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 137 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 137 of 140)
Font size
QR-code for this ebook


IBM and SR knew the Patent Office proceeding was essentially a sham.

16.3.3 IBM and SR were both aware that further appeals were available on
the public use issue and that the Patent Office decision not to hold a public use
hearing was not binding on the courts ; however IBM concluded it could claim



5891

to have leaned over backwards to appear to act in an ethical manner and that
nothing further could be expected of it.

16.3.4 IBM made no further formal attack on the validity of the EXIAC patent
application or patent despite its possession of evidence of and belief in the
invalidity.

[50] 16.4 Settlements are, of course, to be encouraged unless in the proce.ss
the antitrust laws are violated and the public interest harmed.

16.5 I find no violation of the antitrust laws and find lack of proof as to
injury.

16.5.1 On August 21, 1956, IBM and SR entered into an agreement providing
interalia, for a non-exclusive patent cross-license in the EDP field.

16.5.2 Also on August 21, 1956, IBM and SR entered into procedural agree-
ments to dispose of eleven outstanding Patent Office iuterference.s between patents
and patent applications of the two companies. Six of these interferences and
agreements involved the ENIAC patent application and five involved other SB
patent applications.

16.5.3 A Honeywell witness characterized these 1956 IBM-SR interference
procedure agreements as "very conventional," and Honeywell itself (through
Datamatic, which it controlled) later joined in certain of them.

16.5.4 Honeywell has not proven any fraud in connection with the patents
and patent applications involved in the eleven IBM-SR procedural interference
agreements.

16.5.5 Honeywell has not shown that the 1956 IBM-SR agreements on inter-
ference procedures alone constituted an unreasonable restraint on trade or
were otherwise improper.

16.5.6 Honeywell has proven no injury to its business or property caiised by
the 1956 IBM-SR agreements on interference procedure.

16.5.7 Honeywell has not shown that the 1956 IBM-SR agreements on inter-
ference procedures provide any grounds for declaring the ENIAC patent, or any
other SR patent, invalid or unenforceable.

16.5.8 There was suprcssion of the facts concerning the 1956 IBM-SR inter-
ference agreements.

17. 1861 Agreement and SR and BTL

17.1 The 1961 Agreement apparently resolved the issue of priority between
Williams and Mauchly and Eckert.

17.1.1 One of the Principal interferences (No. 85,809) in which the ENIAC
patent application was involved was with BTL's Williams patent.

17.1.2 That interference began in 1952 over the issue of priority of invention
and ended with BTL and SR litigation on the actually academic [to the parties]
question of public use of the ENIAC before the late Judge Archie Dawson of
the Southern District of New York.

17.1.3 On July 1, 1931, prior to the trial of that litigation. Western Electric
Company (acting for the Bell System, including BTL) and SR executed a
complete cross-license of all their patents and patent applications which in-
cluded all EDP patents and applications, of which ENIAC was one.

17.1.4 This Agreement had the effect of disarming BTL's attack on the va-
lidity of SR's ENIAC patent application and put BTL in a position in which it
had nothing to gain by es.tablishing the invalidity of the ENIAC patent
application.

17.1.5 Pursuant to the terms of 1961 SR-Western Electric Agreement, Western
Electric received, among many other things, a license under any patent to issue
on the ENIAC patent application.

17.1.6 Because SR owned the Eckert and/or Mauchly patent and application
rights (including ENIAC), it v.as able to and did gain access to the Bell Sys-
tem's very substantial patent and application portfolio : and this was done after
BTL had first been held to have priority in the Patent Oflice over what SR
later claimed to be the basic computer patent.

17.2 The issue of public use of the ENIAC was submitted to the District
Court and incorrectly decided.

17.2.1 See 18.1.1 through 18.1.5.

17.2.2 BTL never pursued the public use issue vigorouslv in the Southern
District of New York case and certainly not after the execution of the Jnlv, 1961
cross-lisense Agreement with SR.

