dissenting-in-part.
I respectfully dissent from section IV of the majority’s opinion reversing the district court’s denial of Rambus’s motion for judgment as a matter of law on the issue of fraud. In my opinion, substantial evidence supports the jury’s verdict that Rambus committed actual fraud under Virginia state law.
“The species of fraud are numberless, and like a chameleon, fraud is always colored by the context from which it arises. For that reason, it is usually for the jury to determine from the facts of a specific case, whether a fraud was committed.” Hirschberg v. G.W. Motors, Inc., 34 Va. Cir. 55, 60, 1994 WL 1081181 (1994).
Fraud is seldom, if ever, provable by direct testimony, but usually must be shown by circumstances which are sufficient to convince fair-minded men that they would not have occurred without the existence of a fraudulent purpose and design. Fraud is a mixed question of law and fact but, in most cases, is a jury question.
French v. Beville, 191 Va. 842, 62 S.E.2d 883, 889 (1951); Hirschberg, 34 Va. Cir. at 60. In this case, the jury heard direct and circumstantial evidence supporting the conclusion that Rambus committed fraud in the context of its membership in the JEDEC standard setting organization.
Rambus attended its first JEDEC meeting in December 1991 and became a member in February 1992. At the time Ram-bus joined JEDEC, it had several pending patent applications derived from the '898 patent application, which has spawned more than a thousand claims in dozens of continuation and divisional applications. Rambus also had a specific plan for using its pending patent applications against anyone using the SDRAM standard. According to Rambus’s June 18, 1992, business plan:
[W]e believe that Sync DRAMs infringe on some claims in our filed patents; and that there are additional claims we can file for our patents that cover features of Sync DRAMs. Then we will be in position to request patent licensing (fees and royalties) from any manufacturer of Sync DRAMs. Our action plan is to determine the exact claims and file the additional claims by the end of Q3/92. Then to advise Sync DRAM manufacturers in Q4/92.
Rambus did not, in fact, inform anyone at JEDEC about its pending patent applications by the end of 1992. Instead, Rambus continued to attend JEDEC meetings for three more years, watching the SDRAM standard evolve and then amending its patent applications to try to cover features of the standard. Richard Crisp, Rambus’s JEDEC representative, testified at trial about how “Rambus was intentionally drafting claims to intentionally cover the JEDEC SDRAMs”:
[Y]ou’ll- agree as an initial matter, right, that over the years '92, '93, '94 and '95 while you were attending meetings, JEDEC meetings for Rambus, at least during a portion of that time you *1108were also working with the Rambus patent lawyers to change the claims in these applications? Right?
A. Yes.
Q. And you’ll agree, won’t you, sir, that at least on some occasions you went to a JEDEC meeting and then met with the Rambus patent lawyer? Right?
A. Yes. That’s right.
Q. And you’ll agree, won’t you, that in the meetings you had with Rambus patent lawyers after a JEDEC meeting, that one source of the information for changing the Rambus patent claims was what you had seen at JEDEC with respect to the SDRAM standardization? Right?
A. Yes. That’s right.
Q. And what you did in those meetings was work on new claims for the Rambus pending patent applications, and your intent was to make them broad enough that they would cover an SDRAM using the features that you had seen at the prior meetings. Isn’t that a fact?
A. In some cases that was true.
The record is replete with additional and specific instances of Rambus employees attending JEDEC meetings, taking notes of what was discussed, identifying instances where Rambus already had claims covering what was discussed, and then seeking claims to cover what they learned at the JEDEC meetings. Yet Rambus “did not tell the people at JEDEC that what they were proposing for standardization infringed [its] patents.” Instead, after considering whether to “walk into the next JEDEC meeting and simply provide a list of patent numbers which have issued,” Rambus concluded that it was better to remain silent because “we may not want to make it easy for all to figure out what we have, especially if nothing looks really strong.” Rambus was even advised by its patent attorneys “to stop attending JE-DEC” and that “if you go to the JEDEC meetings and stay silent and don’t do anything else, you still have a risk that your patents will be unenforceable if you let the standard go forward and you don’t tell them you have patents.” Rambus was explicitly warned in 1992 that “you cannot mislead JEDEC into thinking that Ram-bus will not enforce its patent.”
