In Re William R. Lueders

RICH, Circuit Judge. This appeal is from the decision of the Patent and Trademark Office (PTO) Board of Patent Appeals and Interferences (Board) affirming the rejection of claims 78-99 of patent application 07/853,356 entitled “Keyboard with Flexible Display and Prompt Capability” (Lueders application). The Board affirmed the rejection of the claims as obvious under 35 U.S.C. § 103 (1994) on grounds other than those relied upon by the Examiner. We reverse. BACKGROUND The Lueders application is a continuation of several prior continuation applications, all having a priority date of August 27, 1986. Claims 78-99 are the only remaining claims in the application. Of these, only Claim 78 and Claim 92 are independent. Claim 78 is representative for purposes of the issues raised in this appeal and is reproduced below. Claim 78: A device comprising: a plurality of data generating elements individually operable and responsive to pressure;* and an electrically programmable display covering said plurality of elements for providing visual indicia associated with said plurality of elements anywhere on said display and operative to flex sufficient [sic] to actuate at least one of said plurality of elements responsive to pressure exerted on said display. Claim 92 differs from Claim 78 in pertinent part by the addition of a processor for applying signals to the display. A preferred embodiment of the Lueders invention is depicted in the following Fig. 2 taken from the Lueders application, and described in the detañed description of the invention. This exploded view shows the major components. Starting from the bottom, there is a printed circuit board (26) 1 on which a plurality of pressure actuated electrical switches (28) are mounted. A static or electromagnetic shield (32) is positioned to completely cover the switches and circuits. A backlighting panel (34) is located on top of the shield, foüowed by a flexible liquid crystal display (36) over the backlight. Lastly, a frame (41) peripherañy clamps the display, backlight, and shield to the printed circuit board. *1571 [[Image here]] The meaning of the claims is explained in the specification and was further elucidated at oral argument. The clause “a plurality of data-generating elements individually ac-tuable in response to pressure” means a pressure responsive keyboard made up of pressure-actuated switches. An electrically programmable display is operatively associated with the keyboard to provide visual indicia associated with the keys. Importantly, the display is sufficiently flexible so that pressure on the display will actuate one of the keys on the keyboard. In,one representative embodiment, the invention is an interactive programmable keyboard comprising a flexible liquid crystal membrane display that fits over a keyboard such that the displays and functions associated with particular keys can change dynamically in response to user input, to prior programming, or even at random. According to the Lueders application’s summary of invention, the display provides the advantage of allowing the user to be interactive with a processing system such that the particular instructions or functional indicia displayed on the screen above each key can change in response to various key depressions and thereby enable a user to enter further and different processing instructions using the same keyboard. The claimed invention is intended to be used for any dynamic input and output. One such use would be as a responsive keyboard where the visible indi-cia corresponding to each key change in response to actuation of the key. Because the keys are claimed generically as pressure-sensitive data-generating elements, however, they could be small dots rather than large keys, in which case the apparatus could be used as a graphical input and output device, like an electronic sketch-pad. The Examiner rejected all claims as obvious over a primary reference, U.S. Patent No. 4,403,965 to William R. Hawkins (“Hawkins”), in view of three secondary references. The Board affirmed the examiner’s rejection, but “for different reasons.” Relying on the general level of skill in the art, rather than on any specific disclosures of secondary references, the Board held that Hawkins sufficiently suggested all elements of the claimed invention to render it obvious. ANALYSIS We review the Board’s ultimate legal conclusion of obviousness de novo. Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561, 1566, 1 USPQ2d 1593, 1596 (Fed.Cir.1987). This legal conclusion, however, rests on several factual findings, including among others, what a prior art reference teaches or suggests — two different inquiries, as well as the secondary indicia of obviousness discussed by the Supreme Court in Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 693-94, 15 L.Ed.2d 545 (1966), see In re Beattie, 974 F.2d 1309, 1311, 24 USPQ2d 1040, 1041 (Fed.Cir.1992); Panduit, 810 F.2d at 1579 n. 42, 1 USPQ2d at 1606 n. 42. We *1572 review these underlying findings of fact for clear error. See, e.g., In re Baxter Travenol Labs., 952 F.2d 388, 390, 21 USPQ2d 1281, 1283 (Fed.Cir.1991). The primary reference in this case, Hawkins, teaches an electronic teaching apparatus designed for use by children, which includes a speech synthesizer that audibly reports the subject of a pictorial representation when it is pressed by the user. For example, when a user presses the picture of a tree, the machine utters the word “tree.” An embodiment of the Hawkins apparatus is shown in the following Figures 2a and 2b, taken from Hawkins. [[Image here]] Hawkins teaches a pressure-responsive keyboard (3) 2 that is programmable through actuation of coding switches (7) on the side of the keyboard. This keyboard is used in con *1573 junction with a plurality of coded flexible overlays (2), each displaying a series of pictorial representations and having a set of four coding apertures (6). The coding apertures mechanically engage the coding switches (7), shown in Fig. 2b, to selectively determine which sounds will be produced in response to activation of the keys underlying the particular pictorial representations on the particular flexible overlay then in use. The Board made the following findings concerning Hawkins: The Hawkins teaching apparatus essentially requires only two features with respect to its inputs and visual outputs. The visual outputs must be variable to give the device sufficient flexibility and practical application. The inputs must be actuable by the user by touching or depressing a specific portion of the display overlay. ' According to the Board, this is a very broad teaching of inputs and outputs that would be combined by one of ordinary skill in the art to form the Lueders invention. The Board found the teaching to be augmented by the suggestion at column 4, lines 43-46 of Hawkins that “other display/input means such as liquid crystal display and a touch-capacitive keyboard system may be utilized.” According to the Board, Hawkins suggests the use of a liquid crystal display in combination with a keyboard. The Board found that “a person having ordinary skill in the art would have recognized that a pressure sensitive keyboard could be used with a liquid crystal display instead of the exemplary touch-capacitive keyboard, because both keyboards were known to be conventionally used in an interchangeable manner as a matter of routine design consideration.” From these findings, the Board reached the ultimate conclusion that the Lueders invention was obvious over Hawkins, which, it said, discloses all elements of the Lueders invention — a liquid crystal output over a pressure sensitive input, in view of the general level of skill in the art. On appeal, Lueders argues that the. invention is more than a simple conglomeration of prior art inputs and outputs. Lueders asserts the invention to be a certain spatial and operative combination of inputs and outputs such that there is dynamic interaction between the two. In his briefs and at oral argument, Lueders persuasively argues several reasons why specific factual findings by the Board are clearly erroneous. In light of the particular record and decision before us, we agree with Lueders that the Board clearly erred on each of these findings. First, Lueders argues that Hawkins does not teach the use of a visual output at all. Output is “data produced by a data-proeess-ing operation.” McGraw-Hill Dictionary of Scientific and Technical Terms 1141 (3d ed.1984). In Hawkins, the visual images on the flexible display are fixed, generated in advance of use, and not in response to operation of the device. Whereas the images of Lueders are output because they are generated in response to user input, the only output in Hawkins is audible output from the voice synthesizer, not visual output from the display. Hawkins’ output consists of synthesized voice sounds. Pictures in, words out. Indeed, while Hawkins may suggest using liquid crystal displays, which are potentially programmable, Hawkins itself teaches only static displays. At column 5, lines 9-11, Hawkins states that “other similar coding means may by utilized to allow differentiation of numerous overlays.” (Emphasis added). This implies that the displays are static and are altered only by replacement of the entire display. For example, in Figure 2a, the pictorial display is changed by replacing plastic sheets (2). Because there is nothing in this record to teach the use of 'a visual output, we agree with Lueders that the Board clearly erred in concluding that Hawkins teaches the use of a visual output. 