United States. Congress. Senate. Committee on the.

The Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) online

. (page 130 of 140)
Online LibraryUnited States. Congress. Senate. Committee on theThe Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) → online text (page 130 of 140)
Font size
QR-code for this ebook


a simple one. and that the ENIAC machine correctly handled the mathematical
processes based upon the information coded into the machine.

13.36.21 Harper's letters of December 1, 1961, and December 5, 1961, as well
as Anderson's response of December 13, 1961, have been in the possession of
defendant's counsel, John P. Dority. since 1963.

13.36.22 On June 13, lfM3S. Honeywell served interrogatories on SR in this
action, seeking rhe identification and ultimate production of all materials in
defendant's possession which involved the operation of the ENIAC machine
during the time period July 1, 1944 through August 2.5, 1946, and which were
not introduced into evidence in any interference or public use proceeding ot
in the litigation between SR and BTL.

13.36.23 The Harper-Anderson correspondence was not identified in SR's an'
swers to Honeywell's interrogatories and was not produced for Honeywell until
after SR had been given notice, in April, 1970, that Honeywell intended to take
the deposition of the custodian of documents at Los Alamos.

13.36.24 On October 16. 1970. Honeywell brought a motion in this action to
abrogate SR's claim of privilege as to all materials relating to the Los Alamos
problem, this motion having been based in part upon Honeywell's external dis-
covery of the existence of the correspondence between Harper and Anderson
which had never previously been produced or identified for either BTL or
Honevwell.

13.36.2.5. Honeywell's motion to abrogate the claim of privilege with respect
to the Los Alamos problem documents was set for hearing on November 23,
1970.

13..36.26 On November 13, 1970, SR attorney Hall in AVashington. D.C. tele-
phoned Anderson to inquire concerning Anderson's December 13, 1961, letter
to Harper: Hall was informed by Anderson that the A.E.C. had in its possession
the letter iby Heyman dated January 4, 1962, and that the letter was available
to him tipon written request.

13.36.27 SR did not produce or report its knowledge of the Heyman letter
of January 4. 1962. at the November 23, 1970, hearing held before this Court
on Honeywell's motion to abrogate the claim of privilege with respect to Los
Alamos documents.

13.36.28 The Heyman letter of January 4, 1962, was first made known to
Honeywell by SR when it was ultimately produced for the court on December IS,
1970. with tiie explanation that SR did not have the letter until subsequent to
the anrument on Novemljer 23. 1970.

13.37 Counsel for defendants chose instead to confine the investigation to
Metropolis and Frankel.

13.37.1 The Los Alamos Laboratory personnel who performed the calculations
with the ENIAC machine at the ?Joore School, and reported their results to
Teller and others, were Dr. Nicholas Metropolis and Dr. Stanley Frankel.

13.37.2 On or about October 29, 19.59, Libman. then in private practice, prepared
an affidavit for the signature of :\Ietropolis for use in opno.sition to the IBM
public use petition, wherein Metropolis was to swear that, although he (Metrop-
olis) had a complex problem, he knew it could not be solved on the ENTAC
machine and tliat he visited the Moore School only in the hope of obtaining
information about the potentialities of an electronic digital computer.



5850

13.37.3 Metropolis refused to sign the October, 1949 affidavit prepared Ity Lib-
mau, but ou November 2, 1959, Metropolis did sign a revised affidavit which
had been prepared by SR and which was thereafter filed in the Patent Office,

13.37.4 The revised. November 2, 1959. Metropolis affidavit differed materially
from the earlier draft, omitting any reference to the complex problem, except
what could be deduced from the newly formulated statement that Metropolis
learned about some of the preliminary qualitative features of the physical
problems the calculational form of which represented a considerable simpliii-
cation.

13.37.5 An affidavit purporting to describe the Los Alamos problem was obtained
from Metropolis on January 31, 1962, and filed with the court in the SR v. BTL
case.

13.37.6 An affidavit of Frankel was obtained on February 6, 1962, which also
purported to describe the Los Alamos problem.