In 1995, members of JEDEC suspected that Rambus may have intellectual property rights related to the SDRAM standard. Richard Crisp “was asked [at a JEDEC meeting] to make a comment about the Rambus intellectual property position as it may relate to [a particular] proposal.” Rambus responded in writing on September 11, 1995, that “[a]t this time, Rambus elects to not make a specific comment on our intellectual property position.” Ram-bus attended its last JEDEC meeting in December 1995, and on June 17, 1996, Rambus formally withdrew from JEDEC. In its farewell letter to JEDEC, Mr. Crisp stated: “Recently at JEDEC meetings the subject of Rambus patents has been raised. Rambus plans to continue to license its proprietary technology on terms that are consistent with the business plan of Rambus, and those terms may not be consistent with the terms set by standards bodies, including JEDEC.” Even after withdrawing from JEDEC, Rambus continued to furtively pursue its scheme to patent the evolving SDRAM standard by receiving reports from undisclosed attendees at JEDEC meetings named “Deep Throat” and “Secret Squirrel.”
Rambus also tried to destroy the evidence of its plan to draft patent claims to cover the SDRAM standard. Rambus implemented a “document retention policy” in 1998 in part “for the purpose of getting *1109rid of documents that might be harmful in litigation.” It also attempted to prevent discovery of relevant documents by failing to list them on its privilege log in this case. Having believed that they had destroyed or disguised the documents evidencing their plan to patent the SDRAM standard, Rambus’s witnesses initially provided “false or misleading testimony.” Rambus, Inc. v. Infineon Techs. AG, 155 F.Supp.2d 668, 681 (E.D.Va.2001). The false testimony was exposed after the court pierced the attorney-client privilege, compelling Ram-bus to produce previously concealed documents. Id. Once the Rambus witnesses were “confronted with documents obtained after the piercing of the attorney-client privilege” and “prodded by reference to the belatedly obtained documents,” they were compelled to admit that they had in fact participated in the prosecution of Rambus’s patent applications based on information learned at JEDEC meetings. Id.
The jury found that Rambus’s thus exposed scheme amounted to fraud under Virginia state law. In Virginia, “[t]he elements of actual fraud are: (1) a false representation, (2) of a material fact, (3) made intentionally and knowingly, (4) with intent to mislead, (5) reliance by the party misled, and (6) resulting damage to the party misled.” Spence v. Griffin, 236 Va. 21, 372 S.E.2d 595, 598 (Va.1988). Fraud may arise from “deliberate concealment or a relationship, contractual or otherwise, that would give rise to a duty to disclose.” Devansky v. Dryvit Sys., Inc., 52 Va. Cir. 359, 361 (2000). “For purposes of an action for fraud, concealment, whether accomplished by word or conduct, may be the equivalent of a false representation, because concealment always involves deliberate nondisclosure designed' to prevent another from learning the truth.” Spence, 372 S.E.2d at 599; see also Norris v. Mitchell, 255 Va. 235, 495 S.E.2d 809, 812 (1998). When fraud is based on a violation of a duty to disclose, the scope of that duty “depends upon the circumstances of each case and the relationship between the parties.” Hirschberg, 34 Va. Cir. at 57. The duty can arise in many ways:
The principle is basic in the law of fraud, as it relates to nondisclosure, that a charge of fraud is maintainable where a party who knows material facts is under a duty, under the circumstances, to speak and disclose his information, but remains silent. Situations evoking the duty of disclosure may arise in various ways in different cases. Generally speaking, however, in the conduct of various transactions between persons involving business dealings, commercial negotiations, or other relationships relating to property, contracts, and miscellaneous rights; there are times and occasions when the law imposes upon a party a duty to speak rather than remain silent in respect of certain facts within his knowledge, and thus to disclose information, in order that the party with whom he is dealing may be placed on an equal footing with him. In such a case a failure to speak amounts to a suppression of a fact which should have been disclosed, and is a fraud. In such circumstances, a failure to state a fact is actually equivalent to fraudulent concealment and amounts to fraud just as much as an affirmative falsehood.
