The opinion of the court was delivered by: FARNAN
JOSEPH J. FARNAN, UNITED STATES DISTRICT JUDGE
In this action, RCA Corporation ("RCA") charged Data General Corporation ("Data General") with willful and deliberate infringement of U.S. Patent No. 3,345,458 ("the Cole patent"). RCA also accused Data General of having willfully breached a licensing agreement entered into under the Cole patent.
Data General denied infringement and asserted a panoply of legal and equitable defenses related primarily to the issues of validity, noninfringement and unenforceability. The Court has jurisdiction over the patent issues pursuant to 28 U.S.C. § 1338(a), and has pendent jurisdiction over the alleged breach of the licensing agreement.
The Court conducted a bench trial in this action and, following submission of Proposed Findings of Fact and Conclusions of Law by the parties, presided over post-trial argument. In accordance with Federal Rule of Civil Procedure 52(a), this Opinion constitutes the Court's Findings of Fact and Conclusions of Law on the issues of validity, infringement, unenforceability and damages.
In order to understand the nature and character of this lawsuit, it is necessary to recount the prior litigation in this district involving the Cole patent. See In re Cole Patent Litigation (hereinafter " HLA "), 558 F. Supp. 937, 959 (D.Del. 1983) (Cole patent held invalid as anticipated by British Dirks patent; however, but for Cole's invalidity, it would be infringed by the accused terminals), rev'd in part, aff'd in part and remanded, RCA Corp. v. Applied Digital Data Systems, Inc., 730 F.2d 1440, 1448, 221 U.S.P.Q. (BNA) 385 (Fed.Cir. 1984) (judgment of invalidity reversed, judgment that patent not proved invalid for obviousness affirmed). In the HLA action, RCA charged Hazeltine, Lear Siegler and ADDS
with willful and deliberate infringement of the Cole patent. The HLA decision has to some extent become intertwined with the instant litigation, although not inextricably so. The earlier case bears most directly on Data General's assertion that, if not invalid, the Cole patent is unenforceable because RCA's victory in the HLA litigation was achieved through the perpetration of a fraud on the HLA court. See RCA Corporation v. Data General Corporation, No. 84-27O-JJF, slip op. at 2-4 (D.Del. October 27, 1986). Additionally, the HLA litigation is pivotal to RCA's argument that the Federal Circuit's decision in HLA is stare decisis on the issues of Cole patent validity, infringement, and alleged inequitable conduct before the Patent and Trademark Office ("PTO").
In late 1959, RCA engineers Cole, Corson, and Stocker first conceived of the Cole invention. At that time, Cole was working on a display in connection with an RCA proposal to sell a display for a NORAD military command post installation in Wyoming. Cole, a systems engineer experienced in digital electronics and familiar with character generation techniques, contacted Corson, a member of RCA's Advanced Development Display Group, and Stocker, the leader of that same group, in connection with designing a display for the NORAD project. The group sought a fast, accurate, reliable, low maintenance system capable of displaying computer messages on a television-type display. Corson testified that the only digital way then known to the inventors to convert the data output of a computer for displays was a Wang Generator used in the Gordon patent.
As to the actual invention of the Cole patent, neither Stocker nor Corson recalls the precise details of its conception. Stocker, however, did recall one turning point in the inventive process: "[Cole] came up with the amazing fact . . . that alphanumeric [video] signals don't have to have half-tones like the video gradation of brightness in a photograph. Its quite true, and I had known it, but I had never realized this. This opens a whole new field of binary systems in handling alpha-numeric symbology." This realization signalled the inventor's move toward digital technology. According to Stocker, this was the "concept of digital video [which] had come to [them]" that the video beam can be "turn[ed] . . . on or off at definite times." In short, the inventors realized that characters could be formed on the screen by sending appropriately timed signals to the electron beam of the tube, turning it on and off at the correct instant. This insight was combined with the inventors' recognition that the timing and synchronization of the display of a television raster could be analyzed in terms of three variables: the code of the character to be displayed ("character code"); the scan line then being traced across the face of the tube ("scan line count"); and the horizontal position of the electron beam scanning across the face of the television tube ("position count").
