849 F.2d 585 | Fed. Cir. | 1988
Lead Opinion
Randomex, Inc. appeals the final judgment of the district court dismissing Randomex’s complaint because of the invalidity of United States Letters Patent No. 3,803,-660 (’660) for failure to disclose the “best mode.” See Randomex Inc. v. Scopus Corp., 2 USPQ2d 1622, 1627 (D.Mass.1987). Randomex also appeals a prior district court order setting aside a jury award of damages to Randomex, and ordering a new trial for damages under Federal Rule of Civil Procedure 59(a). Scopus Corp. and Dennis Haskamp (collectively Scopus) cross-appeal the district court’s failure to hold the ’660 patent unenforceable for patent misuse. We vacate the judgment dismissing the complaint, reverse the holding of invalidity of the ’660 patent, and, having no jurisdiction on the damages issue or the
BACKGROUND
Familiarity with the district court’s opinion is assumed. In mainframe computers information is stored on magnetic disks, several of which are housed in a disk pack. A small particle of dust or dirt on a disk may cause physical damage to the disk or loss of data. Before Randomex’s invention, only large, nonportable machines were available to clean disk packs. The disk packs had to be transported to the machine which was undesirable because the delicate disks could be damaged in transit. All types of these various cleaning systems used some type of cloth or brush in conjunction with a cleaning fluid. One of the nonportable cleaning machines used a 91% alcohol solution to clean the disks; another used diluted surgical detergent.
The ’660 patent is directed to a portable apparatus for cleaning disk packs. A disk pack is removed from the computer and placed in the apparatus where brushes wipe the disks clean, whereupon a cleaning solution is sprayed on the brushes to remove the accumulated dirt. Although the cleaning solution is not claimed specifically in the ’660 patent, it is needed to practice the invention. Thus, the patent disclosed:
[t]he cleaning solution employed should be of a type adequate to clean grease and oil from the disc surfaces, such as a 91 percent alcohol solution or a non-residue detergent solution such as Rando-mex Cleaner No. 50281.
Column 5, lines 49-53.
At first, Scopus was in the business of transporting disk packs to a facility where they were cleaned. When Randomex began to market its portable cleaner, Scopus purchased several and changed the way it did business. Instead of transporting the disk packs, it transported cleaners, cleaning the disk packs on its customers’ premises, thereby greatly reducing costs. After experiencing some trouble with the reliability of the Randomex cleaners, Scopus discontinued buying the Randomex cleaners, built its own “ruggedized” version, and “reverse engineered” Randomex’s cleaning solution by obtaining an analysis of it from a professional chemist.
Randomex sued Scopus for infringement of the ’660 patent. At trial Scopus stipulated to infringement and the jury properly found the infringement willful. After the district court directed a verdict in Rando-mex’s favor on several of Scopus’ defenses, the jury returned a special verdict, Federal Rule of Civil Procedure 49(a), answering three of the questions submitted to it as follows:
I.
2. Was the patent’s disclosure with respect to cleaning fluid so inadequate that a person skilled in the art who did not use plaintiff’s named cleaner would have had to engage in an undue amount of experimentation,
a) To use the invention?
Yes_ No X
b) To find the best mode to use the invention?
Yes_ No X
[Hereinafter Q2]
II.
3. Did the applicants deliberately refrain from informing users of the invention of the best cleaner formula with the intent that, to a substantial extent, users would be led to purchase plaintiff’s cleaner rather than to experiment themselves to find the best?
Yes _X_ No_
a) Was that a reasonable expectation?
Yes _X_ 'No_
[Hereinafter Q3]
III.
6----[W]hat were [Randomex’s] damages?
$1,566,000.00
[Hereinafter Q6]
Post trial activity flourished and, pursuant to its memoranda and orders of July 25, October 3, and December 4,1986, the court
On March 6, 1987, the district court issued another opinion in which it concluded that because Q3, as opposed to Q2, was a better indicator of whether the best mode requirement of 35 U.S.C. § 112 ¶ 1 (1982) had been fulfilled, the best mode of practicing the invention had not been disclosed in the ’660 patent and the patent was invalid. Accordingly, the interlocutory judgment was vacated and final judgment was entered on March 20, 1987, dismissing the complaint because of the invalidity of the ’660 patent.
ISSUES
1. Whether the district court erred in determining that the patent in suit fails to disclose the best mode for practicing the invention.
2. Whether the district court erred in granting a motion for new trial on the issue of damages and in concluding that the patent in suit was not unenforceable for patent misuse.