Among other ways, the obligation to communicate facts may arise from the fact that one of the parties has superior knowledge or means of knowledge; from the fact that confidential relations exist between them; from the fact that a party does something or says something which, for want of the disclosure, is false and deceptive; from the fact that he is placed or places himself in a position where his silence will convey a false *1110impression; or from the fact that a statement or representation has been made in the bona fide belief that it is true, and before it is acted on, the party who has made it discovers that it is untrue.
Id. at 57-58 (quoting 37 Am.Jur.2d, Fraud and Deceit, § 146).
Applying this Virginia state law, we must determine whether Rambus has shown on appeal that it is entitled to judgment as a matter of law that it did not commit actual fraud as found by the jury. Judgment as a matter of law is appropriate when “a party has been fully heard on an issue and there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on that issue.” Fed. R.Civ.P. 50(a)(1). As the appellant, Ram-bus “must show that the'jury’s findings, presumed or express, are not supported by substantial evidence or, if they were, that the legal conclusion(s) implied from the jury’s verdict cannot in law be supported by those findings.” Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 893, 221 USPQ 669, 673 (Fed.Cir.1984). Because the jury returned a general verdict on the ultimate legal question of whether Rambus committed fraud, “the law presumes the existence of findings necessary to support the verdict the jury reached.” Id. at 893, 221 USPQ at 673. There are two issues underlying whether Rambus is entitled to judgment as a matter of law: the scope of Rambus’s duty to disclose and whether Rambus violated this duty.
I
According to the majority, “a reasonable jury could only find that the duty to disclose a patent or application arises when a license under its claims reasonably might be required to practice the standard.” The majority then proceeds to apply this standard by determining de novo whether Rambus had any pending or issued claims while it was a member of JEDEC that read on the final JEDEC standard.
I believe that the evidence in this case supports a broader duty than the one applied by the majority. According to the October 1993 JEDEC Manual of Organization and Procedure, section 9.3.1. titled “Committee Responsibility Concerning Intellectual Property”:
The Chairperson of any JEDEC committee, subcommittee, or working group must call to the attention of all those present the requirements contained in the EIA Legal Guides, and call attention to the obligation of all participants to inform the meeting of any knowledge they may have of any patents, or pending patents, that might be involved in the work they are undertaking (emphasis added).1
In my opinion, this portion of the manual clearly states the duty of disclosure required by all members of JEDEC, which is different from the duty applied by the majority in at least two respects. First, the statement “might be involved in” the standard is much broader than requiring disclosure of only claims reading on the standard. Second, the majority applies the duty to the final standard adopted by *1111JEDEC, whereas the manual requires disclosure based on the “work they are undertaking,” which is much more expansive than the final, completed standard resulting from the work undertaken. The majority’s comparison of pending claims to the final standard does not take into account the possibility that, during the course of its work, the committee considers, debates, rejects and amends various proposals as the standard evolves.
Documents and witness testimony show that the members of JEDEC understood-the JEDEC policy to require that its members disclose patents and pending patent applications that might be involved in the standard setting process. For example, during the development of the SDRAM standard, the committee discussed and then voted on many different features. The ballots for these votes stated that “[i]f anyone receiving this ballot is aware of patents involving this ballot, please alert the Committee accordingly during your voting response” (emphasis added). One witness interpreted the language on the voting ballot as requiring disclosure of “intellectual property that is related to that ballot or to the content of that ballot” (emphasis added). Similarly, the committee’s stated “patent tracking” procedure included “reviewing] items identified as of potential patent interest at each meeting” (emphasis added). In addition, the minutes from the December 6, 1995, committee meeting state that “MOSAID noted that they had a pending patent on DLL and noted that it was a particular implementation and may not be required to use the standard.” Gordon Kelley, the committee chairman and IBM’s JEDEC representative, testified about Rambus’s particular kind of conduct and whether it violated JEDEC’s policies:
Q. As the chairman at least during some period of time of some of the relevant JEDEC committees that we’ve talked about here today, did you think it was — or did you have any understanding as to whether it was acceptable practice for a JEDEC member to attend JEDEC meetings and then write claims to cover proposals in the JEDEC standards without disclosing those — those patent applications or patents that contain those claims?