13.37.7 Such affidavits did not state sufficient facts to prevent their conclusory
content from being misleading as now appears from the record in this case.

13.38 They may have thus withheld information from the Patent Office and the
District Court.

13.38.1 The District Court for the Southern District of New York, based on
the limited evidence presented, found that the work done by the ENIAC machine
on the Los Alamos problem was highly classified, and (erroneously) that the
only two persons who knew the nature of the problem were Metropolis and
Frankel, who submitted affidavits that the results were not checked for accuracy,
that there were almost certainly some undetected errors.

13.38.2 Because no effort was made to obtain information from the A.E.C., and
information may thus have been withheld, all of the essential facts presented at
this trial concerning the Los Alamos problem, had not been placed before Judge
Dawson through the limited affidavits and/or deposition te.stimony of witnesses.

13.38.3 The New York District Court's findings v/as thus directly contrary to
the clear and convincing evidence presented in this action which establishes "that
tlie Los Alamos work on ENIAC was in fact a sub.'^.tantial effort which succes.s-
fully and satisfactorily solved specific problems and that the results were useful
and did not lie dormant.

13.39 The Court finds and concludes that despite the various derelictions of M
and E, defendants and their counsel that the claim of willful and intentional
fraud on the Patent Office has not been proved by clear and convincing evidence.

13.39.1 Honeywell has not proven by clear and convincing evidence any willful
and intentional fraud on the Patent Office concerning the ENIAC patent.

13.39.2 Honeywell has not proven tliat defendants or their predecessors violated
the Sherman Act in the filing and prosecution of the ENIAC patent application
and/or in the enforcement of the ENIAC patent.

13.39.3 Honeywell has not proven by clear and convincing evidence that
Eckert and Mauchly, their attorneys, successors or assigns, committed willful
and intentional fraud ou the Patent Office in connection with the public use of the
claimed subject matter of the ENIAC patent.

13.39.4 Honeywell has not proven by clear and convincing evidence that Eckert
and Manc'hly, their attorneys, sucessors or assigns, committed willful and inten-
tional fraud on the Patent Office in connection with the Los Alamos problem.

13.39.5 Honeywell has not proven by clear and convincing evidence that Eckert
and Mauchly, their attorneys, successors or assigns, committed willful and inten-
tional fraud on the Patent Office in connection with the February, 1946, demon-
stration and dedication.

13.39.6 Honeywell has not proven by clear and convincing evidence that Eckert
and JNlauchly, their attorneys, successor-^ or assigns, committed willful and inten-
tional fraud on the Patent Office in connection with Hartree's use of the ENIAC
machine.

13.39.7 All of the essential farts presented at this trial concerning the investiga-
tion of alleged iiublic use by the attorneys responsible for filing the ENIAC patent
application were not placed before Judge Dawson through the deposition testi-
mony of Max L. Libman.

13.39.8 Honeywell has not proven by clear and convincing evidence that
Eckert and Mauchly, their attorneys, successors or assigns, committed willful
and intentional fraud on the Patent Office in connection with the alleged "on
sale" condition of the inventive subject matter claimed in the ENIAC patent.

13.39.9 Honeywell has not proven by clear and convincing evidence that Eckert



585 1

and Mauclily, their attorneys, successors or assigns, committed willful and inten^
tional fraud on tJie Patent Office in connection with the work of Atanasoff.

13.39.10 Honeywell has not proven by clear and convincing evidence that
Eckert and Mauchly, their attorneys, successors or assigns, committed willful
and intentional fraud on the Patent Office in connection with alleged anticipation
by prior art patents.

l.j.39.11 Iluney well has not proven by clear and convincing evidence that Eckert
and Mauchly, their attorneys, successors or assigns, committed willful and
intentional fraud on the Patent Office in connection with the RCA Reports, tlie
AMP Report and/or the Burks Article.

13.39.12 Honeywell has not proven by clear and convincing evidence that
Eckert and Mauchly, their attorneys, successors or assigns, committed willful
and intentional fraud on the Patent Office in connection with the First Draft
Report.