OPINION
I. Best Mode
A. The Law
“The specification ... shall set forth the best mode contemplated by the inventor of carrying out his invention.” 35 U.S.C. § 112 ¶ 1. This disclosure requirement is directed to persons “of ordinary skill in the art at the time the application was filed.” W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1556-57, 220 USPQ 303, 316 (Fed.Cir.1983), cert. denied, 469 U.S. 851, 105 S.Ct. 172, 83 L.Ed.2d 107 (1984). “Because not complying with the best mode requirement amounts to concealing the preferred mode contemplated by the applicant at the time of filing, in order to find that the best mode requirement is not satisfied, it must be shown that the applicant knew of and concealed a better mode than he disclosed.” Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384-85, 231 USPQ 81, 94 (Fed.Cir.1986), cert. denied, — U.S.-, 107 S.Ct. 1606-07, 94 L.Ed.2d 792 (1987). Although there may be a general reference in the patent to the best mode, “the quality of [that] disclosure may be so poor as to effectively result in concealment.” Spectra-Physics, Inc. v. Coherent Inc., 827 F.2d 1524, 1536, 3 USPQ2d 1737, 1745 (Fed.Cir.), cert. denied, — U.S.-, 108 S.Ct. 346, 98 L.Ed.2d 372 (1987).
B. The District Court Opinion
Considering the experience and eminence of the district court, its vacillations in this case are remarkable. In early July 1986, it posited the question:
Is it a misuse of the patent deliberately to omit the formula for the best use in order to achieve psychologically tie-in sales of plaintiff’s unpatented product that would persuade even a sophisticated user to purchase from plaintiff even though, in fact, the user could have discovered the formula by reasonable experimentation? Put another way, while it is sometimes said that the duty of fairness is owed to the patent examiner, this is but a euphemism — the duty is owed to the public. The question is whether it is inconsistent with that duty to make deliberately limited disclosure in order to sell unpatentable products even though, by experimentation, the public could remedy the deficiency. The argument against it is that it is de minimis. The argument for it is that it was done deliberately— the court expressly asked that word of Mr. Ludka. In Minerals Separation, Ltd. v. Hyde, 242 U.S. 261 [37 S.Ct. 82, 61 L.Ed. 286] (1916), the Court stated that an applicant need only make reasonable disclosure, but there there was good*588 reason why he did not do better. Here, there was only a bad reason.
Does this reflect back to mean failure to state best mode of use as matter of law or, at least, an inconsistency?
In its July 25, 1986 memorandum and order, the trial court answered the question in the negative by stating, “I find the patent not invalid for failure to disclose the best mode.”
In its subsequent order and opinion on appeal here, the district court totally reversed itself. It held that although there were a number of non-residue detergent solutions on the market, “[i]t may be assumed that the inventor considered his own the best, but it is not ‘set forth’ by reference to Randomex Cleaner No. 50281.” Randomex, 2 USPQ2d at 1623. The court opined that Randomex “kept secret [its solution formula] in order to sell the product to users of its device.” Id. It concluded that “the fact that experimentation by obtaining a chemical analysis of plaintiff’s Cleaner solution ... could reveal plaintiff’s formula could not, as matter of law, satisfy the statutory requirement.” Id. It based its conclusion on the following factors: (1) Randomex was under no obligation to make its fluid available for such reverse engineering; (2) there was nothing to keep Randomex from changing its formula; and (3) a patent is supposed to fully educate the public for use following the expiration date and who, after seventeen years, could determine what had been plaintiff’s proprietary Cleaner. Id. The district court held, accordingly, that “[s]o far as best mode is concerned, the reference to ‘Randomex cleaner No. 50281’ was legally insufficient.” Id.
The district court then undertook an equitable approach, id. at 1625, characterizing this case as one involving “intentional concealment for the purpose of making collateral profits on the invention.” Id. at 1626. The court found the testimony that the nondisclosure of the formula “was just an advertising ‘gimmick,’ and that users would soon catch on to the formula, ... was inconsistent with the jury’s answers to questions 3 and 3(a).” Id. Thus, the court concluded “that, viewing the record as favorably to plaintiff as the evidence and the jury findings permit, intentional concealment of the best mode contemplated by the inventor requires that the patent be declared invalid.” Id.
C. The Jury Questions
The district court erred in concluding that Q3, as opposed to Q2, was a better indicator of whether the best mode requirement had been fulfilled. Although Q3 incorporates some of the elements of a proper best mode analysis, the question was not designed to be, and was never intended to be, dispositive of the best mode issue. The district court devised this question to deal with a patent misuse issue. In fact, the district court once denied Q3 had anything to do with best mode. Joint Appendix at 1301.
