Rambus SC Opposition Brief In Full
Posted by kerry_lass (Registered) 12 Aug 2003 09:57 AM

Thanks to twelvesoutherners on yahoo and then richard suckle on SI

'(After file conversions, OCR, clean up, etc., here is Rambus's brief in opposition to Infineon's petition for certiorari at the
Supreme Court. I've gotten the text pretty clean and accurate, but some OCR misreads undoubtedly remain.)'

Opposition Brief

QUESTIONS PRESENTED
1. Whether petitioner failed to prove its state-law fraud
claim by the required clear and convincing evidence.
2. Whether the court of appeals correctly construed a
patent claim term in accordance with its plain ordinary
meaning, where the patent specification did not override
that meaning and the prosecution history affirmatively
supported it.

2 (p. iii)
TABLE OF CONTENTS
Page
QUESTIONS PRESENTED.......................................... i
TABLE OF AUTHORITIES......................................... iv
STATEMENT................................................................ 1
A. Background.................................................. 1
B. District Court Proceedings........................... 4
1. Patent Infringement................................ 4
2. State-Law Fraud..................................... 5
C. Court of Appeals Decision........................... 8
1. Claim Construction................................ 8
2. State-Law Fraud..................................... 10
ARGUMENT................................................................. 13
I. CERTIORARI SHOULD BE DENIED ON
THE STATE-LAW FRAUD ISSUE................. 13
A. Duty.............................................................. 14
B. Breach.......................................................... 17
C. Narrowness.................................................. 20
II. CERTIORARI SHOULD BE DENIED ON
THE CLAIM CONSTRUCTION ISSUE.......... 21
A. Dictionaries.................................................. 22
B. Ordinary Meaning........................................ 23
CONCLUSION.............................................................. 30

3--p.1
RESPONDENT'S BRIEF IN OPPOSITION

Infineon's petition presents only case-specific issues not
warranting further review. The primary issue merely chal-
lenges a record-dependent determination of evidentiary
insufficiency on a state-law fraud cause of action. The
secondary issue at bottom merely challenges a (unanimous)
construction of one particular patent claim term. In both
respects the ruling below is correct and raises no general legal
issue, let alone one that merits certiorari review.1
STATEMENT
A. Background. In 1990, inventors Mike Farmwald and
Mark: Horowitz filed a wide-ranging patent application
describing a wealth of inventions for speeding up the transfer
of data in and out of computer memory. They founded
Rambus to commercialize their inventions through licensing
and technological assistance to manufacturers like Infineon.
The counterpart European patent application, in all material
respects the same as the 1990 U.S. application, was published
by late 1991 and known to Infineon by May 1992. Pet. App.
28a, 31 a, 153a; Ct. App. JA ['CAJA'] 18268.
The United States Patent and Trademark Office (PTO)
quickly recognized that the 1990 U.S. application contained
many 'independent and distinct inventions' (35 U.S.C.
121). It therefore required Rambus to pursue those inven-
tions through numerous separate applications just as an
application describing a car made extremely fast by a host of
independent inventions would have to be divided into distinct
applications separately claiming those which operate inde-
pendently of one another. See Pet. App. 21a-22a. For
Rambus, the extensive process of refining and prosecuting the

1 No other publicly held company owns 10% or more of the stock of respondent Rambus. See Sup. Ct. R. 29.6.

4--p.2

separate applications continued through the 1990s. All the
patents and applications at issue in this case share the original
written description, which, in 1990, described all the invent-
tions later claimed.2 So fertile was the 1990 application that
dozens of patents have now issued to Rambus from that
application. Id. at 2a.
In December 1991, Rambus, a new company invited by
industry leaders, attended its first meeting of the Joint
Electron Devices Engineering Council (JEDEC), which was
part of the Electronic Industries Association (EIA). JEDEC
was a forum in which the major competitors in the
semiconductor industry engaged in standard setting. It had
regular meetings open to anyone paying dues, with the
minutes and proposed standards publicly available (Pet. App.
34a; CAJA1558, 3394-98)to develop common specifica-
tions for certain technologies so that different manufacturers
could make interchangeable parts.
There are numerous standard-setting organizations nation-
ally and worldwide. Such organizations recognize the rele-
vance of intellectual property to technological standards.
(Pet. App. 31 a). They also recognize (a) the legitimate inter-
est of firms in maintaining secrecy of ever-changing, often-
rejected patent-development efforts until they have been
made final through actual issuance of a patent (35 U.S.C.
122 (1994 ed.)), which minimizes pre-issuance priority
contests (35 U.S.C. 135) and jeopardy to foreign patent
rights (Pet. App. 34a n.9), see Infineon CA Br. 21 ("Rambus'

2 To be entitled to the 1990 priority date, later applications introducing new claims shared the original 1990 written description,
which had tosupport the new claims. 35 U.S.C. 120, 121 (continuation and divisional applications). The filing of such
applications has long been the standard procedure enabling inventors to secure protection in properly drafted claims for all
inventions described in an original application. See Johnson & Johnston Associates Inc. v. R.E. Service Co., Inc., 285 F.3d
1046, 1055 (Fed. Cir. 2002) (en banc);In re Hitchings, 342 F.2d 80, 85 (CCPA 1965).

5--p.3
pending patents were of course confidential'); (b) the
antitrust delicacy of competitors sharing information about
their in-process innovations where innovation is a central
aspect of rivalry (Pet. App. 23a n.7; Allied Tube & Conduit
Corp. v. Indian Head, Inc., 486 U.S. 492, 501, 506-07
(1988)); and (c) the interest in ensuring that firms with
relevant knowledge are not deterred, by onerous or unclear
membership duties, from participating in what can be
important public forums (Pet. App. 35a n.lO). They balance
such considerations in a variety of ways to adopt widely
varying duties of disclosure and licensing for members' deve-
loped or nascent intellectual property.3 Notably, the patent
policy of the umbrella organization, the American National
Standards Institute, does not require disclosure of mere patent
applications, limiting disclosure duties to certain issued
patents.4 In the present case, the particular disclosure duty for
members ofJEDEC, as part ofEIA, became a central issue.
Rambus was a member of JEDEC for calendar years 1992
through 1995. It attended its last meeting in December 1995
and confirmed its withdrawal by letter in June 1996. In 1993,
the relevant JEDEC committee adopted a standard for

3 An amicus brief submitted by GlobalPlatform et al. in the Federal Circuit noted (at 9) that 'circumstances vary widely in
standard setting' and there is no ''one size fits all' approach' to standard-setting organizations' disclosure policies. See also
Lemley, intellectual Property Rights andStdiulard Setting Organiialions, 60 Calif. L. Rev. 1889(2002). Elsewhere, counsel for
those amici, while mischaracterizing aspects of the evidence, wrote that the Federal Circuit decision in this case, in 'following
the letter of the law,' was no 'surprise.' www.consortiuininfo.org/bulletins (Feb. 2003).
4 See Testimony of Amy Marasco, at www.ftc.gov/opp/intellect-/020418marasco.pdf (ANSI general counsel explaining that
ANSI's 'Patent Policy does not apply to pending pateat applications'). A recent survey indicates that very few standard-setting
organizations require disclosure of unpublished patent applications, and even fewer require such disclosure from passive
members, like Rambus, who never propose standards. Lemley, 60 Calif. L. Rev. at 1904-05.