[objection]
A. So, for the part of your question with regard to write claims, are you suggesting that someone attended the JEDEC meetings and then subsequent to the JEDEC meeting went and wrote claims that he or she learned at the JEDEC meeting? Is that what you mean?
Q. Exactly, or modified existing patents to cover what was proposed at the JEDEC meetings?
[objection]
Q. And my question is did you have — did you have any understanding at the time as a JEDEC participant and also as the chairman whether that was acceptable behavior or not?
[objection]
A. This letter in January of 1994 to Buf Slay I think documents my position on that, that that cannot be allowed. It’s in complete violation of JEDEC requirements of openness and fairness with regard to notification of patents and pending patents.
In addition, John Kelly, EIA’s general counsel and the person responsible for implementing the EIA/JEDEC patent policy, testified that the JEDEC patent policy “required the early disclosure of patents and patent applications that are or may be required to comply with the standard” (emphasis added). Willibald Meyer, Infi-neon’s JEDEC representative, likewise testified to a disclosure duty that was not *1112limited to claims that read on the standard:
The question is, sir, what was your understanding of the JEDEC patent policy in July, June and July 1992?
THE WITNESS: The understanding was that the holders of a patent or an application should make the committee aware in the case that they were aware of that, the application of the patent which they held or had filed was in relationship to the work in JEDEC that we were doing (emphasis added).
Evidence also shows that even Rambus understood that it was required to disclose something more than only those claims reading on the SDRAM standard. Ram-bus timely disclosed only one of its patents to JEDEC: the '703 patent. However, Rambus admitted that the '703 patent “did not relate to JEDEC’s SDRAM work but [was] directed to the implementation of Rambus’[s] RDRAM products.” Rambus’s compulsion to disclose this one patent is evidence that it broadly interpreted its duty of disclosure (although at the time, Rambus allegedly thought its duty was limited to issued patents, not pending applications).2
Certainly the majority opinion has identified testimony that can be interpreted to support its framing of the duty to disclose. However, the majority has applied the duty as being limited to the issue of whether claims read on the final standard, which is not consistent with the broader duty stated in the JEDEC manual or the other evidence identified above. Having identified substantial evidence supporting a sufficiently broad duty of disclosure to support the jury’s verdict, our job is done. The applicable standard of review does not permit us to go further, reweighing the evidence and determining de novo that the duty should be defined or applied in a different manner. See Comark Communications, Inc. v. Harris Corp., 156 F.3d 1182, 1192-93, 48 USPQ2d 1001, 1010 (Fed.Cir.1998); Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1375, 231 USPQ 81, 87 (Fed.Cir.1986). I respectfully submit that the evidence described above compels us to conclude that there was sufficient evidence for the jury to find that Rambus had a duty to disclose pending and issued patents that might be involved in the development of the SDRAM standard, as stated in section 9.3.1 of the JEDEC manual.
The majority rejects the plain meaning of this section of the JEDEC manual for two reasons. First, the majority interprets Appendix E, not section 9.3.1, as giving rise to the duty to disclose. Second, according to the majority, a plain reading of section 9.3.1 of the manual would “render the JEDEC disclosure duty unbounded. Under such an amorphous duty, any patent or application having a vague relationship to the standard would have to be disclosed.” I disagree with each of these reasons for not following the duty of disclosure stated in section 9.3.1 of the JE-DEC manual.
With respect to the majority’s first reason for rejecting the plain language of the manual, the majority relies on various tes*1113timony about the JEDEC “patent policy” to arrive at the conclusion that the members of JEDEC “treated the language of Appendix E as imposing a disclosure duty.” This conclusion is contrary to testimony at trial showing that members of JEDEC understood Appendix E to describe the procedures to be applied once JEDEC has learned of a relevant patent, which is different from the requirement for disclosing relevant patents and patent applications. For example, Reese Brown testified that the “patent policy” has two distinct components:
Q. Can you tell me what the patent policy is?
A. Well, there are two parts. One that says that whenever material comes up in the committee for discussion and for voting, any members who are aware of any patent position or potential patent positions on the material should and are obligated to reveal that to the committee at that time.... The other portion of the policy has to do with if a specific patent material has been — or patent positions have been identified in connection with a proposal that is in the process of being approved for a ballot of standardization ....