As a patent misuse jury question, Q3 is less than satisfactory. It tells the jury that “the applicants deliberately refrain[ed] from informing users of the invention of the best cleaner formula____” See Q3. The jury only found the obvious when determining that the inventors failed to disclose their cleaner formula for the purpose of leading users to purchase their cleaner.
As a best mode question it is also faulty, because the question is directed not to persons skilled in the art, W.L. Gore, 721 F.2d at 1556-57, 220 USPQ at 316, but to “customers.” See Christianson v. Colt Indus. Operating Corp., 822 F.2d 1544, 1563, 3 USPQ2d 1241, 1255 (Fed.Cir.) (the law requires inventor not to conceal best mode of practicing claimed invention; best mode does not deal with sales to customers having particular requirements), cert. granted in part, — U.S. -, 108 S.Ct. 500, 98 L.Ed.2d 499 (1987). The district court was overly concerned with “users of the device,” Randomex, 2 USPQ2d at 1626, who in this instance are not synonymous with those of ordinary skill in the art.
It is concealment of the best mode of practicing the claimed invention that section 112 TT1 is designed to prohibit. Here the claimed invention is a portable machine for cleaning computer disk packs. Those
D. Quality of Disclosure
Although Q2 is poorly framed and is directed more to enablement than best mode, we must determine if Q2 is so legally deficient that in answering that question the jury would have failed to make the underlying inquiries necessary to support its finding that the inventors fulfilled the best mode requirement. Spectra-Physics, 827 F.2d at 1535-36, 3 USPQ2d at 1745. It is undisputed in this case that (1) a 91% alcohol solution was used in a prior art non-portable machine; (2) diluted surgical detergent — a non-residue detergent solution —was used in the prior art non-portable machines; (3) Randomex Cleaner No. 50281 is a non-residue detergent solution; (4) the applicant knew of no better non-residue detergent solution than Randomex Cleaner No. 50281; (6) the evidence tended to show that a 91% alcohol solution was the “worst mode,” not the “best”; (7) Scopus easily “reverse engineered” Randomex’s solution; (8) the inventor intentionally omitted the formula for Randomex Cleaner No. 50281 from the disclosure. Randomex, 2 USPQ2d at 1622.
First, as an initial matter, “[is] the best mode requirement complied with when an inventor discloses his preferred embodiment indiscriminately with other possible embodiments?” 2 D. Chisum, Patents § 7.05[1] at 7-68 (1987). The Board of Patent Appeals and Interferences has stated:
[t]here is no requirement in 35 USC 112 that an applicant point out which of his embodiments he considers his best mode; that the disclosure includes the best mode contemplated by the applicant is enough to satisfy the statute. There is no concealment of best mode here since one of ordinary skill in the art could readily determine the best operating mode....
Ernsthausen v. Nakayama, 1 USPQ2d 1539, 1549 (Bd.Pat.App.Int.1985), aff'd, Nos. 86-736, 86-923 (Fed.Cir. Nov. 24, 1986) [809 F.2d 787, 788 (tables)]. As applicable to the facts of this case, we agree with this statement of the Board.
The disclosure of the 91% alcohol solution is suspect in light of evidence that one who used it might have had to contend with an explosion. Although the motive for including the 91% alcohol solution in the disclosure is not apparent, Randomex, 2 USPQ2d at 1623 & n. 1, the solution was part of the prior art and suggested an easily procurable inferior substitute for the Randomex Cleaner No. 50281. But if the words “Randomex Cleaner No. 50281” satisfy the best mode requirement, we will not invent a rule that would result in the forfeiture of a patent due to the district court’s novel combination of patent misuse and best mode notions. The indiscriminate disclosure in this instance of the preferred cleaning fluid along with one other possible cleaning fluid satisfies the best mode requirement.
Second, the district court held that the reference to “Randomex Cleaner No. 50281” was legally insufficient. Id. at 1623. By this we assume the district court found that the quality of the disclosure was so poor as to effectively result in concealment. Spectra-Physics, 827 F.2d at 1536, 3 USPQ2d at 1745. We disagree. The complete description is “a non-residue detergent solution such as Randomex Cleaner No. 50281.” Randomex, 2 USPQ2d at 1623 (quoting ’660 patent). Although a trade name alone may be inappropriate in a best mode disclosure when suitable substitutes are unavailable, see White Consolidated Indus., Inc. v. Vega Servo-Control, Inc., 713 F.2d 788, 791, 218 USPQ 961, 963
As the jury found by answering Q2 in Randomex’s favor, the patent disclosed the “best mode contemplated by the inventor of carrying out his invention,” 35 U.S.C. § 112 ¶ 1, as directed to one of ordinary skill in the art. W.L. Gore, 721 F.2d at 1556-57, 220 USPQ at 316. This invention neither added nor claimed to add anything to the prior art respecting cleaning fluid. If the disclosure in the ’660 patent had truthfully said “the cleaning solution employed should be of a commercial type adequate to clean grease and oil from disk surfaces, i.e., a non-residue detergent solution such as Randomex Cleaner No. 50281, or equal,” the best mode requirement would have been met without full disclosure of the formula. Because there is no dispute as to the underlying facts, it follows that nondisclosure of the formula per se did not rise to the level of violating the statutory requirement to disclose the best mode.