6--p4
synchronous dynamic random access memory (SDRAM); by
then, 'all JEDEC members had notice of the written
description of all of Rambus's patents.' Id. at 3 la. After
Rambus left JEDEC, the committee First proposed and later
adopted a standard for a double-data rate form of such
memoryDDR SDRAM. Not until 1998, two years after
leaving JEDEC, was Rambus able to draft and file in the PTO
the patent applications that, upon issuance in 1999 and 2000,
it asserted against products built to the SDRAM or DDR
SDRAM standards.
B. District Court Proceedings. In 2000, Rambus sued
Infineon for infringement of several newly issued patents.
Infineon counterclaimed, alleging that Rambus committed
fraud under Virginia law by failing to disclose, while it was at
JEDEC, certain earlier patent claims that Rambus had on file
in the PTO at that time. Because the filll written description
of all patents and applications at issue had in fact been
disclosed, the only alleged nondisclosure could be of claims
that Rambus had Filed in the PTO before leaving JEDEC.
Infineon's fraud theory relied on such pre-1997 claims, even
though most never issued and none has ever been asserted
against Infineon (or other JEDEC-compliant SDRAM or
DDR SDRAM makers).
1. Patent Infringement. During trial, the district court
granted Infineon judgment of noninfringement, based on a
severely scope-narrowing construction of the claim term
'bus.' That term appears alone, without modification by the
phrases that in other claims require a special kind of bus, in
certain claims to the other speed-enhancing inventions for use
with any 'bus.' But the district court ruled that all of Ram-
bus's claims required a special kind of 'multiplexing' bus
in which each of the lines that cany data also carry address
and control information. Id. at l00a.
Such multiplexing plainly is not part of the term's well-
known ordinary meaning: Infineon itself told the jury that 'It

7-p5
just says 'bus,' but it doesn't mean bus ***.' CAJA2001.
Indeed, the district court accepted that 'bus' had an ordinary
meaning not requiring multiplexing (and that dictionaries
could properly help identify such ordinary meaning). Pet.
App. 80a-81a & n.8. Nevertheless, the court held that, even
without modifiers, the term ''bus' actually has a specialized
meaning conferred by the specification of the patents in suit,'
so that even claims to Ranibus's other independent inventions
must require the multiplexing bus shown in the specification
describing embodiments combining such inventions.5 id. at
81 a, 80a-100a. On that basis, the court removed Rambus's
infringement case from the jury. Id. at 5a.
2. State-Law Fraud. Infineon's counterclaim of nondis-
closure-based fraud went to the jury. Infineon had the burden
of proving all the required elements by clear and convincing
evidence. The threshold issue was precisely what informa-
tion a JEDEC member had a duty to disclose and whether
Rambus had something undisclosed within that duty. Only if
Infineon and other members actually and justifiably expected
a particular fact to be disclosed could silence about it
constitute a false representation of its nonexistence. Because
the common written description was known, claims were the
only possible subject of nondisclosure. Id. at 28a-29a, 3 la.6

5 The description of a particular embodiment that includes multiple
inventions in particular forms responds to the basic statutory requirement
that the written description show 'the best mode contemplated by the
inventor of canying out his invention.' 35 U.S.C. sect. 112, para. 1 (emphasis added).
6 Under Virginia law, which requires a full inquiry when there is reason
to make inquiry, Infineon also had to prove that it reasonably relied on the
asserted nondisclosure of some patent claims from between 1991 and
1996despite Infineon's and JEDEC's having the full 1990 written
description for all the post-1990 claims, knowing that Rambus sought
additional patents (CAJA16524), openly worrying about what patents
Rambus was seeking based on the 1990 application (CAJA 1650-51,6534.
9054), concededly relying on others' assurances that any broad claims

8--p6
Infineon's position always has been that what required
disclosure was some subset of the class of 'pending and
issued patents' held by Rambus during its JEDEC mem-
bership, nothing more. Infineon CA Br. 14; Pet. App. 135a
(JMOL ruling); CAJA11717 (jury instruction). It was undis-
puted that no disclosure duty attached either to applications
filed in the PTO after a member left JEDEC or to a current
member's intention to file applications in the PTO. Pet. App.
34a.7 The latter indisputable limitation is critical because,
even in this Court, Infineon presents as its central grievance
the nondisclosure of plans to file claims broad enough to
capture Rambus inventions being incorporated into the
JEDEC standards. See Pet. 6, 19. The sole question at trial
was which claims within the class of current members' on-file
applications and issued patents had to be disclosed at
JEDEC.8 More specifically, the question was what disclosure

could be invalidated (Infineon CA Br. 7-8; CAJA2775-76), and being told
explicitly by Rambus not to rely on its silence (CAJA9115 C'[o]ur
presence or silence at committee meetings does not *** make any
statement regarding potential infringement of Rambus intellectual
property')), yet never directly asking Rambus to state what patent
protection it planned to seek. The sufficiency of Infineon's proof in this
regard was an issue not reached by the court of appeals.
7 Regarding plans, Infineon's JEDEC representative Meyer testified
(CAJA4637-38): 'Q: *** Did JEDEC patent policy require members to
disclose plans to file patents in the future? *** A. No. *** Q: Do you
recall testifying *** that JEDEC's patent policy did not require its
members to disclose any plans they had to modify applications in the
future? A: That's what I said. It was never mentioned in any of the
discussions. Q: *** [I]t was your understanding that that was not required;
is that right? A. Yes.' See CAJA 19099 (JBDEC official McGhee) (same);
Pet. App. 34a.
8 JEDEC members included IBM, Toshiba, Samsung, Hewlett-
Packard, and other companies collectively receiving thousands of patents
ever)' year. A large number of such patents relate in some general sense
to the DRAM memories with which JEDEC was concerned. Pet. App.
32a-33a; CAJA1580 (committee chairman: IBM itself had 'hundreds, if

9--p7
duty Infineon proved by clear and convincing evidence, i.e.,
what relation particular claims had to have to what particular
activity at JEDHC for the filing of such claims to trigger the
disclosure duty.9
The evidence, as recounted by the court of appeals (Pet.
App. 24a-35a), included JEDEC's written policies on disclos-
ure and, among other things, the testimony ofinfineon's own
JEDEC representative, Meyer, as to his understanding (which
necessarily limited Infineon's disclosure expectations).
Explaining his own nondisclosure of an Infineon patent
application 'related' in some sense to SDRAMs, Meyer made
clear the outer limits of the disclosure-triggering relation,
tying it naturally to the reason JEDEC wanted certain patent-
related information in the first place: a pending or issued
claim had to be related to a particular proposed standard in
the sense that practicing the standard would support an
assertion of infringement and hence a license demand. Id. at
30a.10 Only claims with that relation had to be disclosed.
As to whether Rambus, while at JEDEC, had any pending
or issued claims having that relation to standards under
not thousands,' of patents related to DRAMs). 4643 (Infineon's Meyer:
'lots' of patents related to SDRAMs).
9 There was a serious question whether passive members (those not
proposing standards) had to disclose mere patent applications, rather than
issued patents. Because that distinction was not the basis of the court of
appeals' ruling in this case, it is here assumed arguendo that the duty
encompassed patent applications.
10 "Q: What was your understanding of the relationship that a patent
had to have in order to be disclosed under JEDEC's patent policy? A:
Well, it had to be related to the work at JEDEC in the sense that it
described features that were necessary to meet the standard. Q: In other
words, in order to practice a standard, it would be necessary to use the
feature that was patented, right? A: Yes. Q: So if the patent -wouid not
be required to be wed in order to practice the standard, it didn't have
to be disclosed, right? A: If it wasas I said, if it was a circuit, which
could be done differently, then no.' CAJA4643-44; Pet. App. 30a
(emphases added).

