there may be creativity and ingenuity in devising this source code, a source code is not an entirely new, unique expression of ideas. It is simply a translation from one language to another.

The computer programmer writes in computer language the commands necessary to implement the direction provided in the narrative. The computer language can be compared to any foreign language. Pezzillo, et al. v. General Telephone and Electronics Information System, Inc., 414 F.Supp. 1257 (M.D.Tenn.1976).

Writing the program may be a task requiring a great deal of inventiveness, or it may be merely a clerical function. To a skilled programmer, the conversion of known input, known output, the mathematical expressions needed and the methods of transferring those expressions into computer language is necessarily a mere clerical function. In the Matter of the Application of John W.C. Sherwood, 613 F.2d 809 (USCCPA 1980).

I conclude the source code in this case is not a different and unique expression of a well known idea. It made it possible for an experienced trader to reach a decision in a matter of minutes, rather than an hour. While the skilled programmer can provide flexibility, neatness, and clarity in arranging the order of the system, the programmer, no matter how talented, does not express creativity, imagination, independent thought and uniqueness. While the quality of the program is determined by the quality of the programmer, it still amounts to a translation. If it were otherwise, every expression would be open to copying in source code form.

(2) Was the Plaintiff's Copyright Valid

I conclude that both manuals, the original "Floor Trader's Manual" and the revised "Floor Trader's Commodity Method" were copyrightable works. The manuals were the original creative effort of Williams alone, an arrangement of words he used to express his ideas. O'Neill v. Dell Publishing Co., 630 F.2d 685 (1st Cir. 1980). They were a detailed, step-by-step procedure or process to accomplish a specific desired result. Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240, 1250-1252 (3rd Cir.1983), cert. dismissed, 464 U.S. 1033, 104 S.Ct. 690, 79 L.Ed.2d 158 (1984); 1 Nimmer on Copyright, § 2:03[D] (1984). The FTCM was the corrected and edited final draft of the FTM. Since the FTCM was derived from the FTM, as a derivative work it also fell within the protection of the copyright laws. 17 U.S.C. § 106(2).

I further conclude that notice of copyright on the original FTM was not required because those 11 copies were not publicly distributed. 17 U.S.C. § 405(a). These works were not made available to the public generally. They were limited to a definite, very selective group for a limited purpose on the condition that the contents not be disclosed or further disseminated. Schuchart & Assoc. v. Cold Serve Corp., 220 U.S.P.Q. 170, 181 (W.D.Texas, 1983); 1 Nimmer on Copyright, § 4:13[A]. Further, only 11 copies were distributed, "a relatively small number of copies." 17 U.S.C. § 405(a)(1). In my judgment Williams' action did not amount to publication. E. Harold Munn, Jr. & Assoc., Inc. v. Minority Broadcasting Co. of Midwest, Inc., No. 82-0759 (N.D.Ill. Dec. 14, 1982).

Finally, Williams made a reasonable effort to add copyright notice to those 11 copies by mailing a letter with a sticker to be affixed to those copies. He did not become aware of, or did not recall, the distribution of those copies until more than 3 years later, but, once he recalled the circumstances, he made a reasonable effort to add the notice. 17 U.S.C. § 405(a)(2). E. Harold Munn, Jr. & Assoc., Inc. v. Minority Broadcasting Co. of Midwest, Inc., supra.

Both the FTM and the FTCM have been granted Certificates of Copyright Registration by the United States Copyright Office. Copies of those manuals have been deposited with the Register of Copyrights. 17 U.S.C. § 410(a). The defendants have alleged the affirmative defense of unclean hands because of certain omissions and concealment in the application for copyright registration. The defendants have failed to prove that any omission was an intentional, knowing or purposeful concealment or that the Copyright Office might have rejected the application had it known of the omission. Eckes v. Card Prices Update, 222 U.S.P.Q. 762, 764; 736 F.2d 859 (2nd Cir.1984). As stated herein, Williams did not recall that the FTM which lacked the copyright notice had been sent to Snider until after this litigation began. When he did recall that fact, he took immediate steps to rectify the situation. The defendants are not entitled to the defense of unclean hands under the facts of this case.

Accordingly, as to this issue, I conclude the plaintiff's copyright was valid.