13.39.13. Apart from attorney's fees, Honeywell has not proven that it suffered
any injury to its business or property by reason of the existence of the ENIaC
patent.

13.39.14 Honeywell has not proven that defendants or their predece.ssors
violated the Sherman Act in the filing and prosecution of the EXIAO patent
application and/or in the enforcement of tlie ENIAC patent against Honeywell.

[46] 13.39.15 To recover antitrust damages based on alleged fraud in ontiun-
ing a patent, the plaintiff must prove (1 ) willful and intentional fraud, (2) injury
to business or property caused by the fraudulently procured patent, and (3) rhe
other elements necessary to a Section 2 Sherman Act violation.

13.39.10 Good faith or an honest mistake is a complete defense to an antitrust
action based on fraud on the Patent Office.

13.39.17 To prove an antitrust violation based on fraud on the Patent Office,
the proof of fraud must be by clear, unequivocal and convincing evidence ; a
mere preponderance of the evidence is not enough.

13.39.18 "The road to the Patent Office is so tortuous and patent litigation is
usually so complex, that 'knowing and willful fraud' as the term is used in
Walker, can mean no less than clear, convincing proof of intentional fraud involv-
ing affirmative dishonesty, 'a deliberately planned and carefully executed scheme
to defraud * * *.' "

13.39.19 "A person attacking the validity of a patent for alleged fraudulent
representations must prove that the representations were material — that the-
patent would not have issued but for the representations."

13.39.20 An applicant for a patent need not set up "straw men which he
reasonably and in good faith believes he can knock down."

13.39.21 The requirement of "full, frank disclosure * * * does not mean an
applicant must list out the full spectrum of his knowledge to establish bona fides.''

14. 25 Other Patents and Patent Applications

14.1 These are the so-called 30A patents and patent applications.

14.1.1 Honeywell amended its Amended Complaint (by Order of Court entered-
September 19, 1969) to add paragraph 30A, alleging that SR claimed to own
or control a "Patent Portfolio" containing issued patents and pending I'atent
applications that are subject to the same types of infirmities as he ENIA(>
patent, and that the procurement, licensing and attempted enforcement by SR
of these additional patents and applications constituted a further part of the
oi'iginally alleged fraudulent and conspiratorial pattern of conduct by SR in
restraint of trade.

14.1.2 Honeywell contends that these applications and patents (hereinafter the
30A patents and applications") evidence a pattern of conduct calculated to
proc-ure and maintain a patent portfolio of dominance over the computer indu!=try,
and that the 30A patents and applications have been and are being procured
in disregard of the uncompromising duty of fairness and full disclosure to the
Patent Office and the courts in patent matters.

14.1.3 During the discovei-y stake of this lawsuit, SR produced for Honeywell's
inspection and copying hundreds of SR patent application files.

14.2 Several have been aliandoned, several have expired, and two — BIXAC
and UNIVAC — are still pending.

14.2.1 Honeywell has offered evidence with respect to twenty-five 30A patents
and applications to establish Sll's pattern of i)r()secuting apiilieations and riro-
curing patents known to be barred by printed publications, public uses, sales, or-
other invalidating infirmities.



5852

14.2.1.1 In addition to its cliarges concerning the EXIAC patent, Honeywell
has charged fraud on the Patent Ofiice in connection with two pending patent
applications, two abandoned patent applications, and 19 patents (ten of which
have expired).