10--p8
consideration at JEDEC, Infineon never contended that
Rambus did, and twice Infineon asserted that Rambus did not.
CAJA4798-99, 13195-; see Pet. App. 38a.
In the Face of such evidence, Infineon argued its case to the
jury at a high level of generality, relying on a broad-brush
assertion of a duty to disclose certain claims 'related' in an
unspecified sense to standards JEDEC was discussing. The
jury was persuaded: it found for Infineon on separate
questions for the SDRAM standard and the DDR SDRAM
standard. The district court granted Rambus judgment as a
matter of law as to the DDR SDRAM standard, but allowed
the verdict as to the SDRAM standard to stand, relying (like
Infineon) on an undefined 'related to' duty. Id. at 134a-135a,
143a-146a.11
C. Court of Appeals Decision. 1. Claim Construction.
The Federal Circuit unanimously reversed the district court's
claim construction in three respects and therefore remanded
with Rambus's infringement count reinstated. Id. at 8a-22a.
The court recited the accepted principles that 'a claim term
means 'what one of ordinary skill in the art at the time of the
invention would have understood the term to mean,'' id. at 9a
(quoting Markman v. Westview Instruments, Inc., 52 F,3d
967, 986 (Fed. Cir. 1995) (en banc), aff'd, 517 U.S. 370
(1996)), unless the inventors have 'use[d] the specification
[or prosecution history] to supply implicitly or explicitly new
meanings for claim terms.' Id. The court also noted the
basic principle that, '[w]hile claims often receive their inter-

11 The district court concluded that Infmeon had failed to prove that i
disclosure duty was triggered as to the DDR SDRAM standard, becausi
that standard was not even proposed at JEDEC until after Rambus left
Pet, App. 166a-169a. The court of appeals, without dissent, affirmed tha
grant of judgment as a matter of law (id. at 33a, 40a), which is not
the subject of Infineon's certiorari petition and is not otherwise addressed here.

11--p9
history, courts may not read limitations into the claims.' Id.
at 9a-10a (citing, e.g., Markman, supra). The court
then proceeded to apply those principles to three disputed
claim terms.
Regarding 'bus,' the claim term now at issue, the court of
appeals noted that the district court had not disputed the
ordinary meaning of the term but, rather, had concluded 'that
the patentees had acted as their own lexicographer by rede-
fining 'bus' to be a multiplexed bus'one whose ''signal
lines [are] used to transmit address, data and control
information.'' /(/. at 19a. Reviewing that conclusion, the
court of appeals first noted what was not disputed by the
district court:
The term 'bus' is very common in the electrical arts
and has a well-recognized meaning in such arts, namely,
a set of signal lines (e.g., copper traces on a circuit
board) to which a number of devices are connected, and
over which information is transferred between devices.
Id. (citing The New IEEE Standard Dictionary of Electrical
and Electronic Terms 141 (5th ed. 1993)). As the district
court had accepted, that ordinary meaning requires no
multiplexing. Id. The court of appeals then noted that
'[nJothing in the claims' compels a different construction but
that '[b]efore according 'bus' this meaning, *** [it] must
consider the usage and meaning of the term as used in the
relevant context of the specification.' Id.
The court observed that most references to 'bus' in the
specification do not even arguably suggest a limitation of the
ordinary meaning to a multiplexing bus. Id. It then noted
two statements about 'the present invention' that might
suggest such a limitation (id. at 19a-20a), but it concluded
that, considered in context, they do not. Id. at 20a-21a. In
particular, 'the prosecution history shows that a multiplexing
bus is only one of many inventions disclosed in the [original
1990] application.' Id. at21a. 'Although some of Rambus's

12--p10
claimed inventions require a multiplexing bus, multiplexing is
not a requirement of all of Rambus's claims.' Id. The
original application's claims pointedly added modifiers to
'bus' when claiming the 'inventive multiplexing bus,' con-
firming that 'bus' without such modifiers was not already
limited to multiplexing in the specification. Id. And the
PTO's requirement that Rambus divide its multiple inven-
tions into separate applications, some requiring multiplexing
and others not, 'demonstrated an understanding of 'bus' that
is not limited to a multiplexing bus.' Id. at 21a-22a.
2. State-Law Fraud. The court of appeals also reversed
the district court's denial of judgment as a matter of law
respecting the SDRAM standard. Id. at 22a-39a. 'Infineon
bore the burden of proving the existence of a disclosure duty
and a breach of that duty by clear and convincing evidence.'
id. at 39a. The court concluded that '(n]o reasonable jury
could find' that Infineon had carried its burden. Id.
The court first reviewed what Infineon proved the EIA/
JEDEC disclosure duty to require. Id. at 24a-35a. Infineon,
before the court of appeals, relied entirely on a simple and
unelaborated assertion that the JEDEC duty was to disclose
claims ''relating to' JEDEC standardization' (Infineon CA
Br. 21-22)as it does in this Court, too (Pet. 5, 12). The
court of appeals, however, examined what 'relation' Infineon
proved a patent claim had to have to standards to trigger
disclosure.12 The court observed that the governing EIA/
JEDEC policythe only such language ever shown to
members'identifies the duty to disclose based on the scope
of claimed inventions that would cover any standard and
cause those who use the standard to infringe.' Pet. App. 28a.
The court then recounted the description of the policy by the

12 This Court has noted the 'indeterminacy' of a 'relates to' standarc
without more concrete definition. New York Conference of Blue Cross v.
Travelers Ins. Co., 514 U.S. 645, 655-56 (1995) ("unhelpful text").

13--p11
relevant committee chairman, Kelley, as requiring a patent's '
disclosure when ''design or production of the component that
was being standardized would require the use of that patent'' ,
(id. at 29a). See id. at 30a (''the JEDEC policy *** of
notilying the committee when there are patents issued that
*** read on or apply directly to the activities of a standards
process''). The court also recited the critical understanding of
Infineon's own JEDEC representative Meyer (see note 10,
supra) that, for a patent claim to be within the disclosure
duty, ''it had to be related to the work at JEDEC in the sense
that it described features that were necessary to meet the
standard'' Id.
In accordance with 'the primary JEDEC goal of adopting
open standards that can be practiced without unreasonable
license fees or terms,' the court concluded that Infineon's
own witnesses, together with the documentary evidence,
'indicate that the relevant disclosure duty hinges on whether
the issued or pending claims are needed to practice the
standard.' Id. at 31a (footnote omitted). Only 'claims in
patents or applications that reasonably might be necessary to
practice the standard''claims that a competitor or other
JEDEC member reasonably would construe to cover the
standardized technology'had to be disclosed. Id. at 31a-
32a. Infineon's proposed 'amorphous duty' based on a
claim's 'having a vague relationship to the standard' was
further undermined by its own nondisclosure of an SDRAM-
related patent (because, as Meyer said, its use was not
required) and by the tiny number of patents and applications
actually disclosed by the industry giants participating in
JEDEC. Id. at 32a-33a. The court noted that plans to file
patent applications did not require disclosure (id. at 34a), and
it took care to note that JEDEC 'could have drafted a policy
broad enough to capture a member's failed attempts to mine a
disclosed specification for broader undisclosed claims. It
could have. It simply did not.' Id. at 35a (emphasis added).