(3) Was the Plaintiff's Copyright Infringed

The burden of proof in a copyright infringement action is on the plaintiff. 3 Nimmer, The Law of Copyright, § 13.01 (1984). He must prove each of two elements of copyright infringement: (1) ownership of the copyright; and (2) copying by the defendants. Marcelo Ramos Motta, Inc., et al. v. Samuel Weiser, Inc., 768 F.2d 481 (1st Cir.1985). To prove copying, the plaintiff must show access and substantial similarity. O'Neill v. Dell Publishing Co., Inc., supra, at 686; Sid & Marty Krofft Television Productions, Inc. v. McDonald's Corp., 562 F.2d 1157, 1164 (9th Cir.1977).

I conclude the plaintiff has satisfied his burden of showing both access and substantial similarity. There is no doubt that Arndt had access to FTM, not only from a copy he testified he received from an unknown sender, but also from Snider. As discussed above, the errors or inconsistencies in the text were not significant.

As to substantial similarity, the question is whether Arndt's works can be recognized by a reasonable observer as having been copied from the Williams' manual. Original Appalachian Artworks, Inc. v. Toy-Loft, Inc., 684 F.2d 821, 829 (11th Cir. 1982). As discussed above, it is not a defense that the Arndt program was expressed in a computer source code. That source code contained similarities which generated identical signals in the vast majority of comparisons. Here, Arndt merely translated Williams work from English into computer language (in this case, Basic) which produced substantially similar results.

The most graphic evidence of substantial similarity was the comparison of the FTM with both programs Arndt had offered, Floor Trader and Trend Counter Trend. Picking a commodity — live cattle — in February, 1983, James Stack went through a step-by-step process of both systems, illustrating his steps by graphs and a "chalk" which were received in evidence. The FTM result was essentially the same as the result reached by running the Trend Counter Trend system (Exhibit # 32).

Accordingly, I conclude the plaintiff's copyrights of the Floor Trader's Manual and the Floor Trader's Commodity Method were infringed.

One further note is in order. The basic issue in this case, as in most cases, turns on the question of credibility. In this non-jury trial, that assessment is left to me. Generally, I have accepted the plaintiff's version of the facts as more credible. Because the issues here were hotly contested and, at times, somewhat frustrating, I set out my reasons for accepting the plaintiff's version.

As for the principals, Larry Williams showed an extensive, proven experience in the commodities market as a trader, author, teacher and adviser for 15 years. His testimony as to his background was supported by documents, corroborating testimony, and favorable notices in trade journals. His demeanor impressed me as one of candor, fairness and truthfulness. His work in this field was amply demonstrated, not only by the developing of the manuals at issue here, but by his seminars and publications dating back to 1974.

On the other hand, George Arndt told conflicting stories. When called as a witness by the plaintiff, he said he worked largely as a sales representative for Digital, Techtronics and Raytheon, interrupted by a short period of 14 months as an independent sales representative and antiques auctioneer. He had experience writing computer programs, but had no experience as a commodities trader and never earned a living by providing advice on the commodities market. He testified he received the Floor Trader Manual in the mail from an unknown source, but he used it in connection with his software program and incorporated many ideas from the manual into his software program. He later received the same manual from Snider, together with handwritten notes, and used them, his own experience and program he had written at Christmas, 1981, to write his Floor Trader program. Under this adverse questioning, he admitted his purpose was to generate the same returns as the Floor Trader Manual and he admitted he programmed the manual he received from Snider. After receiving complaints from Williams that he was copying the manual, he testified that he consulted an attorney and continued to distribute his computer program on the advice of counsel. Following this direct examination, he was not examined by his counsel. Because of the press of other court business, the case was adjourned on July 26, 1985, after the plaintiff rested its case. It was resumed on September 5, 1985.

On resumption of the trial some five weeks later, Arndt was called as the first witness in his defense. While his educational and work experience was essentially the same, he told a very different story about his commodities experience. He testified he had traded in the commodities market since he was 14 years old, and continued throughout high school, college and graduate school. He testified he "created" Floor Trader and Trend Counter Trend and before those, he had written 75 programs of trading strategy, with updates provided 4 times a year. Over a period of 10 years, he had written hundreds of programs telling people when to buy and when to sell. Many were offered to the public for sale. In mid-1981, he went into the business full-time, leaving his position as senior sales representative at Digital. He founded Harvard Investment Service, Inc.