17.2.3 At the time when the question of the public use was submitted to Judge
Dawson in the Fall of 1961, BTL had no true interest in establishing public use
as a defense to the validity or issuance of the ENIAC patent.



5892

17.2.4 On December 18, 1961 in the course of the BTL case, SR was advisecl
that the Los Alamos Laboratory of the A. E.G. was prepared to furnish an official
affidavit that the 1945-46 ENIAC work before the critical date was iiseful, im-
..portant, solved real problems and was relied upon ; SR declined the affidavit as
-tiot helpful, did not reveal the A. E.G. offer to BTL or the court and contended

to a contrary effect on the record to the court.

17.2.5 BTL's internal memoranda in connection with the lawsuit prove that it
•desired to complete the pviblic use litigation in the Southern District of New

York at the least expense and with the least possible amount of effort ; SR was
vof a like mind as it never mounted a truly effective effort.

17.2. i) BTL and SR submitted their evidence in perfunctory and conclusory
affidavits and deposition transcripts.

17.2.7 Most of the volume of the evidence before Judge Dawson did not bear
on tlie public use issue at all but on the priority issue which had previously been
eliminated from the case.

17.2.S On September 6, 1962, Judge Dawson found that the ENIAG machine
had not been in public use more than one year prior to filing the ENIAC appli-
cation and directed the Gommissioner of Patents to issue a patent on the ENIAC
application.

17.2.9 After this decision, BTL took a "for-the-record" appeal to the Court
of Appeals for the Second Circuit which dismissed the appeal because the case
Iliad l)een moot all along by the 1961 execution of the cross-license between West-
ern Electric and SR.

17.2.10 Thus, had the case been correctly decided by the District Court, the
ENIAC patent would not have issued and the EDP industry would not have
been threatened with its burden or sanction.

17.3 I find no violation of the antitrust laws and find lack of proof as to
injury.

17.3.1 Honeywell has not shown that it was injured in its business or property
by reason of the 1961 SR- Western Electric Agreement.

17.3.2 Honeywell has not proven that the 1961 SR-Western Electric Agree-
ment was an unreasonable restraint on trade.

17.3.3 Honeywell has not proven that the 1961 SR-Western Electric Agree-
ment, of itself, violated the Sherman Act or injured Honeywell in its business
or property.

£51] 17.3.4 Nonexclusive patent cross-licensing in itself may be proper.

18. 1965 Agreement of SR and IBM

18.1 Following the issuance of the ENIAC patent, the question of whether
further payments were required under the 1956 Agreement required resolution.

I'^.l.l The 1956 Agreement required that IBM pav minimum rovalties to SR of
$10.(XtO,000 in eight $1,250,000 annual installments; an additional royalty of
1 percent of IBM's EDP manufacturing costs (for infringing equipment) from
■October 1. 1956 to September 30. 1964. after crediting the $10,000,000 prepaid,
was due if the ENIAG patent should issue prior to December 31, 1964.

18.1.2 After the ENIAC patent issued on Fel)ruary 4. 1964, SR served notice
on IBM claiming additional royalties due under the 1956 Agreement, and a.sked
IBM to prepare a statement (certified by independent auditors) of additional
royalties due as provided by the 1956 Agreement.

IS.1.2.1 Subsequent to the issuance of the ENIAC patent in 3964. and SR
•engaged in a dispute concerning whether IBM owned SR additional royalties
under a provision of the 1956 IBAI-SR agreement.

is.1.3 On April 24. 1964. J. Blrkenstock of IBM furnished SR with a certified
"Statement of Additional Royalty Due"' which concluded that since IBlM's total
manufacturing costs from the period October 1, 1956 for all EDP equipment,
infringing or not. did not exceed $1,000,000,000 (making 1% of that less than
the $10,000,000 fixed royalty). IBM owed no further royalties other than the
remaining fixed royalty payment.