While still at the conception stage, the inventors were dispersed. Corson was transferred to the west coast in late 1959, Cole was transferred to a facility in the Boston area shortly thereafter, and Stocker, the senior inventor, remained in Moorestown, New Jersey. With the consent of his fellow inventors, Stocker disclosed the Cole invention to RCA's patent department in March, 1960.
Thereafter, the Cole patent disclosure was the catalyst for feasibility studies conducted by an RCA display group located in Van Nuys, California. The object of the feasibility study conducted in Van Nuys was to "determine if there [was] a feasible system for utilizing digital methods to generate video for a display." In a section captioned "Summary," the authors noted that two methods for generating color video by digital techniques had been developed. In the technique called "computer generation of video (CGV) system, video is generated, one television line at a time, as a series of 'ones' and 'zeros' by a digital computer and accumulated in a buffer storage until the display is completed. Then the buffer storage repeats the video lines in a flicker-free raster scan." In the second system, entitled real time generation of video (RTGV) system, the display information is stored in a data store. The information is read from the data store at such a rate that it may be converted by a high speed character generator to video and presented to the display console as a flicker free display. With the RTGV system, the engineers sought to devise a system that would operate with the normal deflection mechanism of the display and generate randomly placed characters. The RTGV system differed from the system described in the Cole patent disclosure in that the latter was designed to display characters in fixed positions as on this page of text.
Claims 1 through 3 of the Cole patent, the claims in suit, are as follows:
1. A display system for generating character patterns for display on a display device that exhibits a television raster scan-line pattern, each character pattern being displayed in one character space,
means responsive to a certain character code for applying to a certain selected lead an output signal having a duration substantially equal to the scanning time in said scan-line direction through one character space,
means for generating scan-line select counts in synchronism with the scan lines of said raster, each scan-line count having a duration substantially equal to that of a raster scan-line,
means for generating position counts which occur successively during a scan along a scan-line through a character space, and
means for causing said output signal appearing on said selected lead, said scan-line counts and said position counts to supply to said display device a selected character pattern.
2. In a system for displaying a message comprising certain character patterns on a display device that exhibits a television raster scan-line pattern, wherein each different character pattern is manifested by a digitally coded data signal corresponding thereto, the improvement comprising generating means responsive to the data signal forming said message applied thereto for digitally generating a video signal for use in displaying said message on said display device, and means for applying said data signals forming said message to said generating means.
The Cole patent describes a system for decoding digital computer symbol codes representing a message and converting them, using digital techniques, into digital video signals for displaying the message on a television-type cathode ray tube ("CRT"). These video signals are then used to blank and unblank an electron beam to display the message as the beam scans across the screen. The beam scans the screen in a television raster scan pattern. The term "television raster scan" describes a standard deflection pattern of the electron beam in the CRT of a standard television monitor and other television-type CRT's.
As the electron beam is deflected, it sweeps across the screen of the CRT. If the video input to the television monitor turns on the electron beam, the beam will illuminate the phosphor coating inside the face of the CRT screen as it scans the area in which the image is to be displayed. This will cause that part of the coating to glow for a short time. If the beam is turned off, it does not strike the coating, and that area of the screen remains dark. Thus the video input can paint a picture on the face of the CRT if it is properly coordinated with the raster scan deflection signals.
In the Cole system, the beam scans across one horizontal line at a time. The horizontal deflection circuits of the CRT cause the beam to sweep rapidly from left to right, horizontally, across the full CRT screen. The beam then "flies" back to the left side of the screen, moving sequentially down the screen to the bottom. Each successive horizontal sweep thus falls below the preceding one. At the end of the vertical sweep, the beam flies back to the top of the screen and repeats the process.
The digital video control signal controls the points at which the beam illuminates the screen during its scan. The beam can therefore be used to form a message or image on the screen. The speed of the beam's movement as it traverses the screen renders it undetectable to the eye. Image flicker is avoided because the beam "refreshes" each bright spot at a high rate.