II. Damages
“An order granting a new trial is interlocutory in nature, and therefore not immediately appealable.” Allied Chem. Corp. v. Daiflon, Inc., 449 U.S. 33, 34, 101 S.Ct. 188, 190, 66 L.Ed.2d 193 (1980). In this case a final judgment was entered, but it was based only on the March 6, 1987, opinion that dealt exclusively with the best mode issue. The October 3, 1986, memorandum and order that granted the new trial on damages was not a basis for the final judgment. Neither was the order for new trial conditional, to become effective only upon reversal of the judgment. See Fed.R.Civ.P. 50(c); 9 C. Wright & A. Miller, Federal Practice and Procedure § 2540 (1971). The same analysis applies to the patent misuse issue on cross-appeal because it arose out of the October 3, 1986, memorandum and order. Thus we have no jurisdiction over the October 3,1986, order.
CONCLUSION
The judgment of the district court is vacated; the holding that the ’660 patent is invalid is reversed; and the case is remanded for proceedings consistent with this opinion.
COSTS
Costs to Randomex.
VACATED, REVERSED, AND REMANDED
For example, if one should invent a new and improved internal combustion engine, the best mode requirement would require a patentee to divulge the fuel on which it would run best. This patentee, however, would not be required to disclose the formula for refining gasoline or any other petroleum product. Every requirement is met if the patentee truthfully stated that the engine ran smoothly and powerfully on Brand X super-premium lead free “or equal." Making engines and refining petroleum are different arts, and the person skilled in the art of making engines would probably buy the suggested gasoline. But if the hypothetical maker or user of the engine did not want to use the Brand X super-premium, he would then explore the "or equal" alternative of the patent disclosure. Practically speaking, he would not buy a test tube of Brand X gasoline and reverse engineer it to determine how Brand X refined it. He would ask dealers what other brands of gasoline are available that, in their view, would do for the patented engine what Brand X did. The user is not driven by lawyer’s ideology. He is a person who elects practical means to accomplish practical ends.
Dissenting Opinion
dissenting.
I dissent because I see a sharper distinction between the best mode and enablement requirements of 35 U.S.C. § 112.
Here, prior to the patent application, one of the patentees and the president of Ran-domex, Ludka, had tried several different cleaning solutions in his quest for the one that worked best. Through experimentation, he learned that “there was a difference in effectiveness in detergent formulae percentages,” and ultimately arrived at an optimum cleaning formula. As he admits, however, he intentionally refrained from disclosing the optimum formula so that the users of his cleaner would also purchase his cleaning solution. 2 USPQ2d at 1626. Ludka believed this intentional non-disclosure was “a good advertising gimmick” that would bring him additional profits from his invention.
The specification of the ’660 patent said, “The cleaning solution employed should be of a type adequate to clean grease and oil from the disc surfaces, such as a 91 percent alcohol solution or a non-residue detergent solution such as Randomex Cleaner No. 50281.” As can readily be seen, this sets out an undifferentiated list of apparently equally effective solutions, no one of which is highlighted as preferred over the others. There would be no problem with this if all of the suggested solutions worked equally well. But using the alcohol solution, an example of an unstated number of other cleaning solutions, did not work well and could produce dangerous fumes:
Ludka ... testified that in experimenting to find the best fluid, he abandoned alcohol after one try. While he maintained at the trial that alcohol was usable, neither he, nor plaintiff’s expert, contradicted defendants’ expert’s testimony that, because plaintiff’s device was enclosed, alcohol could produce dangerous fumes.
2 USPQ2d at 1623 n. 1. Further, the use of certain “non-residue detergent solutions” also could have negative consequences. As Randomex related in a letter to a client:
It is our intent to deliver continuous amounts of our [cleaning] solution across [the disc] surfaces, through the device of the brush bristles, so that the chemical solution itself emulsifies the contamination on the surface, lifts it, and either carries or spins it from the surface____ To achieve proper emulsification, it is mandatory that you provide enough of the chemicals, carried in a deionized water base, to do their emulsifying job. Too little will result in incomplete contaminant removal; the wrong solution will generally result in residue and subsequent head crashes [in the cleaning apparatus]. [Emphasis added in last clause.]