John Kelly’s testimony likewise distinguished between the disclosure requirement and the requirement for obtaining “assurances” from a patentee once JE-DEC has learned of a relevant patent. Gordon Kelley had a similar view:
Q. Between 1991 and 1996 what do you believe that patent policy in JEDEC to be?
A. The stated policy was that, first of all, all member companies would notify the committee of patents that they were aware of that applied to a proposed standard. And another requirement was that they would agree that their licensing practice to all other member companies of JEDEC would be all companies would be licensed, excepting none, and that the license would be either free or offered at reasonable rates, without exception.
Later in his testimony, Mr. Kelley outright rejected the theory that the disclosure duty comes from the language describing what is to happen when JEDEC learns of a relevant patent:
Q. About one line down at the end of the sentence it starts with the word if, if the committee determines that the standard requires the use of patented items, then the committee chairperson must receive a 'written assurance, and it continues.
A. Yes.
Q. Sir, does that language accurately reflect your understanding of when a patent needed to be disclosed?
A. No. The language that I’m seeing here refers to a patent issue that has been raised in the committee....
Thus, according to the understanding of these witnesses, the language of Appendix E is only one part of the “patent policy”— the part that describes the appropriate procedures that the committee must apply once a disclosure has been made. Appendix E does not describe the second part of the patent policy: the obligation to disclose relevant patents and patent applications, as stated in section 9.3.1 of the JE-DEC manual. Moreover, the testimony quoted above from Reese Brown specifically refers to the duty to disclose when voting. The voting ballot parrots the language of section 9.8.1, requiring members to disclose patents and applications “involving th[e] ballot.” The ballot therefore confirms the separate duty of disclosure as stated in section 9.3.1, not Appendix E.
The majority also implies that the members of JEDEC do not use their own man*1114ual to learn about the rules they must follow, including the duty of disclosure. This conclusion, too, is not supported by the testimony at trial. Rambus’s Mr. Crisp described the manual as what “was used to tell people what the rules were.” In addition, both Mr. Kelly and Mr. Brown testified that they learned about the rules of membership from the JEDEC manual. Mr. Kelly testified as follows:
Q. So in the time period of 1991 through September of 1993, if we wanted to know the rules in JEDEC, we would look to [manual] JEP-21-H; is that right?
A. Yes, sir.... The manual contains — without reference to this text, I can tell you that the manual contains a reference to our patent policy EIA JE-DEC patent policy, which required the early disclosure of patents and patent applications that are or may be required to comply with the standard.
Q. In that manual, under 9.3.1, Committee Responsibility Concerning Intellectual Property, were the members of JEDEC told that the chairperson of any JEDEC committee, subcommittee or working group must call attention to the obligations of all participants to inform the meeting of any knowledge they may have of any patents or pending patents that might be involved in the work they are undertaking?
A. Yes, sir, it was.
Mr. Brown confirmed that the JEDEC manual is the appropriate place to look for the JEDEC patent policy. Shortly before the testimony quoted by the majority, Mr. Brown testified as follows:
Q. Okay. Now, is a patent policy in writing so other people can know what it is?
A. I believe that it is in writing.
Q. Where would we find it?
A. Probably in a document called “JEDEC” — something—“21” followed by a letter, which is a council-created documents which is a set of rules.3
In spite of this testimony, the majority concludes that members of JEDEC only learned about the disclosure duty from the viewgraphs displayed during meetings, and that there was no evidence that anything other than Appendix E was displayed at those meetings. Richard Crisp testified to the contrary:
Q. You’re aware, aren’t you, sir, that as of October 1993, JEDEC published a new manual for the memory committee, right?
A. Yes, I know that now.
Q. And you know that in the manual, it specifically states that patent applications have to be disclosed as part of the JEDEC application, don’t you?
A. Yes, I do know that.
Q. And during the time that you were going to these meetings, including in 1992 and in 1993, that new language was put up on the overhead projector at the meetings and dicussed by Jim Townsend, wasn’t it?
A. I think it was.
Q. You were aware that the chairman of the committee put up on the overhead projector at the meetings the new language of the policy which specifically required the disclosure of patent applications, right?