14--p12
The remaining question was whether Infineon proved that
Rambus failed to disclose any claims, pending or issued while
it was at JEDEC, that came within the disclosure duty-claims
that would reasonably support an infringement claim (and
thus support a license demand). Infineon never argued to the
panel that Rambus had any such claims, even when
specifically asked at oral argument and even though Rambus
had, in its opening briefs opening argument, discussed each
of the pending applications the district court had identified as
improperly kept undisclosed (id. at 143a-144a). See id. at 38a
(despite Rambus's 'repeated assertions *** that these claims
were not necessary to practice the SDRAM standard, Infineon
does not directly address Rambus's arguments'). Neverthe-
less, the court of appeals, doing Infineon's work for it,
reviewed those applications and concluded that none could be
reasonably read to be required by the SDRAM standard. Id.
at 35a-38a. Accordingly, 'Infineon did not show the first
element of a Virginia fraud action and therefore did not prove
fraud associated with the SDRAM standard.' Id. at 39a.13
Judge Prost dissented on the record-specific ground that, in
her view, Infineon had presented sufficient evidence of a
breached duty of disclosure regarding the SDRAM standard.
Id. at 44a-66a. She did not dissent with respect to either the
claim construction or the DDR SDRAM judgment as a matter
of law. Infineon sought panel and en banc rehearing, with
support from four amicus briefs, but the requests were denied
with no record of dissent. Id. at 68a.

13 Although Infineon criticizes the court of appeals for asking the
parties for 'documents that were not in the appellate appendix' (Pet. 9
(emphasis added); id. at 2, 17, 18), it cannot and does not say that the
court requested, received, or relied on material outside the district court
record. Securing record material not included in the appendix is routine.
Sec Fed. R. App. P. 30(a)(2) ('Parts of the record may be relied on by the
court or the parties even though not included in the appendix.'); Sup. Ct.
R. 26.2.

15--p13
ARGUMENT
Neither of Infineon's two challenges to the court of ap-
peals' decision warrants review by this Court. The decision
involves no significant legal rulings, let alone federal-law
rulings on which a conflict of the circuits has been or could
be suggested. The court of appeals applied undisputed legal
standards for reviewing evidentiary sufficiency on a state-law
claim and followed established principles in construing a
patent claim term. Its decision on both the state-law issue and
the patent issue is a wholly fact-specific application of settled
legal standards. Most of Infineon's petition, indeed, does not
even purport to present legal issues, but only a high-pitched
disagreement with the result in the particular case. Such a
disagreement, even when accompanied by a baseless assault
on the Federal Circuit's overall fidelity to law, does not
warrant review. In any event, Infineon is wrong, and the
court below was correct, on the intensely record-specific
evidentiary and claim-construction matters it decided. The
petition should therefore be denied.
1. CERTIORAM SHOULD BE DENIED ON THE
STATE-LAW FRAUD ISSUE
Attempting to get around this Court's consistent practice of
not reviewing state-law matters (Wharf (Holdings) Ltd. v.
United Int'l Holdings, Inc., 532 U.S. 588, 596 (2001)),
Infineon attacks the court of appeals' sufficiency-of-the-
evidence ruling. Review for sufficiency is authorized by Fed.
R. Civ. P. 50 and the Seventh Amendment and goes on daily
in federal trial and appellate courts.14 Indeed, such review is
'necessary for the vigorous firnctioning of the jury system'

14 See, e.g., Weisgram v. Marley Co., 528 U.S. 440, 450-52 (2000);
Gasperini v. Center for Humanilies, Inc., 518 U.S. 415, 436 n.20 (1996).
Neely v. Martin K. Eby Conslr. Co., 386 U.S. 317, 321, 324, 327-28
(1967); Galloway v. United States, 319 U.S. 372, 389 (1943); Boyle v.
United Technologies Corp., 487 U.S. 500, 514 (1988).

PART TWO

16--p14
and 'essential to the preservation and proper application of
the applicable substantive law.'15 Whether the evidence is
insufficient in a particular case necessarily depends on the
record in that case-all the evidence. Reeves v. Sanderson
Plumbing Prods., inc., 530 U.S. 133, 150-51 (2000). Evi-
dentiary sufficiency on a state-law question, if ever cert-
worthy, certainly is not in this case. Indeed, the court of
appeals' ruling was correct.
A. Duty. Infineon's first argument does not even purport
to be a legal one. It does not question the court of appeals'
'reasonable jury' standard requiring that the actual evidence
reasonably meet the burden of proof (here, clear and
convincing evidence) on each element of the claim. (Indeed,
Infineon does not question the rulings of the district court and
a unanimous court of appeals finding insufficient evidence as
to the DDR SDRAM standard.) Rather, Infineon simply
challenges the judgment that the particular evidence in this
case failed to establish a breached disclosure duty as the first
essential element of fraud under Virginia law. Pet. 11-16.
That challenge is not certworthy, and the bits of evidence
recited by Infineon in no way undermine the court's con-
clusion as to the maximum disclosure duty supportable in the
record: the 'relation' required for an on-file claim to trigger
disclosure was that it must be reasonably read to cover a
JEDEC standard and hence support a license demand.
Thus, the passing statement by JEDEC official John Kelly
(who had never attended a meeting of the JEDEC committee
at issue and who was not purporting to offer an interpretation
of the disclosure duty) itself speaks of claims "that are or may

15 9A C. Wright & A. Miller, Federal Practice and Procedure 2522
at 244-246 (2d ed.l995) (internal quotation marks omitted), cited it
Gasperini, 518 U.S. at 436 n.20; see Warner-Jcnkinson Co. v. Hilton
Davis Chem. Co., 520 U.S. 17, 39 11.8 (1997) (importance in patent cases)
Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209
(1993).

17--p15
be required to comply with the standard.' CAJA3360.
Gordon Kelley, the committee chairman, even in the
Infineon-quoted excerpt spoke of claims that 'cover' JEDEC
standard proposals (Pet. App. 53a; CAJA15975) and
elsewhere confirmed that what triggered disclosure was that
'the design or production of the component that was being
standardized would require the use of that patent' (Pet. App.
29a; CAJA19057; see also id. 19055). Infineon quotes the
district court's summary of JEDEC consultant Reese Brown's
testimony (Pet. App. 141a-142a), which used 'relating to' as
a shorthand in focusing on other issues and, when indicating
what the required relation was, spoke of members having a
'patent position or potential patent positions on the material'
in a standard proposal. CAJA19031, 19032. Infineon'sown
JEDEC representative Meyer, while similarly using the short-
hand 'in relationship to' (Pet. App. 53a-54a; CAJA2745),
made crystal clear -what the relationship had to be-as
quoted at Pet. App. 30a and note 10, supra. (All emphases
above added.)
When Infineon turns to the actual EIA/JEDEC patent
policy from EIA/JEDEC documents, it simply ignores the
only policy that was ever shown to members-the EIA policies
shown as slides at every meeting Rambus attended. See Pet.
App. 25a-27a. That language, as Infineon does not even
bother to deny, 'links the disclosure duty to patents or appli-
cations whose claims cover the proposed JEDEC standard.'
Id. at 27a-28a. Instead, Infineon quotes a directive to
committee chairpersons that (a) was not shown to members as
the disclosure duty and (b) was not by its terms 'a statement
of the duty, but *** a requirement [for] the chairperson to
point members to the duty in' the above-referenced policy
that was shown to members (the one Infineon ignores). Id. at
27a. And the ballot language, aside from being limited to
'patents' (not applications), speaks of ''patents involving this
ballot'' (id. at 52a) without changing the understanding of