3 Second, Lueders argues that Hawkins does not teach the dynamic interactive connection between the keyboard and display of the invention. The PTO Solicitor argues that the suggestion in Hawkins to use a liquid crystal display anticipates this element because it was well known in the art that such displays *1574 are programmable. Lueders does not dispute the known programmability of liquid crystal displays, but rather points to the aforementioned express statement at column 5, lines 9-11 of Hawkins that the display and the keyboard interact via the coding method in each fixed display. Lueders argues that this statement contemplates a set of passive, static displays rather than a single, actively changing display. Conceding that the plastic overlay of Hawkins could be replaced by a liquid crystal display, Lueders argues that Hawkins teaches away from Lueders because all displays in Hawkins are static and only change by manual replacement of the entire display. We are persuaded by Lueders’ argument that all displays in Hawkins, programmable or not, are nevertheless not dynamically responsive to keyboard input. 4 As before, we find no evidence offered by the Board or the PTO Solicitor to refute this factual assertion by Lueders. As a result, we are persuaded by Lueders that the Board clearly erred on this point. Third, Lueders disputes the Board’s finding that one of skill in the art would know to use a pressure sensitive keyboard with a liquid crystal display. 5 According to Lued-ers, the liquid crystal displays of the art were rigid, not flexible, and therefore would prevent actuation of the keys if placed over a pressure sensitive keyboard. Lueders points to examples in the art, like Figure 9 of U.S. Patent 4,885,580, to Noto et ah, and argued that at the time of invention a liquid crystal display would only have been used beneath a keyboard that is touch sensitive, not pressure sensitive. At oral argument, we repeatedly asked the Solicitor what evidence there was, if any, that formed the basis for the Board’s finding that a person of skill in the art would know to use a flexible liquid crystal display over a pressure sensitive-keyboard. Each time, the. Solicitor answered that there was no such express mention of evidence in the Board’s opinion, but argued that because we shoúld review the decision of the Board, not its opinion, we can disregard the lack of express mention of evidentiary support in the opinion as harmless error and find implicit evidence through inductive reasoning. We are unable, however, to find any such implicit support. Indeed, the only implicit support we do find in the record is itself clearly erroneous. It appears from the Board’s reasoning that it misinterpreted the above phrase from column 4, lines 43^16 of Hawkins concerning “other display/input means”. The Board must have read this phrase as if it were “other display and/or input means”. While Hawkins does suggest using a touch capacitive keyboard and a liquid crystal display, it does not suggest using both a pressure sensitive keyboard and a liquid crystal display. Rather, we are persuaded by Lued-ers’ arguments which are based on examples from the art; absent any contrary evidence cited by the Board in its opinion, wé reverse the Board on this point. Because we agree with Lueders that the Board clearly erred in these three factual findings based on the record before us, we need not address the Board’s reasoning based on those findings by which it reached its ultimate conclusion of obviousness. Instead, we reverse because the Board’s conclusion is not supported by the evidence of •record. Having concluded that the Board’s decision must be reversed because of clear error in its findings of fact, we further note that only the court sitting in banc can determine whether the Board’s decision should be reviewed under a different standard of review. The Solicitor recognizes that we are bound by our precedents to apply the “clearly erroneous” standard in reviewing the Board’s findings of fact, see, e.g., In re Baxter Travenol Labs., 952 F.2d at 390, 21 USPQ2d at 1283, but *1575 argues, however, that, sitting in banc, we should overrule our precedents and apply the more deferential standards of review provided by the Administrative Procedure Act, 5 U.S.C. §§ 701-706 (1994) (APA). More particularly, the Solicitor argues that either the APA’s “substantial evidence” or “arbitrary and capricious” standard should apply. We, as a panel, decline, however, to accept the Solicitor’s invitation to refer this issue to the whole court for in banc review. The APA’s more deferential standards of review have never been applicable to the Board’s decisions and we are not persuaded to change this practice. Throughout the history of this court, of this court’s predecessor court, the Court of Customs and Patent Appeals (CCPA), and of that court’s predecessor, the Court of Appeals of the District of Columbia, 6 the decisions of the PTO boards 7 have been reviewed on the record, by the same standards as applied to a decision from a district court, without enhanced deference. 