15.1.4 On May 1. 1964. c' INIcTiernan of SR informed IBM that in SR's opinion
IBM had not fulfilled its obligation under the 1956 Agreement as regards fur-
nishing a royalty statement : accordingly. SR requested a formal, exact, detailed
presentation which would enable it to certify IBM's accounting.

18.1.5 During 1964 and 1965. IB]\I and SR continued to negotiate in an attempt
to arrive at a satisfactor.v resolution to their controversy: the main points at
apparent issue, aside from the validity of the ENIAG patent, were the inter-
pretation of IBM's obligations under the 1956 Agreement, the scope of the ENIAO



589'3

patent claims as they related to IBM equiiiment (which SR claimed covered some
TAB machines) and the accounting measures used by IBM to arrive at its
manufacturing costs for infringing equipment.

18.2 In 1904 or 1965 IBM was still contending that the ENIAC was invalid
for a num))er of reasons.

18.2.1 At a December 15, 1964 meeting between IB:M and SR patent represen-
tatives, Charles Walker, outside patent counsel of IBM, contended that if IB:\I
should prevail on any one of four stated positions in a court contest, the ENIAO
patent would be declared invalid ; the four contentions were : '

.1 The EXIAC patent fails with respect to public use ;

.2 Even if SR should persuade the Court that use up to a certain time was
experimental, the public use was nevertheless more than a year before claims
to the system were first introduced ;

.3 Although IBM is a licensee, it has the right to construe a patent and the
background against which it should be read ; and

.4 Certain claims from the EXIAC patent read ou the Phelps patent. [See 6.1
through 6.1.7]

18.2.2 At a February 5, 1965 meeting of representatives of SR and IB:M, IBM
outside coun.sel again contended that the ENIAC machine had been in public
use and that IBM retained the right under the 1956 Agreement to raise the
public use defense against the EXIAC patent: IBM also correctly contended
that the public interest would be to have an invalid patent declared invalid,
and that .ludge Dawson in the SR v. BTL case made a clear error in finding
experimental instead of public use.

18.2.3 At an August 11, 1965 meeting of SR and IB:\I representatives, IBM
again contended that the EXIAC patent was invalid by reason of public use
and the Phelps patent and reiterated the late claiming defense [See 10.1] and
IBM's position that the scope of the ENIAC patent should not include input
and output units: at this meeting IBM gave SR a report, prepared as a result
of earlier meetings with SR, detailing these contentious.

18.3 It had. however, in 1959 or 1960 given up its strongest argument on iQ-
validitv liecause of public use.

18.3.1 See 16.2.1 through 16.2.10 and 16.3.1 through 16.3.4.

18.4 The dispute resolved into an accounting issue and IBM agreed to furthei
payment of $1,100,000.

18.4.1 See 18 through 18.1.5 and 18.5 through 18.5.4.

18.5 This payment was conditioned liy the 1956 agreement on the issuance of
the EX^IAC apVilication. and it appears that much of the total $11,100,000 i)aid
by IBM to SR represented EXIAC royalties.

18.5.1 The 19.56 Agreement contains a specific definition of the EXIAC api)lica-
tion and also provides sjiecifically and separately for an additional royalt.v to
1)0 paid in the event EX'IAC finally ripened into a patent: the EXIAC applica-
tion was a .significant is.sue. was negotiated separately from the other issue.s,
was eventually settled separately from the other issues and was the sultject
of a specific provision in the 1956 Agreement and again in the 1965 renewal
or extension agreement.

18.5.2 The 19.56 Agreement called for IBM to pay minimum royalties to SK
of .$10,000,000 plus an additional royalty of 1% of IBM's EDP manufacturing cost.<?
ffor infringing equipment) from October 1, 1956 to September .30. 1964. after
crediting the $10,000,000, if the ENIAC patent should issue prior to December 31,
1964.

18.5.3 lender the 1956 Agreement no additional royalties were payable by IBM
if the ENIAC patent did not issue or was invalidated or held unenforceable.