18--p16
how the patent had to involve the balloted standard, as
indicated in the policy actually shown to members.
Infineon criticizes the court of appeals for noting the
'amorphous' and 'unbounded' nature of the wholly unde-
fined 'related to' duty that Infineon attributed to JEDEC.
Pet. 14. But that fact is relevant to evidentiary insufficiency:
it renders Infineon's proffered duty 'implausible' on the
record showing so few patents or applications disclosed at
JEDECa record that Infineon has to explain away as
showing that industry giants in JEDEC pervasively violated
the supposed disclosure duty. Id. In contrast, the underlying
concern with whether license demands will be made against
products built in conformity with an adopted standard fits the
duty to disclose as shown by the evidence as summarized by
the court of appeals. Pet. App. 3 la. Contrary to Infineon
(Pet. 15-16), standardized products are built under adopted,
not rejected, standard proposals; and license demands require,
at a minimum, patent claims that reasonably read on adopted
standards. That inquiry, while not perfect in predictability, is
the one familiar to technical people dealing daily with
technical specifications and technical patents. Indeed, patent
claims 'must be understood and interpreted by the court as
they would be understood and interpreted by a person in that
Field of technology.' Multiform Desiccants, Inc. v. Median,
Ltd. 133 F.3d 1473. 1477 (Fed. Cir. 1998).16

16 Infineon says that the disclosure-duty issue was not properly raised
by Rambus below. Pet.15 n.4. That contention is not part of the question
presented and is wrong, as shown by the Federal Circuit's unanimous
treatment of the argument on the merits. Rambus's very first argument on
fraud was that the disclosure duty, whatever it might affirmatively
encompass, reached no farther than filed or issued claims supporting a
license demand, of which Rambus had none while at JEDEC. Rambus
CA Br. 26-29. Infineon recognized and addressed the argument, asserting
a broader duty. Infineon CA Br. 16-18, 21-22. This was again Rambus's
primary argument in its reply brief (at 1-8 & 30 n.17) and at oral

19--p17
B. Breach. Turning from what the duty was to whether
Rambus breached it, Infineon's primary point about the court
of appeals' insufficiency holding is framed as a legal argu-
ment. It suggests that a remand was required for the district
court to consider whether Infineon should receive a new trial.
Pet. 16-18. This argument is incorrect.
The very decisions Infineon cites make explicit that, under
28 U.S.C. sect. 2106 and Fed. R. Civ. P. 50(d), an appeals court,
upon finding evidentiary insufficiency, acts properly in
ordering judgment as a matter of law without such a remand,
where, as here, the verdict winner had an unimpeded
opportunity to present its evidence. See Weisgram, 528 U.S.
at 449-452 (affirming appellate JMOL); Neely, 386 U.S. at
321, 324, 327-28 (affirming appellate JMOL); see also. New
York. N.H. & H. R. Co. v. Henagan, 364 U.S. 441 (1960);
Herdman v. Pennsylvania R. Co., 352 U.S. 518 (1957). In
such circumstances, bringing the claim to an end ensures
fairness to both sides and serves the essential Rule 50 policy
'to speed litigation and to avoid unnecessary retrials' (Neely,
386 U.S. at 326). A remand may be necessary where, for
example, the jury failed to rule on the plaintiffs expressly
presented alternative grounds of liability (so that
insufficiency of evidence on the limited grounds addressed by
the jury cannot end the case), [acurci v. Lummus Co., 387
U.S. 86 (1967), or where the evidentiary insuffic-iency may
be due to exclusion rulings or other actions of the trial court
(Neely, 386 U.S. at 327). See Weisgram, 528 U.S. at 451 n.7,
456 n.l3. Even then, such specific grounds warranting a
remand for consideration of a new trial must be raised before
the appeals court. Neely, 386 U.S. at 327-28; see Fed. R. Civ.
P. 50(d). But no such circumstances are remotely present, or
were identified to the court of appeals, here: the jury fully
addressed Infineon's contentions and Infineon incurred no

argument, as to both the SDRAM and DDR SDRAM issues. Infineon
said nothing about waiver until its rehearing petition below.

20--p18
impediment to full presentation of its case at trial. Infineon is
simply demanding a second bite at the apple after failing to
prove its case, which basic principles of fairness, efficiency,
and finality preclude. See, e.g., Boyle, 487 U.S. at 513-14;
Burson v. Carmichael, 731 F.2d 849, 854 (Fed. Cir. 1984).
Infineon's secondary argument criticizes the court of
appeals for 'scouring the record to determine whether'
Infineon proved a breach of the maximum disclosure duty the
record would support. Pet. 18. Infineon offers no legal
citation for this criticism, which is badly misplaced. The
court could have simply held Infineon to its admission, by its
silence on appeal, that Rambus had no undisclosed claims
supporting a license demand. In responding to Rambus's
plain contention that it had no undisclosed pending claims
assertable against practice of a proposed standard at JEDEC
(Rambus CA Br. 27-28, discussing specific pending claims
based on district court's citations), Infineon, before the panel
(in its briefs and at oral argument), never even denied
Rambus's contention, much less identified a single pending
claim purporting to disprove it. It chose instead to rely solely
on arguing for a broader disclosure duty. Appellees have a
duty to raise their arguments and identify supportive parts of
the record in their brief, not save arguments for rehearing or
foist on appeals courts the task of combing the record for
themselves.17 The Federal Circuit, in examining the sole
pending Rambus claims that had even arguably been placed
in issue, thus gave Infineon fuller consideration of the merits

17 See Fed. R. App. P. 28(a)(9)(A), (b); Pentax Corp. v. Robison, 135
F.3d 760, 762 (Fed. Cir. 1998); FDIC v. Massingill, 30 F.3d 601, 605 (5th
Cir. 1994); Costo v. United States, 922 F.2d 302, 302-03 (6th Cir. 1990);
Morrison Knudsen Corp. v. Fireman's Fund Ins. Co., 175 F.3d 1221,
1239 n.l8 (10th Cir. 1999); Johnson v. City of Ft. Lauderdale, 126 F.3d
1372, 1373 (11th Cir. 1997).

21--p19
On the merits, Infineon rehearses an evidentiary-suffi-
ciency contention (Pet. 18-20) that is wholly case-specific
and, in any event, meritless. Dispositively, Infineon does not
now identify even a single undisclosed patent claim that
Rambus had pending while at JEDEC that could reasonably
support infringement (and hence a license demand). In the
district court, hifineon submitted a 'Statement Identifying
Specific Pending Rambus Patent Applications that Related to
JEDEC's SDRAM and DDR SDRAM Standardization Work'
(July 6, 2001), from which the district court's discussion on
JMOL evidently was drawn (Pet. App. 143a-144a). But that
submission spoke only of claims 'related to' JEDEC
standardization, and Infineon twice told the district court that
Rambus did not get broad enough claims filed until at least
1998. See p. 8, supra. Before the appellate panel, Infineon
again relied entirely on its assertion of a broader 'related to'
duty, without identifying any undisclosed pre-1998 Rambus
claim that could reasonably be read to cover a JEDEC
standard. In its rehearing petition, Infineon finally pointed to
one Rambus application, but Rambus, in its response, showed
that the contention was incorrect and waived (e.g., this
application was not even listed in Infineon's above-cited
district court submission). Infineon has now abandoned all
pretense of being able to identify such a claim.
Where the universe of on-file patent claims was sitting in
the record and available for direct examination, anything less
than identification and direct discussion of individual claims
could not suffice to establish whether, objectively, they could
reasonably be read to cover relevant JEDEC standards. The
inquiry, like an infringement inquiry, must involve element-
by-element comparison of individual claim terms to the
particular JEDEC standards. See Warner-Jenkinson, 520
U.S. at 29. Infineon cannot here make up for its utter failure
to do so by now pointing to (a) certain company officials'
unconfirmed, soon-corrected, assumptions about claims they
had not even directly read and compared to standards (e.g., a