8 At several times in this long history, Congress has spoken on topics directly relating to the interaction among the PTO, the courts, and the law of patents, but each time Congress avoided changing the then pertinent appellate court’s standards of review. In 1929, Congress created the CCPA. Pub.L. No. 914, 70th Cong. (Mar. 2, 1929). The statute was simple: it changed the name of the Court of Customs Appeals to the Court of Customs and Patent Appeals, and then provided The jurisdiction now vested in the Court of Appeals of the District of Columbia in respect of appeals from the Patent Office in patent and trade-mark cases is vested in the United States Court of Customs and Patent Appeals. Pub.L. No. 914, § 2(a). No change was made to the standard of review. For the next seventeen years the CCPA continued to review the PTO’s decisions without enhanced deference. Then, in 1946, the APA was enacted to provide a uniform procedure to remedy most errant administrative actions. While the intended scope of the APA was extensive, as expressly noted by its drafters at the timé, the APA was not designed to alter the review process of the CCPA. As explained by then Attorney General Clark, while discussing the legislative history of what became the APA’s provision on reviewable actions, 5 U.S.C. § 704: Furthermore, this provision does not apply to situations where the Congress has provided special and adequate review proce-dures____ Thus, the Customs Court and the Court of Customs and Patent Appeals retain their present exclusive jurisdictions. 15 Attorney General’s Manual on the Administrative Procedure Act 101 (1947). Importantly, the Attorney General’s Manual has frequently been given deference by the Supreme Court: “[bjeeause of the role played by the Department of Justice in drafting the [APA],” Vermont Yankee Nuclear Power Corp. v. Nat’l Resources Defense Council, Inc., 435 U.S. 519, 546 [98 S.Ct. 1197, 1213, 55 *1576 L.Ed.2d 460] (1978) (Rehnquist, J.), [and] because Justice Clark was Attorney General both when the APA was passed and when the Manual was published. Steadman v. SEC, 450 U.S. 91, 102, n. 22, 101 S.Ct. 999, 1008 n. 22, 67 L.Ed.2d 69 (1981) (also giving deference to the letters and statements of then Attorney General Clark). Indeed, as expressly noted during the flurry of activity preceding the APA’s enactment, Staff reports were completed in connection with all the agencies [herein listed] except the following which are affected by special circumstances: ... Patent Office, Department of Commerce (the highly specialized character of the Patent Office and. the insufficiency of the Committee’s staff led first to postponement and then to abandonment of plans to study this agency). Report Of The Committee On Administrative Procedure, Appointed By The Attorney General At The Request Of The President, To Investigate The Need For Procedural Reform In Various Administrative Tribunals And To Suggest Improvements, S. Doc. No. 8, 77th Cong. 1st Sess. at 4, note 2 (1941). 9 After the 1946 enactment of the APA, several more years passed with the CCPA continuing to review the PTO’s decisions without enhanced deference 10 before Congress once again set its hand to work on the patent law — this time in the 1952 Act. Pub.L. 593, 82nd Cong., 2nd Sess., Chap. 950, 66 Stat. 792, approved July 9, 1952, now 35 U.S.C. § 1 et seq. (The ’52 Act). The effort was extensive; all of Title 35 (Patents) was codified and many changes were made. In addition, all prior statutes were repealed. The drafting of the ’52 Act was surely an opportune time to reconcile, if necessary, the CCPA’s review practices with the APA. Both matters were plainly before Congress. The CCPA’s review practices had existed since well before the creation of the CCPA, and Congress had recently spent several years studying, drafting, and enacting the APA. Moreover, the two seemingly incompatible regimes had co-existed since 1946. Yet, there was never even a suggestion that the CCPA’s standard of review should be changed, even in light of the recently enacted APA. After the ’52 Act, another ten years passed with the CCPA continuing to review the PTO’s decisions without enhanced deference before the next round of Congressional actions in the field of patents — this time expressly incorporating a citation to the APA into Title 35. In 1962 Congress added subsections (b) and'(c) to 35 U.S.C. § 135 (Interferences). Pub.L. 831, 76 Stat. 958, 87th Cong. The last paragraph of Section (c) provides Any discretionary action of the Commissioner under this subsection shall be reviewable under section 10 of the Administrative Procedure Act 35 U.S.C. § 135(c) ¶ 3 (1962). 