Arndt then turned to the FTM. Because people knew he was writing his own software system, he assumed someone sent him the FTM. He skimmed it and found it impossible to understand and full of mistakes. When he received the Snider manual, he found similar inconsistencies and mistakes. The signals were conflicting, and contained no prioritization. Using Snider's "ideas" and "concepts," he continued to work on his own program to make it "unique." He testified it took a tremendous amount of commodity market experience and a tremendous amount of "creativity." Finally, after over 500 hours of work, and 38 hours of uninterrupted work at one stretch, the "expression of his ideas" came out on the computer screen. In his "creative frenzy" he had "created" his own program. He spent 500 more hours perfecting his system. Basically, he testified that he never read the Larry Williams system, and the Snider manual was of no use. He had started from scratch. This appeared to me to be a total abandonment of his earlier contention, namely, that the source code was such a different expression, there could never be substantial similarity.

I looked for some corroboration of Arndt's version and found nothing of substance. Arndt did not produce one program to support his assertion that he created hundreds. He did not produce any supporting evidence of market experience, even losses. He produced no witnesses, no pamphlets, no charts, no notices to support his assertions of a lifetime of investment advice experience. In short, he asked me to take his testimony, unsupported and unadorned. I found his testimony not credible.

Turning to the expert testimony, I was presented with the same conflicts. The plaintiff's single expert owns and operates an investment research firm. He met Larry Williams in 1979 and in 1981, Larry Williams asked him to program his commodities market program "Striker." He became aware of the Floor Trader Manual when people called in on the "Hot Line" for information or with questions, a service that Williams made available to his purchasers. He explained the Williams system and its unique features. Of the thousands of trading systems with which he is familiar, only one has these same 7 features — Arndt's Trend Counter Trend. He performed the calculations set out in the Floor Trader Manual by hand, then ran the Trend Counter Trend program on his computer and did the same with Arndt's early Floor Trader, and, in his opinion, they were substantially the same. They contained at least 6 of the unique features and generated signals that were 75% the same. The inconsistencies, he testified, were errors which would be obvious to the experienced investors who read the system as a whole. This expert then demonstrated the step-by-step procedure, using data from the Wall Street Journal, achieving the same result. His calculations were demonstrated by exhibits received in evidence as well as "chalks" he used to perform his comparison.

He then ran the Arndt program, Trend Counter Trend, and the result was the same. The prioritization in the Arndt program was an additional feature which resolved conflicts. Whereas a person would resolve these conflicts by making a choice of alternatives, the computer did it for the user. It was a desirable feature, but not a significant one, he testified. In his opinion, the source code meant nothing since the program can be written in various ways. But when the output is the same, the program is the same. The expert supported his testimony by graphs and charts outlining his work.

The defendant produced two experts. Both had experience with computers. The first had no experience in the commodities market. He had compared the source code with the Floor Trader Manual and found them quite different. The FTM was not a complete document from which to make a complete program. He came to the opinion that the person who wrote the source program had read the manual, but had not followed it one hundred percent.

The second expert had minimal experience in writing programs although he traded in commodities on a full time basis. He examined the Floor Traders Manual and found it was not an effective system because it did not have prioritization, was confusing and contained ambiguities. He came to the opinion that the Floor Trader Manual and the source code were entirely dissimilar, organized and structured in a completely different way and the FTM did not contain prioritization. He ran the systems and found that 70% of the signals were the same. He kept no records, had no charts or graphs and had no results to support his testimony.

On the basis of preparation, corroboration, fairness, candor and experience, I accepted the plaintiff's expert as clearly more credible.

Having concluded that the plaintiff's copyrights were infringed, the next question is whether the defendants were innocent infringers and thus shielded from liability for actual or statutory damages under 17 U.S.C. § 405(b). I believe they are not so protected because they have failed to satisfy their burden of proof that they were innocent infringers. 17 U.S.C. § 405(b). Arndt did not prove that he acted in good faith or reasonably relied upon the absence of a copyright notice. 2 Nimmer on Copyright, § 7.14[B][i] and [ii]. Beginning in early 1982, from his conversations with Snider, and certainly by June, 1982 when Williams wrote directly to Arndt and Stack had called him, Arndt was on notice that he was copying the Williams system. He attempted to make some arrangement with Williams and, rebuffed in this attempt, he nevertheless continued to distribute his program. Despite the absence of the copyright notice on the FTM he received from Snider, Arndt had "... reason to think otherwise." H.R.Rep. No. 1476, at 148, U.S.Code Cong. & Admin. News 1976, pp. 5659, 5764.