22--p20
1992 Rambus business plan. Pet. App. 60a; CAJA16382, or
an email slating 'I believe' Rambus had certain claim but
noting 'we need to check,' Pet. App. 61 a; CAJA16050,
16054), or (b) testimony about plans to add certain claims
(Pet. App. 62a, 145a, quoting a mere question by Infineon's
counsel, answered negatively, 'I'm not sure that I did that'
and 'I didn't know what we had claimed or what we hadn't
claimed,' 5/2/01 Trial Tr. 133-34), or (c) testimony about
having claims on undefined 'aspects' of certain technologies,
which does not even begin to suggest coverage of specific
JEDEC versions (Pet. App. 63a; CAJA13202-04). The court
of appeals here correctly found such evidence insufficient,
given that subjective beliefs about patent coverage are
'largely irrelevant' to their objective scope,18 that Infineon
conceded the absence of Rambus claims broad enough to
cover the standards, and that Infineon failed to identify a
single undisclosed pending Rambus claim that could
reasonably be read as covering the standards.
C. Narrowness. The court of appeals thus correctly held
that Jnfineon failed to prove either the broader duty on which
it relied or the existence of filed Rambus claims within the
maximum pre-1996 JEDEC duty Infineon proved. The court
of appeals' decision, even in its state-law aspects, is no
broader than that. It does not reach other standard-setting
organizations, or even any post-1996 policies of JEDEC
itself, and it expressly does not limit any standard-setting
organization's ability to define its disclosure policies. Pet.
App. 35a ('JEDEC could have drafted a patent policy with a
broader disclosure duty. *** It simply did not.'). The
decision is thus narrow as well as correct, and it presents no
issue warranting certiorari review

23--p21
II. CERTIORARI SHOULD BE DENIED ON THE
CLAIM CONSTRUCTION ISSUE
Infineon's second question tries to make a general legal
issue out the court of appeals' unanimous construction of the
claim term 'bus.' Pet. 20-29. But the court of appeals
followed a perfectly correct methodology, over which there is
no dissension in the Federal Circuit or in other circuits.19 It
started with the ordinary meaning of the claim term, which
was not disputed even by the district court (and was effec-
tively conceded by Infineon), then analyzed the specification
and prosecution history to determine whether the inventors
had adopted a different meaning. The court found that the
specification did not alter the ordinary meaning and that the
alteration urged by Infmeon was positively incompatible with
the prosecution history. This claim construction (returning
the case, currently in an interlocutory posture, to proceed on
infringement) is correct and presents no issue for review.
In trying to make a general legal issue out of this familiar
course of claim construction, Infineon's entire presentation
rests on a conflation of two distinct issues. One is whether
the ordinary meaning of a claim term (to a person skilled in
the relevant art) presumptively governs its proper construc-
tion, subject to a different meaning being established, for
example, in the specification or prosecution history. The
second is whether, in ascertaining the ordinary meaning, the
court may look to relevant dictionary definitions if needed.
With those questions separated, it is clear that there is no

19 Infineon asserts that 'there can be no circuit split' on patent issues
(Pet. 2), but that is not so. See Holmes Group, Inc. v. Vornado Air
Circulation Systems, Inc., 535 U.S. 826 (2002) (case involving patent
counterclaim appealable only to regional circuit); Telecomm Technical
Servs. v. Siemens Rolm Communications, Inc., 295 F.3d 1249 (Fed. Cir.
2002) (transferring to regional circuit appeal in case involving patent
infringement counterclaim).

24--p22
issue in need of any review, least of all in this case, where an
unimpeachable approach to claim construction was applied.
A. Dictionaries. Infineon seems to rely principally on a
criticism of the use of dictionaries, but that criticism is a red
herring in this case. The ordinary meaning of the very
common term 'bus' in the relevant field is not in dispute.
The district court accepted that the ordinary meaning did not
require multiplexing; the unanimous court of appeals made
the same point; Infineon admitted as much when it stated its
position that the term 'bus' in these patents 'does not mean
'bus''; and no dictionary definition was introduced that
included a multiplexing element, the only point in dispute.
The court of appeals' reference to a dictionary, thus, was
mere confirmation of an ordinary meaning that was already
obvious and beyond dispute: the question below was only
whether that meaning governed or whether it was displaced
by some narrower meaning compelled by the specification or
prosecution history. Any questions about the use of dic-
tionaries that might arise in other cases are utterly beside the
point in this case.20
In any event, there is no meaningful general issue about the
use of dictionaries to help illuminate a claim term's ordinary
meaning (if it has one). As with statutory construction,
dictionaries can be usefill-perhaps technical dictionaries are
especially useful in construing general terms well known to
technicians but less familiar to lay judges. That is so even
though dictionaries might, of course, not supply a unique
answer and must be applied in the relevant technological con-
text and not through "cherry-picked" or "random" selection.

20 Infineon, even while repudiating 'extrinsic' evidence, asserts, with
carefully ambiguous language, that Rambus's expert did not contest that
the 'speciFication uses the term 'bus' as Infineon construes that term.'

Pet. 28. All that this does or could mean is that Rambus's expert said that
certain parts of the specification described & particular inventive bus (by
using modifying phrases) as involving multiplexing.

25--p23
Pet. 28, 2. Not one word in any decision of this Court cited
by Infineon, or any other decision to our knowledge, is to the
contrary. The district court here recognized the appropri-
ateness of using dictionaries to illuminate ordinary meaning.
Every active judge of the Federal Circuit has written
opinionsand joined still others-confirming the usefulness
of dictionaries in this task.21 And the Federal Circuit has
been fully sensitive to the nuances of making appropriate use
of dictionaries. See, e.g., Brookhill-Wilk I, LLC v. intuitive
Surgical, Inc., 326 F.3d 1215, 1221-22 (Fed. Cir. 2003);
Texas Digital, 308 F.3d at 1203-04.
B. Ordinary Meaning. On the separate question of ordi-
nary meaning's presumptively governing status, Infineon has
tried to manufacture an issue by changing the subject,
presenting the issue as a dispute over the primacy of 'intrinsic

21 See, e.g., Abbott Laboratories v. Syntron Bioresearch, Inc. 2003 WL
21544291 (Fed. dr.), at *5; Onega Engineering. Inc. v. Raytek Corp.,
2003 WL 21517751 (Fed. dr.), at 6-7; Boehringer Ingelheim
Vetmedica. Inc. v. Schering-Plough Corp., 320 F.3d 1339, 1346 (Fed. Cir.
2003); Altiris, inc. v. Symantec Corp., 318 F.3d 1363, 1369 (Fed. Cir.
2003); Middleton, inc. v. Minnesota Mining & Manufacturing Co., 311
F.3d 1384, 1387 (Fed. Cir. 2002); Inverness Medical Switzerland GmbH
v. Princeton Biomeditech Corp., 309 F.3d 1365, 1369 (Fed. Cir. 2002);
Texas Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed. Cir.
2002), cert. denied, 123 S. Ct. 2230 (2003) (claim constniction question);
Union Carbide Chemicals & Plastics Technology Corp. v. Shell Oil Co.,
308 F.3d 1167, 1177 (Fed. Cir. 2002); Telefla, Inc. v. Ficosa North
America Corp., 299 F.3d 1313, 1325 (Fed. Cir. 2002); Rexnord Corp. v.
Laitram Corp., 274 F.3d 1336, 1344 (Fed. Cir. 2001); Dow Chemical Co.
v. Sumitomo Chemical Co., Ltd., 257 F.3d 1364, 1372-73 (Fed. Cir.
2001); Tate Access Floors, Inc. v. Maxcess Technologies Inc., 222 F.3d
958, 965 (Fed. Cir. 2000); Hill-Rom Co., Inc. v. Kinetic Concepts. Inc.,
209 F.3d 1337, 1340-41 (Fed. Cir. 2000); Optical Disc Corp. v. Del Mar
Monies, 208 F.3d 1324, 1335, 1338 (Fed. Cir. 2000); EMI Group M.
Amer., Inc. v. Intel Corp., 157 F.3d 887, 892 (Fed. Cir. 1998); Vitronics
Corp. v. Conceptronic. Inc., W F.3d 1576, 1584 n.6 (Fed. Cir. 1996). See
also Cybor Corp. i>. FAS Technologies. Inc., 138 F.3d 1448, 1459 (Fed.
Cir. 1998) (en banc).