14.2.2 These other patents and patent applications within SR's EDP portfolio
are as follows :

Patent No. or Issue date

application No. or condition

If no patent Filing if no patent

EiVl No. has issued date Name has issued

1 2,629,827 Oct. 31,1947 Regenerative memory Feb. 24, 1953

8 2,969,478 June 10, 1949 Electrostatic stoiage memory Jan. 24,1961

14 2,687,^73 Apr. 13. 1950 Cycling unit Aug. 24, 1954

19 2,708,554 July 29, 1950 Uniservo tape unit May 17, 1955

19a 3,189,290 do do June 15, 1965

19b 2,901,734 do do Aug. 25,1959

20 2,686,299 June 24, 1950 BINAC function table Aug. 19. 1954

21 2,600.744 Oct. 21, 1950 Serial binary adder June 17,1952

22 SN 179, 782 Aug. 16, 1950 BlNACsystem (')

23 2,646,501 Oct. 21,1950 Unit adder _ July 21,1953

25 - 2,673,293 do Seiial binary full adder Mar. 23,1954

26. _ 2,655,598 do Parallel binary adder Oct. 13,1953

27 SN 194,112 Nov. 4,1950 Excess-3 binary coded decimal adder (.-)

23 . SN 213, 072 Feb. 23, 1951 fanllel excess-3 binary coded decimal adder (2)

29 2,590,950 Nov. i6, 1950 Function table full aduer Apt. 1,1952

39 SN 279, 710 Mar. 31, 1952 UNIVACsystem (0

40 2,860,756 do Uniprinter _ Nov. 18,1958

41 2,850,325 do Unityper Nov. 11,1958

42 . 3,133,190 do UNIVAC arithmetic unit May 12, 1964

43 2,781,446 Mar. 28, 1952 UNIVAC cycling unit Feb. 12, 1957

44 3,056,947 Mar. 31, 1952 Card to taoe unit Oct. 2,1962

48 . 2,748,270 . do UNIVACcl'ck gateand pulseformer May 29,1956

2, 842, 6£3 June 10, 1955 Binary signal comparator.. July 8,1958

2,915,966 June 13,1955 High speed printer ___ Dec. 8,1959

2,938,193 June 10, 1955 Binary signal encoder May 4,1960

1 Still pending.

2 Abandoned.

14.3 Plantiff has not been directly charged with infringement of any of the
issued patents.

14.3.1 SR has raised no counterclaim in this case charging Honeywell with
infringement of any patents, but ISD has raised a counterclaim for infringement
of the ENIAC patent.

14.3.2 SR and ISD have represented to this Court that no charge of infringe-
ment has been directed against Honeywell under any patents other than the
ENIAC patent.

14.4 Plaintiff claims, however, that these patents and patent applications wei'e
referred to as part of defendants' portfolio in the negotiations which preceded this
lawsuit.

14.4.1 When Honeywell and other competitors within the EDP industry raised
their own patents against SR as trading material to offset ISD's charges of in-
fringement of the EXIAC patent, SR responded by corre.spondingly raising other
patents ajid pending applications, including its 30A patents and applications, for
bargaining purposes in the negotiation of any possible overall cross-license settle-
ment of the EXIAC patent controversy.

14.4.2 For example. ISD had sought a one and one-half percent royalty from
National Cash Register Co. (XCR) for license rights under the EXIAC patent;
and in the negotiation for a possible overall cross-lieen.se between XCR and SR.
including the EXIAC patent being asserted by ISD. SR contended that all equiil
exchange was inappropriate because additional royalties should be accorded to
the wei.ght of its EDP patent portfolio including its 30A patents and applications,
over XCR's portfolio.

14.4.3 SR had called over 1000 of its patents to the attention of GE, including
the EXIAC patent and its 30A i)atpnts and anplieations, and sought royalties of
8.5 million dollars ; SR internally registered doubt as to the enforceability of the
EXIAC patent, and estimated that the "Chance of winning in court is less than
50%."



5853

14.4.4 When a one and one-lialf percent royalty was sought from RCA under the
EXIAC patent, S^R contended that the rest of its EDP portfolio, including its 30A
patents and applications, had "a gi-eater value than the ENIAC patent."

14.4.5 SR and IISD had contended in their negotiations with RCA that SR's
portfolio of other patents, including the 30A patents and such 30A applications as
BINAC and UXIVAC I, gave SR the "basic position in the systems area."