*1115[objection]
A. Yes.
Based on my reading of the record, I therefore believe that there was more than substantial evidence for the jury to have concluded that the disclosure duty is stated in section 9.3.1. of the JEDEC manual, not Appendix E, and that the members of JEDEC knew about this duty from the manual, voting ballots, and meetings.
The majority also rejects the language of section 9.3.1 as being overly broad and ambiguous. However, JEDEC was free to formulate whatever duty it desired and it is not this court’s job to rewrite or reinterpret the duty on the basis that it is unbounded (which I do not think it is). JE-DEC clearly knows how to draft rules and procedures with specificity when it so chooses. This point is amply demonstrated by the language of Appendix E and other sections of the manual describing in detail the rules that must be followed once JEDEC learns of a relevant patent or patent application. The fact that JEDEC chose not to use the same kind of language when stating the duty of disclosure indicates that it did not desire to have a bright line rule, such as the one the majority has now imposed upon it. Instead of creating a duty that it believes JEDEC should have adopted, the court need only determine that there was sufficient evidence of what the duty is such that a jury could apply the duty to the conduct at issue and determine whether the duty was violated. In my opinion, there was sufficient evidence for the jury to have concluded that the duty to disclose was stated by the plain text of section 9.3.1 of the JEDEC manual, requiring the disclosure of patents and pending applications that might be involved in the work of the committee.
II
Given the duty to disclose as stated in section 9.3.1 of the JEDEC manual, the next issue is whether substantial evidence supports a finding that Rambus faded to disclose pending and issued patents that might be involved in the development of the SDRAM standard. In my opinion, there is an abundance of such evidence.
The jury heard repeated admissions from Rambus that it had pending claims that not only related to the developing SDRAM standard, but even covered particular features of the standard.4 For example, Rambus’s business plan stated that “Sync DRAMs infringe claims in Ram-bus’[s] filed patents and other claims that Rambus will file in updates later in 1992.” In a March 14, 1995, email, Richard Crisp wrote from a JEDEC meeting that “[t]aken along with the fact that they rely on an externally bussed reference, this should be anticipated by some of our claims. I would say that the proposal may well infringe our work.” Mr. Crisp wrote many emails during JEDEC meetings noting instances where he believed Rambus had pending claims that covered features of the evolving standard. In one such email, he wrote: “Note that many of the SDRAMs use an externally supplied reference voltage in the input buffers. I believe we have a claim we added to cover this configuration.” Later in that same email, he said:
Suggested that one NC be used for a Vref for a high performance interface, (again we need to check claims about “DRAM with input receivers using an externally supplied reference voltage”). We may be able to slow down or stop (or at least collect from) all of the CTT, GTL, and HSTL devices if this claim is allowed (Allen, I believe this is one of *1116the claims you, Lester, Tracy and I wrote up in late '91, right?)
A 1993 document reporting on the status of drafting new claims derived from the '898 application stated: “(1) Writable configuration register permitting programmable CAS latency[.] This claim has been written up and filed. This is directed against SDRAMs.... (4) DRAM with multiple open rows .... This is directed against SDRAMs.” At trial, Richard Crisp testified at length about a particular instance of a claim drafted to read on a low voltage swing feature of the SDRAM standard proposed at one of the JEDEC meetings:
Q. So when we look at this here, we see Texas Instruments made a presentation for a synchronous DRAM 16 megabit, right?
A. Yes.
Q. If you look at one of [the] features they had on their 16 megabit presentation, it talks about they wanted a low voltage swing. Do you see that?
A. Yes.
Q. And then a few days later, Jim Gasbarro [a Rambus engineer] is meeting with the Rambus patent lawyer talking about low swing signals on a DRAM. Do you see that?
A. Yes.
Q. Then in February of 1992, the lawyer sends you the draft amendment to the claims, correct? If you want to refresh your recollection, you can look at Defendants’ Trial Exhibit 1531. You’ll see it’s a February 19th, 1992, letter.... And he refers to the enclosed revised draft preliminary amendment. Do you see that?
A. Yes, I do.
Q. So if you look at the preliminary amendment [at Plaintiffs Trial Exhibit 26], this is the final version that was filed on March 5th, 1992. Do you see that?