18.5.4 On X^ovember 15. 1965. SR and IB]\I entered into a complete cross-
license on all information handling systems patents and patent applications filed
between that date and November 15. 1970 and agreed that IBM should pay SR
.'';i .100.000 to settle the dispute over additional royalty due because of the issuance
of the ENIAC patent ; this accomplished "peaceful co-existence" between the co-
conspirators for five more years, each being free to infringe each others' patents
without risk.

18.5.4.1 After more than a year's negotiation, IBM and SR on November 15. 1065
reached an agreement resolving, among other things, the dispute concerning the
19.56 IRM-SR Agreement.

18.5.4.2 Under the provisions of their 1965 Agreement IBM and SR exchansrec^
paid-up nonexclusive cross-licenses in the field of "information handling" under
patents and future patent applications filed prior to N^ovember 15, 1070. Moreover,



5894

IBM and SR released each other from all claims under the 1956 agreement, and
IBM paid SR $1,100,000 upon the execution of the agreement.

18 6 What is difficult to measure is what part in the 1956 agreement the fact
of dismissal of the SR antitrust suit against IBM and other factors played.

18 7 IBM was then, and should be now, apprehensive about antitrust lawsuits.

IsIt.I See 15.26.4 through 15.26.30. .

18.8 The fact is that plaintiff and the newcomers in the EDP market in 1956,
including Burroughs, NCR, RCA and others, should have in 1956 or later sued
SR and IBM which then controlled about 95% of the EDP market.

18.8.1 Had such a suit or suits been commenced by filing prior to May 26, 1963,
judicial notice may be taken that they could not have resulted in a trial in less
than two (2) to four (4) years nor in a decision in less than four (4) to six (6)
years after commencement.

18.8.2 Had such a suit or suits been commenced, judicial notice may be taken
that evidence of damage in any one year which flowed from non-access to the
shared technology would not have been available until after the end of at
least that year or one prior thereto and projections as to tlie future from depriva-
tion in the past would have been subject to vigorous objection as speculative
and conjectural even though broad-based injunctive relief could have been
awarded as protection for years in the future.

18.8.3 To illustrate, had the lawsuit been filed in 1957 or 1958 and come to trial
and decision in 1960 (a most unlikely event), evidence of the technology-shared
damage could have been offered for the years 1957, 1958 and 1959 but estimates
of any "pour-over" until 1961 or later would have been met with the objection of
guesswork even though the plaintiff (Honeywell for instance) was assumed not
only to have been granted free access by discovery to see but also the right to
use the shared technology by 1960 as a result of a final decree and no appeal.

18.9 I find that regardless of whatever further accommodation of the interests of
SR and IBM resulted from the 1965 Agreement, that the agreement did not
violate the antitrust laws.

18.9.1 Honeywell has not proven that the 1965 IBM-SR Agreement was an \m-
reasonable restraint on trade.

18.9.2 Nonexclusive patent cross licensing in itself may be proper.

18.9.3 Honeywell has not proven that the 1965 IBM-SR Agreement violated
the Sherman Act or injured Honeywell in its business or property.

18.10 Plaintiff has failed to prove injury.

18.10.1 Honeywell has not proven that the 1965 IBM-SR Agreement, in itself,
had any effect on Honeywell.

18.10.2 Honeywell has not shown that it was injured in its business or property
by reason of the 1965 IBM-SR Agreement.

19. Discriminatory Licensing

19.1 Plaintiff claims a violation of Section 2 because two competitors of SR
have been treated differently.

19.2 The SR and IBM agreements of 1956 and 1965 resulted in payments by
IBM to SR of $11,100,000.

19.2.1 See 18.5.1 through 18.5.4.

19.2.2 One percent of IBM's EDP equipment manufacturing cost for the
period 1957-64 was agreed by SR and IBM to be $11,100,000.

19.3 The ENIAO application was an important factor in the considerations
exchanged.

19.3.1 See 15.24, 15.24.2. 15.24.13. 15.24.21, 15.24.25, 15.24.30, 16.2.2., 18.1.2, 18.2.1
through 18.2.3 and 18.5.1 through 18.5.4.