11 Thus, in 1962, Congress once again had the opportunity to reconcile the CCPA’s review practices with the APA, but instead chose to incorporate a mandatory citation to the APA into Title 35 that is so expressly, limited that it only applies to subsection (c) of Section 135. *1577 Indeed, the express application of the APA to this limited subsection of Section 135 strongly supports the conclusion that the APA should not apply elsewhere in Section 135 and surely not to other sections in Title 35, like Section 141 (Appeal to Court of Appeals for the Federal Circuit). 12 Yet another twenty years passed with the CCPA continuing to review the PTO’s decisions without enhanced deference before Congress once again set its hand to work on the patent law — this time in the Federal Courts Improvement Act, Pub.L. 97-164, 96 Stat. 25 (1982). Like the ’52 Act, the Federal Courts Improvement Act was a significant venture. In addition to merging the CCPA with the Court of Claims, the Federal Courts Improvement Act consolidated in this court, the Court of Appeals for the Federal Circuit (CAFC), nationwide jurisdiction over all appeals from patent cases in the district courts in addition to the CCPA’s existing jurisdiction over direct appeals from the PTO boards. Rather than reconciling some alleged conflict between the APA and the practice of appellate courts reviewing the PTO’s boards without enhanced deference, the Federal Courts Improvement Act ratified, albeit through inaction, a practice that extends back beyond the beginning of this century. Indeed, by so doing, the Federal Courts Improvement Act avoided a very real problem that would be presented if we were to apply the APA’s more deferential standards to appeals from the BPAI while applying the clearly erroneous standard to appeals from the district courts. 13 By enacting the Federal Courts Improvement Act, Congress made it altogether likely, if not intended, that we would find ourselves sitting in review,of two or more rulings by two or more bodies on, for example, the invalidity of the same patent for obviousness over the same prior art. One such body could be the Board, after a reexamination initiated as a result of a district court action, 14 while one could be a different district court. If we were to adopt the standard urged by the PTO, then we would be forced to give heightened deference to the factual findings of the Board in such a case while giving lower deference to the findings by the district court. We might thereby be compelled to hold the same patent both invalid and not invalid over the same prior art simply because of the differing standards of review. 15 Surely Congress’s acceptance of the established practice intentionally avoided this result by entrusting us with review on a uniform basis of patent appeals, wherever they are from, in recognition of this court’s “judgment in this area of its special expertise.” Warner-Jenkinson Co. v. Hilton Davis Chem. Co. — U.S. -, -, 117 S.Ct. 1040, 1054, 137 L.Ed.2d 146 (1997). Finally, stare decisis argues against a change in our standard of review. A longstanding practice should not be set aside absent substantial reason to do so. The sum of our historical and precedential inquiry shows our standard of review’s underpinnings to be unweakened in any way. Generations of inventors and investors have acted on the assumption that the standard will continue, no change in statute or facts having developed in the past few years to merit the *1578 PTO’s recent assertions that we should overrule our precedents. No such reason justifies a change and it is within the bounds of stare decisis analysis, then, to continue to apply our established standard of review. See Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833, 861, 112 S.Ct. 2791, 2812, 120 L.Ed.2d 674 (1992). CONCLUSION For the foregoing reasons, we conclude that the Board clearly erred in factual findings that underlie its conclusion of obviousness, and reverse the decision of the Board that the claimed invention would have been obvious. REVERSED. MAYER, Circuit Judge, concurs in the judgment. Footnotes [1] The numbers in parenthesis correspond to the reference numerals in Fig. 2 of the Lueders application. [2] The numbers in parenthesis correspond to the reference numerals in Figs. 2a and 2b of Hawkins, [3] Similarly, there is nothing in the record to suggest (rather than teach) the use of a visual output other than the Board’s naked assertion that such a suggestion is made. Absent reasoned analysis supported by citations to particular art, we are not