A. ... Yes.
Q. And if you look at the claims that were being added on March 5th, 1992, particularly claim 151, and if you look at the B element, it talks about low voltage swing signals, right?
A. Yes, it does.
Crisp also testified that in May of 1992, he met with Rambus’s patent attorney to “add claims to our patent application broad enough to cover the SDRAM if the SDRAM uses mode register and programmable CAS latency.” In perhaps the clearest admission of the case, Crisp testified that the features of double edge transfer, mode register, programmable CAS latency, programmable burst length and PLL, DLL “were discussed there [at the JEDEC committee meetings] in some form or another, and we certainly had patent applications that covered aspects of those, of those technologies.”
In my opinion, this evidence, which is just a portion of what Infineon presented at trial, is more than sufficient to support the finding that Rambus did in fact have pending patent claims related to, and even reading on, aspects of the SDRAM standard.5 The majority, however, requires a *1117different Mnd of proof than the clear admissions Rambus made through witness testimony and internal documents. By limiting the application of the duty to disclose to the issue of whether pending claims read on the final standard, the majority requires an element-by-element comparison of the limitations of a pending claim to the text of the SDRAM standard.6 Infineon did not call an expert witness at trial to make such a comparison; nor does it appear that Rambus presented a witness to prove the negative — that none of its pending claims ever read on any feature of the SDRAM standard discussed at the JE-DEC meetings.
The district court, however, did identify six instances where Rambus had pending claims related to the SDRAM standard. See Rambus, Inc. v. Infineon Techs. AG, 164 F.Supp.2d 743, 752-53 (E.D.Va.2001). Rambus argues on appeal, and the majority accepts the argument, that none of these pending claims actually reads on the SDRAM standard. The majority has gone so far as to make a de novo comparison of the pending claims to the JEDEC standard in order to conclude that no claims could possibly read on the standard. I do not believe that we, as an appellate court of review, are in a position to make this finding because neither party appears to have given the jury the necessary evidence to make such an analysis in the first instance. See Fromson v. W. Litho Plate & Supply Co., 853 F.2d 1568, 1570, 7 USPQ2d 1606, 1608 (Fed.Cir.1988) (“Obviously, a finding- not made cannot be reviewed!.]”). Rambus points to no trial testimony supporting the argument it now makes on appeal. See Shell Petroleum, Inc. v. United States, 182 F.3d 212, 218 n. 13 (3d Cir.1999) (“The appellant is required to provide a record to support the claims it makes on appeal.”). Certainly it was Infineon’s burden to prove up fraud by clear and convincing evidence, and I believe Infineon did so based in part on Rambus’s admissions; I do not believe In-fineon was limited to proving its case by a limitation-by-limitation claim analysis. To the extent Rambus wanted to rebut Infi-neon’s fraud case on the theory that Ram-bus did not actually have any pending claims that read on the standard, then it was incumbent on Rambus to prove this point at trial. It does not appear from the record before us that Rambus did so.
In addition, Rambus’s briefing on appeal only addresses five of the six instances of relevant pending claims identified by the district court. Rambus makes no chal*1118lenge to the district court’s conclusion that “the evidence shows that, when JEDEC discussed adding a 2-bank design and burst-length technology to the SDRAM standard, Rambus had pending claims relating to those technologies pending in its first patent application, the '898 application.” Rambus, 164 F.Supp.2d at 752. Absent any argument to the contrary from Rambus, I presume that the district court is correct, notwithstanding the majority’s independent conclusion based on its review of the pending claims, which in my opinion is beyond the scope of our review. See Fromson, 853 F.2d at 1570, 7 USPQ2d at 1608 (“This is the eighty-fourth case in which the court has been forced, ad nause-um, to remind counsel that it is a court of review, i.e., that it will not find the facts de novo, that it is not a place for counsel to retry their cases, [and] that its judges do not participate as advocates to fill gaps left by counsel at trial.... ”); Atl. Thermoplastics Co. v. Faytex Corp., 5 F.3d 1477, 1479, 28 USPQ2d 1343, 1345 (Fed.Cir.1993) (“Fact-finding by the appellate court is simply not permitted.”). Moreover, the majority concedes that the '898 application has claims relating to two-bank and burst length technology so long as “related to” is “construed more broadly than the duty” as defined by the majority.