26--p24
evidence.' But the court of appeals here did not question that
primacy, which is not in dispute. As Infineon itself recog-
nizes, the first piece of 'intrinsic evidence' is 'the claim
language.' Pet. 21 ('the claim language, specification, and
prosecution history'). Here. the claim language has an
indisputably clear meaning to persons of skill in the art: a
'bus' does not require multiplexing. Giving that language its
unambiguous meaning to skilled artisans in the field is giving
primacy to the primary piece of 'intrinsic' evidence, which,
therefore, is furnishing the required 'public notice' of patent
scope (id. at 26).
Indeed, Infineon appears to recognize that such an unam-
biguous meaning presumptively governs. Pet. 29. With that
proposition not really in dispute, hifineon's petition is wholly
dependent on a case-specific disagreement with the court of
appeals' unanimous conclusion that 'bus' had a clear ordi-
nary meaning. That disagreement, not even developed in the
petition, presents no general legal issue.
In any event, Infineon never directly contends that a
claim's ordinary meaning is less lhan presumptively
controlling. Nor could it so contend, without confusing the
statutory role of the claims with the statutory role of the other
parts of the specification. The claims define the invention (35
U.S.C. 112,1(2), whereas the lengthy specification
preceding the claims serves different firnctions: it picks out
and identifies 'the best mode' of carrying out the invention
and, more generally, provides a detailed enough teaching to
enable a skilled artisan to make and use the invention
(35 U.S.C. 112, 1). It is particularly important not to
confuse those roles where, as here, the specification describes
many 'independent and distinct' inventions (35 U.S.C.
121)-so that individual aspects of what is described in
combination in the specification can be and are claimed

27--p25
Contrary to the petition (Pet. 22), nothing whatever in this
Court's decisions undermines the presumptive primacy of
ordinary meaning in construing claims. The Court has long
made clear that 'it is the[] claims, not the specifications, that
afford the measure of the grant to the patentee' (Milcor Steel
Co. v. George A. Fuller Co., 316 U.S. 143, 145-46 (1942))
and that 'while the courts may construe the same in view of
the specifications and the state of the art, they may not add to
or detract from the claim.'22 This Court's decision in
Markman, like Warner-Jenkinson afterwards (520 U.S. at 24,
29, 33, 37), notes the important distinction between the
claims as defining the invention and the written description
(517 U.S. at 373, 379), and it nowhere denies the primacy of
ordinary meaning.23 Nor is there any inconsistency between
presumptive primacy of ordinary meaning and the principle
that a claim construction must comport with the whole
document (Markman, 517 U.S. at 389-90), just as there is no
inconsistency among the stalutory-construction principles
that a word presumptively bears its ordinary meaning and
must be construed in the context of the whole statute, which

22 Cimiotti Unhairing Co. v. American Fur Refining Co., 198 U.S. 399,
410 (1905) (emphasis added); see Continental Paper Bag Co. v. Eastern
Paper Bag Co., 210 U.S. 405, 419 (1908). Whether to narrow claim
meaning to features in a specification is a two-way street, for breadth of
claims, while expanding scope, can defeat validity, yet '[i]t is axiomatic
that a claim must be interpreted the same way in determining infringement
and invalidity' (5A D. Chisum, Patents 18.03[2][h] (1998)). See Horn
Machine Co. v. National Needle Co., 134 U.S. 388, 394 (1890) (rejecting
narrowing of claim meaning by patentee to avoid prior art).
23 Indeed, although this Court in Markman decided only that claim
construction was for judges rather than juries, the en banc decision it
affirmed (without criticism on this point) explained that, to overcome
ordinary meaning, 'any special definition given to a word must be clearly
defined in the specification. The written description part of the specifi-
cation itself does not delimit the right to exclude. That is the function and
purpose of claims.' Markman, 52 F.3d at 980 (citation omitted).

28--p26
can overcome ordinary meaning.24 And the other precedents
cited by Infineon (Pet. 22), even as it describes them in
parentheticals, likewise say nothing to undermine the pre-
sumptively controlling character of a claim's ordinary mean-
ing where it has one, with the descriptive portion of the
specification being examined for consistency with such a
meaning and helping to resolve ambiguity where it exists.25
Infineon is also wrong in arguing that the Federal Circuit is
in internal conflict over these principles. 'The rules are by
now well known'; ''[i]t is the claims that measure the
invention''; there is 'a heavy presumption that a claim term
carries its ordinary and customary meaning'; and 'claims are
not perforce limited to the embodiments disclosed in the
specification.' Amgen Inc. v. Hoechst Marion Roussel, Inc.,
314 F.3d 1313, 1324-25, 1327, 1328 (Fed. Cir. 2003).26

24 See, e.g.,Johnson v. United States, 529 U.S. 694, 706-07 & n.O, 715-
17 (2000); FDIC v. Meyer, 510 U.S. 471, 476 (1994); Smith v. United
States, 508 U.S. 223, 228-29 (1993).
25 Infineon invokes the interest in consistency with PTO practice (Pet.
27), but it cites absolutely nothing to suggest that there is a PTO practice
inconsistent with the presumptively controlling nature of ordinary
meaning.
26 These are old, repeatedly applied principles. See, e.g., Prima Tek II,
LLC v. PolyPap, SARL, 318 F.3d 1143, 1148-49 (Fed. Cir. 2003):
Johnson Worldwide Assw., Inc. v. Zebca Corp., 175 F.3d 985, 988-90
(Fed. Cir. 1999); Renishow PLC v. Marposs Swiela ' Per Azioni, 158 F.3d
1243, 1249-50 (Fed. Cir. 1998); Kegel Co., Inc. v. AMF Bowling, inc.,
127 F.3d 1420, 1427 (Fed. Cir. 1997); York Prods., Inc. v. Central
Tractor Farm & Farming Ctr., 99 F.3d 1568, 1572 (Fed. Cir. 1996)
('Without an express intent to impart a novel meaning to claim terms, an
inventor's claim terms take on their ordinary meaning.'); Markmon, 52
F.3d at 979-980 (en bane]. In re Paulsen, 30 F.3<1 1475, 1480 (Fed. Cir.
1994); Hoganas AB v. Dresser Indtis.. Inc., 9 F.3d 948, 951 (Fed. Cir.
1993); Intdlicall. Inc. v. Phonometrics. Inc.. 952 F.2d 1384, 1388 (Fed.
Cir. 1992); ZMI Corp. v. Cardiac Resuscitator Corp., 844 F.2d 1576,
1579 (Fed. Cir. 1988); Evirotech Corp. v. Al George, Inc., 730 F.2d 753,