19.3.2 During the course of the 1956 negotiations between IBM and SR. IBM
insisted that any overall settlement include a license to IBM under SR's ENIAC
patent application.

19.3.3 The 1956 SR-IBM Agreement as consummated included a cross-license on
EDP patents and applications filed prior to October 1, 1956. The ENIAC patent
application was included in this group.

19.3.4 Under the 1956 IBM-SR Agreement, SR received, among other things,
licenses under IB:M's tab patents and applications, and also received technical
information on. all IBM tabulating machines announced or released to production
a.s of October 1, 1956.

19.3.5 SR officials hoped that the 1956 IBM patent licenses and technical infor-
mation would include patent licenses and technical information on a proposed



5895

IBM World Wide Accounting Machine (WWAM), and it had been estimated that
the rights to this WWAM tabulating machine would be worth $20,000,000 to SR.

19.3.6 SR also received pursuant to the 1956 IBM-SR Agreement patent li-
censes under IBM's EDP patents and applications, as well as EDP technical
information, as of October 1, 1956.

19.3.7 Among the patent licenses each party received pursuant to the 1956
IBM-SR Agreement were licenses xmder the patents or applications that were
involved in eleven interferences with SR patents and applications.

19.3.8 Each party also received from the other, pursuant to the 1956 IBM-SR
Agreement, a release for past patent infringement, and IBM agreed to dismiss
its counterclaim against SR for infringement of 35 patents.

19.3.9. SR also received, pursuant to the 1956 IBM-SR Agreement $10,000,000
from IBM, plus a covenant by IBM to pay an additional royalty under the
ENIAC patent, if a patent issued on the ENIAO patent application prior to Janu-
ary 1, 1965. This additional royalty was to be 1% of the maniifactunug cost of
each EDP machine embodying any invention covered by the claims of the ENIAO
patent manufactured by IBM within the United States between October 1, 1956
and October 1, 1964 — after deducting the prepaid $10 million as a credit.

19.3.10 Frost, SR's general counsel and its chief negotiator, had asked his pat-
ent people in 1956 to give him an estimate of the values of the respective patent
positions of IBM and SR but they were unable to do so. Frost concluded that the
only premise he could adopt was that the patents were equally balanced in value.

19.3.11 As to the entire noncash package SR received from IBM in 1956, Frost
had been advised by UNIVAC personnel that it was worth from $30,000,000 to
$40,000,000.

19.3.12 Shortly after the ENIAC patent issued on February 4, 19M, Birken-
stock, IBM's chief negotiator, wrote to T. J. Watson, Jr., IBM's chief executive :

"The issuance of the 'ENIAC patent to Sperry Rand after ten years of litiga-
tion makes our $10 million settlement look many times better than we figured
it to be in 1956."

19.3.13 After issuance of the ENIAC patent, IBM and SR became involved in
a dispute concerning whether IBM owed additional royalties under the 1% pro-
visions of the 1956 IBM-SR Agreement.

19.3.14 After more than a year's negotiation, IBM and SR on November 15,
1965 reached agreement.

19.3.15 Under the provisions of the 1965 IBM-SR Agreement, IBM and SR
exchanged paid-up nonexclusive cross-licenses in the field of "information
handling" under patents and patent applications filed prior to November 15, 1970.
Under the broad definition of "information handling" in the 1965 Agreement, SR
received licenses under IBM's electric typewriter patents as well as under EDP
and tab patents. IBM and SR also released each other from all claims under the
1956 IBM-SR Agreement, and IBM paid SR $1,100,000 upon the execution of the
agreement.

19.3.16 The ENIAC patent issued on February 4, 1964.

19.3.17 On April 8, 1964, SR assigned its rights in the ENIAC patent to ISD, a
wholly-owned subsidiary of SR.

19.3.18 In June of 1964, ISD sent out notices of infringement of the ENIAC
patent, with an offer to license "on reasonable terms" to various companies, in-
cluding Honeywell. ISD offered to license these alleged infringers at 1%% of the
selling price of equipment covered by the ENIAC patent. Although ISD contin-
ued correspondence and discu.ssions with various alleged infringers, no license
agreements were concluded.