Thus, in my opinion, substantial evidence supports a finding that Rambus failed to disclose pending patent applications that might be involved in the SDRAM standard. Rambus made numerous, unambiguous admissions to that effect and failed to prove anything to the contrary at trial.
CONCLUSION
This case is not an easy one, and I appreciate the majority’s efforts to find a bright line rule for what constitutes fraud in the context of standard setting organizations. But the majority s application of its rule, that only claims reading on the standard need be disclosed, is not the JEDEC standard. JEDEC’s disclosure policy required its members to disclose patents and pending applications that “might be involved in the work they are undertaking.” While the majority rejected this standard as unbounded, nothing required JEDEC to formulate its policy with precision and clarity. And, while the majority may believe that JEDEC’s “might be involved” standard is impossibly amorphous, the majority’s restatement of the JEDEC policy might prove impossibly complex. The majority’s application of its rule arguably requires a Markman claim construction, application of the doctrine of equivalents, a Festo analysis, and perhaps even a Johnson & Johnston analysis before anyone can say for sure whether a claim reads on a standard. As a result, an action for fraud will become more a federal patent case than a case arising under state law.
In any event, as I read the record, there is more than sufficient evidence upon which the jury could have concluded that Rambus had a duty to disclose pending and issued patents that might be involved in JEDEC’s development of the SDRAM standard and that Rambus violated that duty. I respectfully submit that the jury’s verdict should stand and I would therefore affirm the district court’s denial of Ram-bus’s motion for judgment as a matter of law.
. At trial, the parties disputed whether the duty to disclose pending applications existed prior to October 1993, when JEDEC revised its manual to explicitly require disclosure of both patents and pending applications. Witnesses testified at trial that the duty required disclosure of patent applications prior to October 1993. I therefore focus on the issue of whether the duty to disclose was limited to claims reading on the standard, as discussed by the majority opinion, or to patents and applications that might be involved in the work on a standard.
. The majority opinion states that to the extent the duty to disclose may encompass situations where an application describes but does not claim technologies under discussion at JEDEC, Rambus satisfied that duty by disclosing the '703 patent. I do not necessarily agree. Rambus's disclosure of a patent with clearly irrelevant claims does not absolve it of disclosing other applications with the same disclosure that might have relevant claims. I also note that there is evidence that Rambus’s disclosure of the '703 patent was deceptive because JEDEC members were led to believe that Rambus had nothing else of relevance to disclose.
. The particular JEDEC manual referred to herein is manual JEP-21-I, "JEDEC Manual of Organization and Procedure.”
. Thus, even under the duty to disclose as defined and applied by the majority, I believe that substantial evidence shows that Rambus violated that duty.
. The majority discounts Infineon's evidence that Rambus believed it had pending claims covering the standard. According to the majority, the JEDEC standard does not have a subjective belief component and a "member’s subjective beliefs, hopes, and desires are irrelevant.” I respectfully disagree. Rambus's statements that it believed it had pending claims covering the SDRAM standard is evidence that Rambus did in fact have claims covering the SDRAM standard. Moreover, Rambus’s belief is a critical component of the *1117overall fraud action, which includes the component of an intent to mislead. Rambus's beliefs about the scope of its duty are also relevant to what that duty actually is, just as the testimony cited in this dissent and in the majority opinion — where witnesses explain what they believe the duty to mean — is evidence of what the duty actually is.
. The majority states that its disclosure duty does not "require a formal infringement analysis,” even though the majority then proceeds to determine whether the pending claims read on the standard. While determining whether there is a "reasonable expectation that a license is needed to implement the standard” allows some degree of latitude beyond a rote comparison of pending claims to the relevant standard, the majority’s application of its standard is narrower than the duty it has defined. Its application of the duty is confined to consideration of whether or not pending claims read on the final standard.
In my opinion, the evidence I have identified in this opinion is sufficient to show that a competitor might reasonably expect that it should obtain a license from Rambus, regardless of whether or not any claim reads on the standard. For example, a reasonable competitor could conclude that Rambus could obtain claims reading on the standard or that Ram-bus had a plausible claim construction or doctrine of equivalents theory that is not readily apparent from a simple reading of the claims.