29--p27
What is required to overcome the presumption of ordinary
meaning is inevitably context-specific: 'Whether an invention
is fairly claimed more broadly than the 'preferred embodi-
ment' in the specification is a question specific to the content
of the specification, the context in which the embodiment is
described, the prosecution history, and if appropriate the prior
art.'27 That context-specific analysis is precisely what the
court of appeals performed. It correctly concluded that the
particular inventive multiplexing bus was required only in
claims that used modifiers to the general term 'bus' to
identify a particular version of a bus, but not where the term
'bus' stood unmodified in claims to the other inventions
described in Rambus's multiple-invention application, as
confirmed by the original application and the PTO's
insistence on dividing it into separate applications claiming
the 'distinct and independent' inventions.28
The Federal Circuit decisions cited by Infineon (Pet. 22-24)
positively confirm this approach. Infineon's main authority,
Multiform Desiccants, 133 F.3d at 1477-78, expressly states

759 (Fed. Cir. 1984); Universal Oil Products Co. v. Globe Oil & Refining
Co., 137F.2d3, 6 (7th Cir.943),aff'd, 322 U.S.471 (1944).
27 Teleflex, 299 F.3d at 1327; see Sunrace Roots Enterprise Co. v.
SRAU Corp., 2003 WL 21665253, at '8 (Fed. Cir.); 5A D. Cilisum,
Patenis 18.03[2][b](ii] (1998); VLTCorp. v. Lambda Eleci., Inc., 238 F.
Supp. 2d 347, 350-51 (D. Mass. 2003) (so recognizing, cited at Pet. 25);
Borg Waner, Inc. v. Mew Venture Gear, Inc., 237 f. Supp. 2d 919, 928
(N.D. III. 2002) (same).
28 In SciMed Life Sys Inc. v. Advanced Caridovascular Sys., Inc., 242
F.3d 1337 (Fed. Cir. 2001)-from which Infineon cites Judge Dyk's
concurrence (Pet. 2Stthe construction, as the Federal Circuit has made
clear, rested on facts absent here: the specification made a 'broad and
unequivocal' assertion that 'all embodiments' included a particular fea-
ture and distinguished prior art on that basis. E.g., Honeyweli Inc. v.
Victor Co., 298 F.3d 1317, 1325 (Fed. Cir. 2002) (panel including Judge
Dyk); Swrace, supra, at *7-*8. See also Inverness, supra (per Dyk, J.)
(primacy of ordinary meaning).)

30--p28
that any 'special meaning' imparted by the inventor to a term
in the specification 'must be sufficiently clear in the
specification that any departure from common usage would
be so understood by a person of experience in the field of the
invention' and that ''[a] technical term used in a patent
document is interpreted as having the meaning that it would
be given by persons experienced in the field of the invention,
unless it is apparent from the patent and the prosecution
history that the inventor used the term with a different
meaning.'"29 The decision in Vitronics, supra-v/hose author
later wrote the Amgen decision quoted at page 26, supra
also recognizes that 'words in a claim are generally given
their ordinary and customary meaning,' to be overcome only
if a 'special definition of the term is clearly stated in the
patent specification or file history.' 90 F.3d at 1582; see also
id. at 1584 n.6 (approving use of dictionaries). Infineon's
three other authorities are fully in accord with this principle.30

29 Multiform Desiccants involved the claim term 'degradable,' which,
unlike 'bus' in this case, the court treated as having two possible relevant
ordinary meanings (dissolvable or breakable) on the point in dispute. The
court held that the patent required the former (the more natural one),
based partly on how the inventor distinguished prior art. 133 F.3d at
1477-78. Multiform Desiccants has been cited for the principle that '[a]
term of technical art, unless defined otherwise, by the patentee, has the
meaning by which it would be understood by persons experienced in the
field oflhe invention' (Talbert Fuel Systems Patents Co. v. Unocal Corp.,
275 F.3d 1371, 1375 (Fed. Cir.), vacated on other grounds, 123 S. Ct. 70
(2002); see Toro Co. v. White Consol. Indus., Inc., 199 F.3d 1295, 1299
(Fed. Cir. 1999)).
30 Infmeon cites Belt & Howell Document Mgmt. Prods. Co. v. Altek
Sys., 132 F.3d 701, 705-06 (Fed. Cir. 1997), only for the primacy of
'intrinsic evidence,' a non-issue here. Renishaw, 158 F.3d at 1249-50,
recognizes the presumptive primacy of ordinary meaning, subject to
rejection if it 'flies in the face ofthe patent disclosure.' Toro, 199 F.3d at
1299, citing Multiform Dessicants, recognizes 'the general rule that words
in patent claims are given their ordinary meaning in the usage ofthe field
of the invention, unless the text of the patent makes clear that a word was

31--p29
There is thus no need to clarify any claim-construction
principles relevant to this case. It is a commonplace that
claim-construction principles, like principles of statutory
construction, apply differently to different records in different
cases. The specification may be decisive where, for example,
the claim term lacks an ordinary meaning resolving the point
in dispute, or where a claim term has an ordinary meaning
that was redefined or otherwise disclaimed, expressly or by
clear implication, by the patentee. None of that was so here;
to the contrary, the claim has an ordinary meaning, there is no
disclaimer or redefinition, and the prosecution history, which
shows 'the framework of the invention as viewed by the
inventor' and 'the issues of patentability as viewed by the
patent examiner' (Multiform, 133 F.3d at 1477), precludes
Infineon's claim-narrowing treatment of all the independent
inventions as dependent on one of the inventions (a particular
configuration of bus). The Federal Circuit's reliance on
ordinary meaning in this case, which was consistent with
the specification and confirmed by the prosecution history,
warrants no further review.31

used with a special meaning.' (Toro on one issue rejected a narrowing
construction and on another found the terms 'cover' and 'including' too
open-ended, considered by themselves, to resolve the point in dispute.)
31 All four of the law-review articles cited by Infineon (Pet. 25 n.5)
reiterate the soundness of the principle presumptively giving a claim its
ordinary meaning. Hill & Cote, Ending the Federal Circuit Crapshoot:
Emphasizing Plain Meaning in Patent Claim Interpretation, 42 IDEA 1, 2
(2002); Gallagher, The Federal Circuit and Claim Construction:
Resolving the Conflict Between the Claims and the Written Description, 4
N.C. J. L. & Tech. 121, 121, 129, 142 (2002); Bender, Uncertainty and
Unpredictability in Patent Litigation: The Time Is Ripe for a Consistent
Claim Construction Methodology, 8 J. Intell. Prop. Law 175, 218-21
(2001); see also Hatterbach, Chickens, Eggs. and Other Impediments to
Escalating Reliance on Dictionaries in Patent Claim Construction, 85 J.
Pat. & Trademark Off. Soc. 181, 183 (2003). The Hatterbach article
focuses on the timing of considering dictionary definitions, which, if it
presents a significant issue at all, is a non-issue in this case.

32--p30
CONCLUSION
The petition for a writ of certiorari should be denied.
Respectfully submitted,

WILLIAM K. WEST, JR.
HOWREY SIMON ARNOLD &
WHITE, LLP
1299 Pennsylvania Ave,NW
Washington, DC 20004
GREGORY P. STONE
KRISTIH LINSELY MYLES
MUNGI-R TOLLES & OLSON LLP
355 South Grand Ave.
Los Angeles, CA 90071

August 6,2003

RICHARD G. TARANTO
(Counsel of Record)
FARR & TARANTO
122019thStreet,NW,#800
Washington, DC 20036
(202) 775-0184
MICHAEL 1. SCHAENOOLD
PAITON BOGOS LLP
2550 M Street, NW
Washington, DC 20037
JOHN DANFORTH
ROBERT KRAMER
RAMBUS INC.
4440 El Camino Real
Los Altos, CA 94022