persuaded. [4] While we note that the potential for dynamic interaction between the display and keyboard is not expressly set forth in Lueders’ claims, it is inherent in them. Moreover, it is clearly part of the claims when read in view of the specification. This is a conclusion that was urged by Lueders at oral argument and that we accept. Perhaps such a limitation might be better suited for express inclusion in issued claims. [5] That is, Lueders disputes the Board’s implicit findings that flexible liquid crystal displays existed in that art and that they were sufficiently flexible to be used over a pressure sensitive keyboard — two distinct questions of fact. [15] In the Attorney General’s memorandum to the Senate Committee, he stated that " 'courts’ includes the Tax Court, Court of Customs and Patent Appeals, the Court of Claims and similar courts. This act does not apply to their procedure nor affect the requirement of resort thereto.” Sen. Rep. p. 38 (Sen.Doc. p. 224). [6] Before 1929, appeals from the PTO went to the Court of Appeals of the District of Columbia and customs cases were appealed to the Court of Customs Appeals, predecessor to the CCPA. [7] The Board of Patent Appeals and Interferences (BPAI) and the Trademark Trial and Appeal Board (TTAB). Formerly called the Patent Office, the PTO’s name was changed in 1975 to "Patent and Trademark Office." Pub.L. 93-596, 93rd Cong., 2nd Sess. (1975). For a brief history of the name, see S.Rep. No. 93-1399, (1974), reprinted in 1974 U.S.C.C.A.N. 7113. [8] See Fed.R.Civ.P. 52(a) (“Findings of fact, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous.”); see also Morgan v. Daniels, 153 U.S. 120, 129, 14 S.Ct. 772, 775, 38 L.Ed. 657 (1894) (“It is enough to say that the testimony as a whole is not of a character or sufficient to produce clear conviction that the Patent Office made a mistake.”). [9] The PTO erroneously asserted the contrary in its brief: "In fact, the [PTO] was one of the agencies specifically studied when drafting the APA." In support of this statement, the PTO cited the 1946 House Report, H.R.Rep. No.1980, 79th Cong., 2d Sess. (1946), reprinted in 1946 U.S.S.C.A.N. 1195, 1199, which summarizes the Committee Report quoted in the main text, above. Importantly, the House Report’s summary is misleading or inaccurate. The House Report contains a list of agencies studied by the Committee, and the Commerce Department is on this list. Following the Department of Commerce there is a parenthetical list of agencies, within the Department of Commerce. This list includes the PTO. The House Report summary cites pages 3-4 of the Committee Report, the same portion that we quote in the text above. But as we point out in the text above, the Committee Report expressly did not consider the PTO component of the Department of Commerce. [10] See Ranney v. Bridges, 38 C.C.P.A. 1044, 188 F.2d 588, 596, 89 USPQ 419, 427 (1951) (Garrett, C.J.) ("It will be understood that in considering allegations of error as to findings of fact made by the tribunals of the Patent Office this court follows the proper and well established practice of appellate courts in refusing to reverse such findings unless we are convinced by our own study of the record that the findings are manifestly wrong against the weight of the evidence.”), [11] The text of this paragraph, 35 U.S.C. § 135(c) ¶ 3, remains the same today. [12] The doctrine inclusio unis est exciusio alterius instructs that where law expressly describes a particular situation to which it shall apply, an irrefutable inference must be drawn that what was omitted or excluded was intended to be omitted or excluded. [13] See note 8, supra. [14] Congress clearly knew that appeals were available from reexaminations since it had created the entire reexamination system only two years earlier by adding a new chapter (Chapter 30), including seven new sections (§§ 301-307), to Title 35 (Patents), Pub.L. 96-517, 94 Stat. 3016, 96th Cong. 2nd Sess. (1980), and had done so, as explained in the House Report, because the new "Section 306 grants the patent owner the right to pursue the same appeal routes available to patent applicants ... [including appeal] to the Board ... [followed by appeal] to the U.S. Court of Customs and Patent Appeals or de novo review ... in the United States District Court for the District of Columbia.” H.R.Rep. No. 96-1307(1), 96th Cong. 2nd Sess. (1980), reprinted in 1980 U.S.C.C.A.N. 6460. [15] Another anomaly is presented if a party were to bring suit in the United States District Court for the District of Columbia under 35 U.S.C. § 145 (Civil Action to Obtain Patent) or 35 U.S.C. § 146 (Civil Action in Case of Interference), in which case the district court would conduct a trial de novo and we would review for clear error.