14.4.6 During negotiations with Honeywell, SR discussed ten patents from its
EDP portfolio, including the EM-1 regenerative memory in '827 patent and the
high-speed printer or '966 patent among the 30A group ; the fact of the pendency
of SR's UXIVAC application, within the 30A group, was also considered, SR con-
tended that its patent portfolio, exclusive of EXIAC but including the 30A patents
and applications, was at least as valuable as the entirety of Honeywell's EDP
patent rights.

14.5 Plaintiff claims also that they were used as leverage in obtaining the IB:\I
and BTL license agreements.

14.5.1 On August 21, li)56, IBM and Sperry Rund entered into an agreement in-
cluding :

1. a ci'oss-license under all of their respective EDP and tabulating ( TAB )
equipment patents and patent applications, including the EXIAC application and
the 30A patents and applications :

2. a settlement of all EDP and TAB Patent Office interferences ; and

3. an exchange of secret and proprietary EDP and TAB equipment know-how.

14.5.2 BTL and Western Electric entered into a cros.s-liceuse with SR dated
July 1, 1961.

14.6 Plaintiff claims that the fraudulent procurement of the portfolio was part
of an overall scheme to monopolize and restrain trade.

14.7 The emphasis in plaintiff's claim of inUrmities is on public use and on sale.

14.8 Other claims rest on derivation from Atauasoft", incomplete application at
time of execution, and omission of co-inventors.

14.9 As to publication, plaintiff relies on the First Draft Report (already
found to be a printed publication before the critical date) [Finding.s> 7.1-7.1.6
above], the EDVAC report of September 30, 1945, the ED VAC report of June 30,
1946, three lectures of the Moore School lecture series, and the report on the
UXIVAC.

[47] 14.10 I tind that as part of an action in antitrust that plaintiff has
standing to assert these claims, that no claims of infringement have been made by
defendants on the issued patents, and that defendants have made no demands for
royalties on the applications.

14.11 I find that the publications referred to above (In Finding 14.9) were
printed publications before the critical dates, that the claims as to public use and
on sale before the critical dates have been proved, that the claims a.«! to deriva-
tion from Atanasoff as to EM-1 have been proved, that the claims as to incom-
plete execution have been proved, and that the claim as to the omission of a
certain inventor has been proved.

14.11.1 Publications.

The June 30, 19J,6 EDVAC Report

14.11.1.1 The Progress Report on the EDVAC, dated June 30, 1946. (hereafter
called the EDVAC Report) became a printed publication more than one year
prior to June 10, 1949, the filing date of the EM-S application .

14.11.1.2 The Court notes that in prior litigation, SR swore in answer to an
interrogatory that the publication date of the EDVAC Report was June 30,
1946.

14.11.1.3 The EDVAC Report was completely declassified on or before Febru-
ary 13, 1947.

14.11.1.4 During the spi'ing of 1947, persons skilled in the computer art were
notified about the declassification of the EDVAC Report.

14.11.1.5 The decla.ssification of EDVAC information, such as the ED'\\\(J
Report, was widely publicized during March of 1947, including press reports in
the Xew York Times, Philadelphia Bulletin and Philadelphia Inquirer.

14.11.1.6 By the spring of 1947, Eckert and Mauchly were aware of the decla.s-
sification of the EDVAC Report.

14.11.1.7 By March 1948, the EDVAC Report had been widely distributed and
was available to all persons having an interest in the computing arts, as a result
of its unrestricted classification, its availability to persons attending the 1946
Moore School Lecture Series and its availability in the Library of Congress.

40-927 — 75 65



5854

14.11.1.8 By the end of June, 1948, copies of the ED VAC Report were available
in the Moore School Library and were being loaned by that libraiT to other in-
stitutions, including Johns Hopkins University ; a review of the EDVAC Report
was published in the January 1949 edition of Mathematical Tables and Others
Aids to Computation, and Eckert cited the EDVAC Report in an article (which
he eoauthored) published in the August 1949 edition of the Proceedings of the
I.R.E.