The characters which comprise the message are formed on the screen by patterns of dots created when the beam is on. The Cole specifications indicate that each character can be represented by the dots in a rectangular dot matrix having fixed dimensions (e.g., five dots wide by seven scan lines high). A character is displayed on the television screen within a character space that includes the dot matrix of the character and additional blank space to separate the characters on the screen. As illustrated in Figure 4 (see Appendix C) of the Cole patent, each horizontal slice of a character dot matrix corresponds to the dots for that character along one television raster scan line. Thus, to form the character shown in Figure 4, dots 103-105 (for the character "A") and dots 106-109 (for the character "B") will be written as the electron beam moves along scan line 9 in Figure 4. Thus, each character is formed slice by slice, in a fixed character space on the screen consisting of a matrix of dots.
The invention described in the Cole specifications purports to provide several important advances over prior art devices. RCA attributes the advantages of the Cole invention to a "unique combination of features." One such feature is Cole's ability to circumvent the "scan conversion" problems associated with the use of a standard television raster scan CRT without resort to intermediate storage of the video signal output of the character generator. As Judge Stapleton noted in his examination of the Cole patent, most prior art display systems "[wrote] each character in its entirety on a CRT screen before going on to the next character to be displayed." In re Cole Patent Litigation, supra, 558 F. Supp. at 942. Most of these systems used miniature ("mini") raster scan, dot matrix symbol pattern character generators. In systems using the mini-raster scan pattern, the beam scanned vertically from top to bottom left to right, over the position on the CRT where the character was to appear. As in Cole, the character itself was written a slice at a time. However, in a device employing a mini-raster scan, the beam scanned through only one character at a time, completing that character before proceeding on to the next.
Judge Stapleton identified two disadvantages associated with display systems employing the mini-raster scan pattern: (1) the requirement of special circuits to cause the electron beam to deflect in a mini-raster scan pattern; and (2) the necessity for intermediate storage. To use such a system on a standard television monitor, a message written in character-by-character order had to be stored in a memory and read out a scan line at a time in order to avoid the "scan conversion" problem mentioned above and in order to be compatible with the television raster scan pattern. In re Cole Patent Litigation, supra, 558 F. Supp. at 942. The invention disclosed in Cole eliminated the need for special deflection circuitry and for "intermediate storage between the character generator and the CRT." Unlike systems using the mini-raster scan pattern, the character generator described in the Cole system involved no intermediate storage and produced the video signal in a manner compatible with the television raster scan pattern. Thus, the Cole system was the first entirely digital system capable of "real time" or "on the fly" electronic generation of a display of a message composed of alphanumeric characters on an ordinary TV screen.
As it did in HLA, RCA presses here that the purely digital system described in Cole represented a significant improvement over prior art display devices utilizing analogue character generators.
C. Overview of Display Technology.
In his discussion in HLA concerning the Cole patent specifications and the relevant prior art devices, Judge Stapleton noted four technological limitations relevant to a system for receiving digital coded data for display in decoded form on a CRT. RCA Corp. v. Applied Digital Data Systems, supra, 730 F.2d at 1442 (citing In re Cole Patent Litigation, supra, 558 F. Supp. at 943). For the purpose of assessing the relevant prior art and the advances, if any, made by Cole, I set out the technological limitations considered by Judge Stapleton in HLA :
1. The type of scan pattern. As noted above, the two types of primary concern here include one in which the scan covers one character space at a time (a miniature raster scan pattern), and one in which each line of the scan covers a horizontal slice of each character in a row, as the beam scans across the entire width of a CRT screen (television raster scan pattern).
2. The type of CRT. The two principal types are the memory tube, which can hold a picture for minutes, and the non-memory type (including some with high-persistance phosphors), of which a TV tube is an example, that needs to be "refreshed" at a sufficient rate to make the picture appear continuous.
3. The type of character generator. The two broad categories are analog and digital, as related to the type of signal in and signal out, whether continuously variable (analog) or variable only in discrete increments (digital).
4. Storage. A storage or memory is required in a system employing a non-memory CRT because the video signal must be applied to the CRT a number of times a second. The memory may, however, be either one that stores the character code prior to decoding or one that stores the video bits produced by the translation process. When the former is used, the system is sometimes characterized as an on the fly system to indicate that the video bits are applied to the input of the CRT as each one is generated by the translator in contrast to a system that has storage of the video bits.