Finally, the defendants have moved to dismiss Count V of the plaintiff's complaint asserting that this Court lacks subject matter jurisdiction over any claim under M.G.L. c. 93A. They based their motion on two grounds. First, they say c. 93A is pre-empted by federal law. 17 U.S.C. § 301. Secondly, they say the transaction in issue did not occur primarily and substantially in Massachusetts as required by M.G.L. c. 93A, § 3(1)(b). I believe they are wrong on both counts, but I do not need to engage in an extensive discussion of those issues because I conclude that the defendants' actions did not amount to conduct which "Attain(s) a level of rascality that would raise an eyebrow of someone inured to the rough and tumble of the world of commerce." Levings v. Forbes & Wallace, Inc., 396 N.E.2d 149, 153, 8 Mass. App. 498 (1979). I have in mind that, while Arndt was on notice of the copying claim in June, 1982, he had received the FTM from Snider which did not contain a copyright notice. And, when he heard from Stack and Williams, he sought the advice of an attorney and proceeded on the basis of counsel's advice. While acting on counsel's advice does not excuse him as an innocent infringer, it does negate, in my judgment, the type of unscrupulous, intolerable and unethical behavior proscribed by c. 93A. J.A. Westerbeke Corp. v. Onan Corp., 580 F.Supp. 1173 (D.Mass.1984).

As to the defendant Arndt's counter-claim, I conclude Arndt has failed to prove by a preponderance of the evidence that he was defamed by the publication of the Williams letter in the October 1983 issue of Futures. The contents of the letter were not false. Further, there was no credible evidence that Arndt was injured or discredited in any way.

IV.

The final matter is the subject of damages for the infringement. The copyright owner is entitled to recover actual damages suffered and, in addition, any profits of the infringer. 17 U.S.C. § 504(b). I have found that the defendants sold 46 copies of the FTM program at $1,000 each for a total of $46,000 and 98 copies of the Trend Counter Trend program for a total of $245,000. The total gross profit realized by the infringing defendants was, then, $291,000. On the facts above, I cannot credit the defendants with any deductible expenses. Arndt kept no credible records. He claimed to have recalculated his expenses and income from the numerous checks he claimed were available to him and to his brother in Chicago, an accountant. The income statements for the years 1982, 1983 and 1984 (Exhibit # 45) were completely unsupported, were prepared for use in this litigation and, in short, were completely unreliable. The defendants have not proven any deductible expenses and, therefore, I award the gross revenue from the sale of the infringing programs. Russell v. Price, 205 U.S.P.Q. 206, 213 (9th Cir.1979); Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 609 F.Supp. 1307, 1322 (D.C.PA 1985).

At the same time, I decline to award any damages for the plaintiff's claim of lost revenues from reduced attendance at his seminars, or purchase of his written material. Given the dissemination of his own manuals, as well as other programs, I have no substantial basis for any such claim and any award would be uncertain and speculative.

In summary, I award compensatory damages in the amount of $291,000. I do not award attorneys' fees. Again, I have in mind that the defendants acted pursuant to counsel's advice and, in the exercise of my discretion, I believe this case presents sufficient unusual circumstances to avoid attorneys' fees. I do, however, award costs to the plaintiff. 17 U.S.C. § 505.

The award of damages and costs will be against both defendants jointly and severally. The evidence is clear that Arndt is the founder, sole stockholder and only operating officer of Harvard Investment Service, Inc. and that the facilities of Harvard Investment Service, Inc. were used freely by Arndt in his infringing activity.

Conclusion

In accordance with the above, judgment will be entered as follows:

(1) Judgment on Counts I, II, III and IV is entered in favor of the plaintiff, Larry Williams, and against the defendants, George Arndt and Harvard Investment Services, Inc., jointly and severally, in the sum of $291,000.

(2) The defendants, George Arndt and Harvard Investment Services, Inc., jointly and severally and individually, are enjoined and restrained from marketing, advertising, selling or licensing the corporate programs Floor Trader and Trend Counter Trend.

(3) The defendants, George Arndt and Harvard Investment Service, Inc., jointly and severally, shall pay plaintiff's costs.

(4) On Count V, judgment will be entered for the defendants.

(5) On the defendant's counterclaim for defamation, judgment will be entered for the plaintiff, defendant-in-counterclaim.

(6) The plaintiff, Larry Williams, is awarded prejudgment interest.

SO ORDERED.