19.3.19 Some alleged infringers of the ENIAC patents, including Honeywell,
insisted on including SR and its patents in the discussions.

19.4 After the issuance of the ENIAC patent, the defendants initially de-
manded $250,000,000 from plaintiff for royalties.

19.4.1 In a meeting on July 21, 1965, SR informed Honeywell that the ENIAC
license royalty rate was iy2% of the net selling price of each Honeywell EDP
system sold or leased with no paid up limit ; SR later estimated that Honeywell's
liabilitv to it under this proposal for the 1904r-1981 life of the ENIAC patent
would be in excess of $200,000,000.

19.4.1.1 SR and ISD communicated with Honeywell in 1965 and 1966, but
negotiations dragged on into 1967.

19.4.2 At a January 26, 1967 meeting, John Dority of SR stated that he had
computed a potential liability of Honeywell approximating $250,000,000 using
the lyo percent figure as applied to past and expected future shipments ; Honey-



5896

well indicated to SR representatives tliat tliis was quite disproiportionate to the
$10,000,000 asked of IBM under the 1956 cross-license Agreement with IBM.

rj.4.2.1 At the urging of John Dority, UNIVAC patent counsel, a meeting
was held on January 2(5, 1967. At that meeting, SR indicated to Honeywell that
SR considered its EDP patent portfolio (excluding ISD's ENIAC patent) as
more valuable than Honeywell's and that disparity alone should warrant the
payment of a l-i/4% royalty by Honeywell.

19.4.2.2 Nevertheless, SR and ISD, in the cynically expressed hope of promptly
getting the matter resolved, told Honeywell at their January 26. 1907 meetint;
that, in the context of a cross-license including the ENIAC patent, they would
accept a lump-sum settlement of $1,250,000 annually for 15 years, plus a negoti-
ated settlement for past infringement.

19.4.2.3 Honeywell neither accepted the 1067 SR-ISD offer nor made a counter-
ofifer. No license agreement was ever concluded and Honeywell has never paid
any i-ovalties to SR or ISD.

19.5 This was later reduced to about $20,000,000.

19.5.1 During the January 1967 meeting. SR indicated to Honeywell that it
would accept, in the context of an EDP patent cross-license between Honeywell
and SR and ISD, a lump sum settlement with Honeywell paying SR $1.250,(X>0
annually for fifteen years, a total of $18,750,000, plus settlement for past lialjility.

19.6 Defendants demanded royalties from the other members of the EDP
industry.

19.6.1 After issuance of the ENIAC patent on February 4. 1964, defendants
systematically demanded royalties for a license under the ENIAC patent at a
rate of I-I/2 percent of the selling price of the affected equipment from all EDP
industry members, including, among others, GE, Burroughs, RCA, NCR, CDC.
Philco-Ford and numerous manufacturers of peripheral equipment: a royalty of
1-1/2 percent of the selling price (in dollars) of equipment sold or leased in the
years 1964-1967 by the main-frame manufacturers appears in Column A below
and that four year total royalty multiplied by 4.25 produces a 17 j'ear conserva-
tively estimated royalty which appears in Column B as follows :



17-yr.

4 yr. established

royalty royalty

General Electric . $5,848,406 $24,855,726

RCA - 6,619,317 28,132,097

CDC 9,857,210 41,893,142

NCR 6,364,531 27,049.257

Philco-Ford 420,840 1,78«',, 570

Burroughs- 6,044,316 25,688,343

Total 35,154,620 149,407,135

19.6.1.1 Liunp-sum settlements in the context of cross-licenses with SR and ISD
were offered to tho.se alleged infringers with whom negotiations continued,
including RCA and Honeywell, but no license agreements were concluded.



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 137 of 140)