14.11.1.9 In 1957, SR's patent agent, English, who was as.signed to an EM-8
interference, withheld the existence of the EDVAC Report from the Patent Of-
fice "'since it might be used as an anticipating publication against our applica-
tion EM-S."'

14.11.1.10 The stated purpose of the EDVAC Report was to serve as a disclo-
sure to the Patent Oflice of possibly patentable ideas, and Eckert, Mauchly and
defendants" predecessors and lawyers used the EDVAC Report to this end ; in
October 1946. Eckert and :Mauchly asked for and received permission to inspect
the EDVAC Rei^ort for the purpose of preparing patent applications; the fol-
lowing applications or patents were derived from portions of the EDVAC Report
by Mr. Eltgroth, a patent attorney for Eckert-Mauchly Computer Corporation
team (EMCC) and Remington-Rand (RR), both predecessors of SR : EM-21, 23,
25, 26, 27, 28 and 29; two attorneys, Messrs. Eltgroth and Light, handled the
prosecution of each of these patent applications ; examples of the correspondence
of the EDVAC Report with the applications or patents are hereinafter found
and set forth.

14.11.1.11 Figure 1 of the EM-21 patent has a one-for-one relationship in all
significant I'espects to EDVAC Report drawing PY-0-105, described at pages 1-
1-2 through 1-1-5 of the EDVAC Report ; Figure 1 of the EM-23 patent shows
the same circuit, except for immaterial variations, as that shown and described
in EDVAC Report drawing PY-O-105 and pp. 1-1-2 through 1-1-5 ; Figures 5,
6 and 7 of the EM-25 patent show the same circuits as are shown and described
in EDVAC Report drawings Py-0-177, PY-(>-178 and PY-0-174 and at pp. 1-1-11
through 1-1-13; Figure 2 of the EM-26 patent shows the same circuit as that
shown and described in EDVAC Report drawing PY-0-177 and page 1-111 of
the EDVAC Report.

14.11.1.12 During August, 1952, the patent examiner assigned to the EM-27
application located the EDVAC Report unaided by defendants or their attorneys
and found that: "This reference [the EDVAC Report] discloses a circuit which
is manifestly identical in all respects to the instant alleged invention" ; the
Court concurs with the examiner's finding; Figure 1 of the EM-27 application
shows the same circuit as that shown and described in Report drawing PY-0-
108 and pages 1-1-27 through 1-1-29 of the EDVAC Report; RR's attorney
prosecuting EM-27, B. J. Light, also concurred, stating that the EDVAC Report
was "a publication identical to the disclosure" of EM-27.

14.11.1.13 Figure 1 of the EM-28 application illustrates the same circuit
shown and described in EDVAC Report drawing PY-0-181 and pages 1-1-29
through 1-1-30 of the EDVAC Report.

14.11.1.14 Figures 1 and 2 of the EM-29 case were based upon drawings PY-0-
101 and PY-O-102 of the EDVAC Report : when the EM-29 application was
drawn into an interference (No. 85,958), RR's patent attorney, Light, reported
to Eckert and Mauchly that the adder described in the EM-29 patent was de-
scribed in the EDVAC Report.

14.11.1.15 Yet RR's attorneys failed to state to the Patent Office at any time
during the EM-29 prosecution and interference that the EM-27 examiner had
found the EDVAC Report to be a printed publication more than one year prior
to the EM-29 filing date.

The Moore School Lecture Series

14.11.1.16 During July and August of 1946, a lecture series was held at the
Moore School entitled "Theory and Techniques for Design of Electronic Digital
Computers."

14.11.1.17 The lectures were i-ecorded and edited by the respective lecturers
for publication, and were published in four volumes between September 19. 1947
and June 30, 1948.

14.11.1.18 These published lectures (hereafter called the "Lecture Series")
were made available in the Moore School Libi-ary, and Lectures 1 through 48



Online LibraryUnited States. Congress. Senate. Committee on theThe Industrial reorganization act. Hearings, Ninety-third Congress, first session [-Ninety-fourth Congress, first session], on S. 1167 (Volume pt. 7) → online text (page 130 of 140)