In re Cole Patent Litigation, supra, 558 F. Supp. at 943.
II. EFFECT OF THE HLA DECISION
Before examining the issues of validity, infringement, and unenforceability, the Court must first confront the question of whether the doctrine of stare decisis binds the Court to the decisions reached by the United States Court of Appeals for the Federal Circuit and by this District Court in prior Cole patent litigation. Throughout this litigation and in its posttrial brief, RCA has argued that it is entitled to the stare decisis effect of the HLA decision. According to RCA, the doctrine of stare decisis requires the Court to hold RCA's Cole patent valid, infringed, and not unenforceable because of alleged inequitable conduct before the PTO. While I am prepared to observe the limitations imposed on the Court by stare decisis, I disagree that respect for and adherence to the doctrine compels the result urged by RCA.
The Federal Circuit in Stevenson v. Sears, Roebuck & Co., 713 F.2d 705, 218 U.S.P.Q. (BNA) 969 (Fed. Cir. 1983) addressed the application of the doctrine of stare decisis to patent cases. The Court stated:
Furthermore, we note that the doctrine of stare decisis is generally an inappropriate one in patent litigation. As stated earlier, patents cannot be held "valid" under all circumstances. Rather, a court merely decides in a particular case that the one attacking validity has not overcome the statutory presumption of validity. . . . To be sure, a prior holding of "validity" should be given weight in a subsequent suit on the issue of "validity." But the prior holding does not necessarily have stare decisis effect.
Id. at 711. The Court added that the weight to be accorded a prior holding of "validity" depends on the additional evidence of prior art or patentability introduced in a second lawsuit. Id.
I find that the record in the instant case differs in significant respects from the HLA record and, therefore, conclude that RCA may not rely on the HLA decision for stare decisis effect. This conclusion is based on the fact that some of RCA's witnesses in the HLA action did not testify in the instant action, and at least one RCA witness gave testimony at the first trial that was inconsistent with his testimony at the second trial. Additionally, Data General has introduced documentary evidence not in the HLA record, as well as the expert testimony of engineers of ordinary skill in the art. Because of these substantial differences in the records of the two cases, I conclude that RCA is not entitled to the stare decisis effect of the HLA decision.
III. VALIDITY OF THE COLE PATENT
Data General contends that RCA's submission of a proposal to the Federal Aviation Administration ("FAA proposal") on October 8, 1962, placed the Cole invention "on sale" more than one year prior to the October 16, 1963, filing date of the Cole patent application. In asserting this on-sale bar claim, Data General "has the burden of proving that there was a definite sale or offer to sell more than one year before the application for the subject patent, and that the subject matter of the sale or offer to sell fully anticipated the claimed invention or would have rendered the claimed invention obvious by its addition to the prior art." UMC Electronics Co. v. United States, 816 F.2d 647, 656, 2 U.S.P.Q.2D (BNA) 1465 (Fed. Cir. 1987), cert. denied, 484 U.S. 1025, 108 S. Ct. 748, 98 L. Ed. 2d 761 (1988). If Data General can make this prima facie showing, then RCA must "come forward with an explanation of the circumstances surrounding what would otherwise appear to be commercialization outside the grace period." Id.
In August, 1962, the FAA sent ten prominent companies a Request for Proposal ("RFP") and an Engineering Requirement ("ER") for character generation equipment. The statement called for the contractor to provide "all necessary qualified personnel, facilities, materials, equipment and services and . . . design, develop, fabricate, test and furnish an alpha-numeric and symbol generation equipment for use with rectilinear raster scan type of radar brights displays." RCA was among the six companies that responded with proposals. On October 8, 1962, RCA submitted a proposal for a system called real time generation of video ("RTGV"). The proposal was composed of two distinct parts: (1) a 68-page technical description with background information and a "detailed delivery schedule and a description of rate of completion;" and (2) financial data and cost analyses. This second part of the proposal was never produced by RCA during discovery in the instant case. The RTGV system was designed primarily by RCA engineer Walter Helbig of RCA's Data Systems Division in Van Nuys, California.