Thus, not only did Ludka fail to point out that the Randomex solution was the best mode, he also disclosed solutions that he knew could be harmful and even dangerous. This is the antithesis of the good-faith full disclosure that is mandated by section 112’s best mode requirement. See In re Nelson, 280 F.2d 172, 184, 47 CCPA 1031, 1048, 126 USPQ 242, 253 (1960) (“One cannot read the wording of section 112 without appreciating that strong language
It is irrelevant that there was a variety of non-residue detergent solutions on the market that might work because Ludka knew that among those available his was the most effective. Yet he only named it as one among many possible solutions, and then only by its trade name, not generically. He did not disclose which of the suggested modes was the “best mode contemplated by him” for practicing the invention. In re Gay, 309 F.2d at 772, 135 USPQ at 315; Hybritech, Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384, 231 USPQ 81, 94 (Fed.Cir.1986). His disclosure does not satisfy section 112 because he buried his best mode in a list of less satisfactory ones. Cf. Spectra-Physics, Inc. v. Coherent, Inc., 827 F.2d 1524, 1536, 3 USPQ2d 1737, 1745 (Fed.Cir.1987) (“Even though there may be a general reference to the best mode, the quality of the disclosure may be so poor as to effectively result in concealment”).
In view of this, question 2(b) was, as the district court concluded in the end, irrelevant. In answer to that question, the jury said that the “disclosure with respect to the cleaning fluid [was not] so inadequate that a person skilled in the art who did not use plaintiff’s named cleaner would have had to engage in an undue amount of experimentation ... to find the best mode to use the invention.” Implicit in this finding is the undeniable premise that Randomex had not disclosed the best mode and therefore one would have to “find” it. But if there is a best mode known to the inventor he must say so; he cannot require the public to hunt for it. See DeGeorge v. Bernier, 768 F.2d 1318, 1324, 226 USPQ 758, 763 (Fed.Cir.1985) (“Not complying with the best mode requirement amounts to concealing the preferred mode contemplated by the applicant at time of filing”). This special verdict did not find that the disclosure was sufficient to teach one skilled in the art how to practice the best mode. It found only that the disclosure was sufficient to teach one skilled in the art how to find the best mode. That is a horse of a different color.
To find the best mode one might have to try any number of solutions akin to 91% alcohol, hopefully without the same damaging results of alcohol, as well as any number of non-residue detergent solutions, among them Randomex Cleaner No. 50281. Having done all that, the searcher presumably would conclude that the Randomex solution was best. But the end of that quest is only the beginning of another. He would then have to “reverse engineer” the Randomex solution, which may or may not be a simple proposition. As the district court observed, Randomex said in a letter to a potential customer that its cleaning solution took “some two years” to develop and its “consulting chemist continues searching for improvements....” See Memorandum to Counsel in Scopus Corp. v. Randomex, Inc., No. 82-1032-MA-A (Oct. 19, 1987).
Question 3 is more pertinent to our inquiry. The best mode aspect of section 112 contemplates that inventors not apply for a patent while “concealing from the public preferred embodiments of their inventions.” In re Gay, 309 F.2d at 772, 135 USPQ at 315. “The public” is everyone, not just those skilled in the art, and embraces “users,” very likely the members of the public who will practice the invention. While a word more comprehensive than “users” might have been used in question 3, the district court was surely correct that it was adequate. In response to Rando-mex’s belated complaint about use of the word to instruct the jury, the trial court said, “While a communication sufficient to inform persons skilled in the art must be sufficient, or there would be no end to the required disclosure, the principle remains that it is the public that is meant to be informed.” 2 USPQ2d at 1626 n. 7.
In any event, Randomex did not timely object to the question and must therefore be deemed to have accepted it, see Fed.R.Civ.P. 51 (“No party may assign as error the giving or failure to give an instruction unless he objects thereto before the jury retires....”); Allen Organ Co. v. Kimball Int’l, Inc., 839 F.2d 1556, 1566, 5 USPQ2d 1769, 1777 (Fed.Cir.1988) (party waives ob
“Compliance with the best mode requirement, because it depends on the applicant’s state of mind, is a question of fact subject to the clearly erroneous standard of review.” Spectra-Physics, 827 F.2d at 1536, 3 USPQ2d at 1745. On the evidence here, I see no basis for holding that the district court’s determination that Randomex failed to disclose its best mode was clearly erroneous.