The Court finds that RCA's submission of the proposal to the FAA was a definite offer to sell the RTGV system. The FAA was not seeking to fund research programs; instead, it was looking for a "prototype system" to be followed by "production quantities." Chuck Fanwick, RCA's liaison with the FAA, testified that RCA viewed the FAA proposal as a purely commercial endeavor. According to Fanwick's testimony, RCA viewed the FAA as a "prospective customer" who could provide "a major piece of business" resulting in revenue to RCA. The Court concludes that RCA's proposal to the FAA was an offer to sell solely for the purpose of commercial gain.
The Court must next determine if the subject matter of the FAA proposal fully anticipated the Cole invention or rendered the invention obvious. In order to address this question, it is necessary to first recount RCA's development and refinement of the Cole invention in 1962.
Although RCA conducted feasibility studies on Cole or variations of Cole in Van Nuys, California, RCA authorized no funds to build the invention until Arthur Stocker, one of the inventors, disclosed it to RCA's Canadian subsidiary in 1962. In March of 1962, Stocker met with representatives of RCA's Canadian subsidiary ("RCA Canada") and Trans Canada Airlines ("TCA", now Air Canada). At this meeting, TCA informed Stocker of its desire to replace its manual menu board/television camera display system with a more automated system for displaying flight information. Following this meeting, Robert Clark, an engineer at RCA Canada, began working on replacements for TCA's menu board system. The system that was ultimately built was labelled DIVCON, an acronym for digital-to-video-conversion. The facts surrounding the building of the DIVCON system are the subject of a great deal of controversy.
In preparation for the HLA trial and during the trial, before certain documents emerged in this case which placed the validity of the Cole patent in jeopardy, RCA called Clark to testify regarding the DIVCON system. The relevance of that line of inquiry was counsel's assertion that the "opposing parties are making [certain arguments] concerning the commercial feasibility or lack of it of the Stocker-Cole-Corson invention." At the HLA trial, after having reviewed documents to refresh his recollection of the events of 1962, Clark testified that the following events took place. In April, 1962, Clark conferred with Stocker and began to "try to implement [Cole] as detailed as possible in the next couple of weeks to see if in fact it looked like a practical and economical approach to the problem." In April, 1962, Clark prepared a circuit diagram of the DVG-2 system (i.e., Cole Figure 2). Up until the middle of May, 1962, Clark worked on a proposal to Air Canada which utilized the DGV-2 system. By mid-May, Clark had formulated a block diagram that constituted a "digital implementation of [Stocker's] concept." In June, Clark personally presented the proposal to Air Canada. According to Clark's HLA testimony, TCA was "very pleased" but wanted to see something working to demonstrate that Clark could "indeed, provide a digital video display using semi-conductor logic and have that synchronized to a standard TV source . . . ." Clark testified that by the end of June, 1962, he had no doubt that he could demonstrate the feasibility of the Cole concept. In Clark's own words, "I had no question that I could go ahead and build that system." Toward the end of June, Clark began to build a system using scrap components. Using "whatever [they] could get [their] hands on," he and his coworkers "had a completed breadboard by the end of August." In September, 1962, the flight information display utilizing DVG-2 was demonstrated to Air Canada. Clark testified in the instant case that the system he demonstrated to Air Canada in 1962 met all of the elements of claim 1. This recitation of the events culminating in Cole's reduction to practice is consistent with Clark's deposition testimony given approximately one year before the trial in HLA. The version of events Clark gave in HLA was reaffirmed after the HLA trial at his first deposition in this case in early 1985.
Following RCA's production of the FAA proposal, Clark revised his testimony on the timetable concerning the reduction to practice of DVG-2. RCA produced the FAA proposal in the fall of 1985, after the conclusion of the HLA case and during discovery in this case. The existence of an FAA proposal was an operative fact in the HLA litigation -- indeed, RCA relied on the testimony of Clark and Helbig, the RCA engineer intimately connected with the FAA proposal, to establish commercial success and feasibility. However, the actual document was not produced in the HLA case and did not surface in this case until late in 1985. Only after production of the document itself, which revealed that the proposal had been submitted more than one year prior to the filing of the Cole patent application, did Clark's HLA testimony on the DVG-2's operability and reduction to practice as of fall 1962 emerge as a potential liability.
In this case, Clark's testimony regarding the development and timetable of a working Cole system stands in stark contrast to his prior testimony. Here, Clark testified that, after the proposal to Air Canada in June, 1962, he began to work on the timer and control unit. Clark stressed that the timer and control unit was "a particularly difficult piece of logic to develop . . . [since] we had to synchronize those counters to a sync generator." Clark further testified that it took him until mid-November of 1962 to wire the symbols needed to produce a partial flight display. Clark stated that he was unable to wire the symbols needed to generate the message "the quick brown fox jumped over the lazy dog" until January of 1963. In a bewildering departure from his earlier testimony, Clark testified that he never used the DVG-2 in a full system.
In the ebb and flow of trial, the Court is often called upon to assess a witness' credibility. These determinations are often difficult, but this one is not. Throughout the HLA trial and well into the discovery phase in this case, Clark stood by one critical fact -- that he and his fellow workers at RCA Canada "had a completed breadboard [of the DVG-2] by the end of August 1962," that the device operated "very satisfactorily" and that the device he demonstrated to Air Canada in September 1962 generated a screen of complete flight information. His stunning about face in this trial is simply implausible, especially in light of the recent emergence of the potentially damaging FAA proposal.
RCA attempted at opening argument and attempts in its post-trial brief to explain Clark's testimonial flip-flop. According to RCA, after reviewing documents with RCA lawyers, Clark realized that his HLA testimony had "telescoped" the sequence of events in RCA Canada that occurred in late 1962 and 1963. The evidence that emerged on cross-examination, however, casts this explanation in a somewhat different light. In the fall of 1985, a few months after the production of the FAA proposal, Clark met with RCA counsel. Clark was shown selected documents "and [was] pressed on the subject" of reduction to practice. After that meeting, Clark's former testimony was completely disavowed. The new version, as recounted by Clark, appears not to be a recollection, but rather a reconstruction from the selected documents he was shown. In fact, Clark testified at trial that his actual recollection of events comports with that given in the HLA case and prior to the discovery of the FAA proposal. It lends no credence to Clark's revised testimony that many of those selected documents were reviewed before he testified in the HLA case. Further, none of these selected documents reveals what was actually demonstrated in September of 1962.
Further, other documents undercut Clark's revised testimony. An October 19, 1962 memorandum from inventor Stocker to one O.V. Mitchell concerning the digital generation of video signals states that RCA Victor Ltd. of Montreal had developed a circuit based on these ideas with the thought of selling it to commercial airlines. Additionally, an affidavit from Clark, dated March, 1966, when his memory was undoubtedly more refreshed, refers to a 1962 proposal made to the Department of Transport in Ottowa, which proposal contained a photograph of the complete flight display. Finally, Clark's testimony in this trial contradicts his HLA testimony that RCA attorney Norton incorporated the DVG-2 in figure 2 of the Cole application in August 1963 in order to have "a design that was working." As Data General correctly points out, if the DVG-2 system was never working, as Clark now asserts, Norton by his own account would not have incorporated it. In the face of this evidence there is but one reasonable, plausible inference for the Court to draw: Clark built the DVG-2 system by September, 1962. That is, for the purposes of assessing a possible on sale bar, Clark had built a functional device coming within the scope of the claims by September, 1962.
Since the RCA patent had reduced the Cole invention to practice in September, 1962, the Court must examine the subject matter offered for sale in the FAA Proposal.
At trial, Michael L. Dertouzos, Professor and Director of the M.I.T. Laboratory for Computer Science, an expert offered by Data General, testified that the system used in RCA's proposal to the FAA "uses a scheme . . . [which] drives a Rom which in turn dumps its contents into a shift register, which in turn through some additional circuitry goes to the CRT." Dertouzos further testified that if the Cole patent covers a Rom/shift register approach, such as that found in Data General's accused terminals, then it also covers the FAA proposal. Helbig, the designer of ...