Decision of August 22, 2002,
United
States District Court - Southern District of New York
MEMORANDUM
AND ORDER GRANTING SUMMARY JUDGMENT
McMahon,
J.
Plaintiff
British Telecommunications ("BT") asserts that Defendant Prodigy Communications Corp. ("Prodigy"), through its
business activities as an Internet Service Provider
("ISP"), directly infringes claims 3, 5, 6, and 7 (the "Asserted
Claims") of U.S. Patent No. 4,873,662 (the
"Sargent Patent" or "’662 Patent"). BT also alleges that
Prodigy induces and contributes to infringement by
Prodigy subscribers who infringe the Sargent patent by accessing the
Internet through the Prodigy service.
The
Court has already construed the claims of the patent in its Markman Opinion.
British Telecommunications PLC v. Prodigy
Communications Corp., 189 F. Supp. 2d 101 (S.D.N.Y. 2002)
("Markman Op."). Prodigy now moves for summary judgment of
non-infringement under Rule 56 of the Federal Rules
of Civil Procedure. For the reasons stated below, I
find that as a matter of law, no jury could find that Prodigy infringes
the Sargent patent, nor that Prodigy contributes to infringement of the Sargent
patent, nor actively induces others to infringe
that patent. I therefore grant Prodigy’s motion for summary
judgment.
Summary
Judgment Standard
A
party is entitled to summary judgment when there is no "genuine issue of
material fact" and the undisputed facts
warrant judgment for the moving party as a matter of law. Fed. R. Civ. P.
56 (c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (1986). In addressing a
motion for summary judgment, "the court must
view the evidence in the light most favorable to the party against
whom summary judgment is sought and must draw all reasonable inferences in [its] favor." Matsushita Elec. Indus. Co. Ltd. v. Zenith
Radio Corp., 475 U.S. 574, 587 (1986). Whether any
disputed issue of fact exists is for the Court to determine. Balderman v. United States Veterans Admin., 870 F. 2d 57, 60 (2d Cir. 1989).
The moving party has the initial burden of
demonstrating the absence of a disputed issue of material fact. Celotex v.
Catrett, 477 U.S. 317, 323 (1986). Once such a
showing has been made, the non-moving party must present "specific
facts showing that there is a genuine issue for trial." Fed. R. Civ. P.
56(e). The party opposing summary judgment "may
not rely on conclusory allegations or unsubstantiated speculation."
Scotto v. Almenas, 143 F. 3d 105, 114 (2d Cir. 1998). Moreover, not every disputed factual issue is material in light of the
substantive law that governs the case. "Only disputes
over facts that might affect the outcome of the suit under the governing law
will properly preclude summary judgment."
Anderson, 477 U.S. at 248.
As
a general rule, infringement is a question of fact. SRI Int’l v. Matsushita
Elec. Corp. of Am., 775 F.2d 1107, 1116 (Fed. Cir.
1985). However, summary judgment is appropriate if the court,
drawing all reasonable inferences in favor of the patentee, concludes that no
reasonable jury could find infringement.
Warner-Jenkinson Co., Inc. v. Hilton Davis Chem. Co., 520 U.S. 17,
39 n. 8 (1997).
RELEVANT
LEGAL PRINCIPLES
Infringement
Determining
whether a device infringes another’s patent is a two step process. First, the Court construes the claims to determine their scope and
meaning. The Court did this in its Markman Opinion
dated March 13, 2002. British Telecommunications PLC, 189 F. Supp. 2d 101.
The next step is to compare the allegedly infringing device against the claims
as construed to determine whether the device
embodies every limitation of the claims. Cybor Corp. v. FAS Techs.,
Inc., 138 F.3d 1448, 1454 (Fed. Cir. 1998) (en banc).
A
device literally infringes a patent, when it "embodies every limitation of
the asserted claims." IMS Tech, Inc. v. Haas
Automation, Inc., 206 F.3d 1422, 1429 (Fed. Cir. 2000); Laitram
Corp. v. Rexnord, Inc., 939 F.2d 1533, 1535 (Fed. Cir. 1991). "Literal
infringement of a claim exists when each of the
claim limitations ‘reads on,’ or in other words is found in, the accused
device." Allen Engineering Corp. v. Bartell Indus., Inc., No. 01-1238, 2002
WL 1765989 (Fed. Cir. Aug. 1, 2002) (citing Baxter
Healthcare Corp. v. Spectramed, Inc., 49 F.3d.3 1575,
1583 (Fed. Cir. 1995); Amhil Enters. Ltd. v. Wawa, Inc., 81 F.3d 1554, 1562 (Fed.
Cir. 1996)).
Even
if a device does not literally infringe a patent, it may still infringe under
the doctrine of equivalents. Infringement under the
doctrine of equivalents applies when there are insubstantial
differences between the claimed invention and the accused product. If an
allegedly infringing device performs substantially
the same function as the patented invention, in substantially
the same way, to yield substantially the same result, it may infringe under the doctrine of equivalents. Warner-Jenkinson Co., 520 U.S. at
21; see also Graver Tank & Mfg. Co., v. Linde
Air Prods., Co., 339 U.S. 605, 608 (1950); Atlas Powder Co. v. E.I. duPont de Nemours & Co., 750 F.2d 1569, 1579 (Fed. Cir. 1984).
The doctrine of equivalents focuses on the "role
played by each element in the context of the specific patent claim."
Warner-Jenkinson Co., 520 U.S. at 40.
Whether
infringement is established literally or under the doctrine of equivalents,
every element, or its substantial equivalent, set
forth in claim must be found in the product in question. See
Penwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931, 935 (Fed. Cir. 1987). No
structural or functional limitation in the claim
may be ignored. Warner-Jenkinson Co., 520 U.S. at 25. The
Supreme Court recently clarified the law regarding the doctrine of equivalents
in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki
Co., Ltd., 122 S. Ct. 1831 (2002). In Festo, the
patentee held a patent on a magnetically coupled, rodless cylinder used in machinery. Id. at 1835. As often occurs, the patentee
amended the claims of the patent during the
prosecution. Id. The amended claims stated that the invention contained a pair
of sealing rings, each with a lip on one side to
prevent impurities from getting onto the piston assembly. The
amendment also added that the outer shell of the device would be made of a
magnetizable material. Id.
SMC
began making a similar device that differed from Festo’s patent in two ways:
it had one sealing ring with a two-way lip instead
of two sealing rings with a one-way lip, and the outer shell
of SMC’s device was made of a non-magnetizable material. Festo claimed that
SMC infringed its patent under the doctrine of
equivalents.
SMC
asserted that prosecution history estoppel prevented Festo from claiming that equivalents infringed its patent, because the additional
recitations in the amendment narrowed the claim to
embrace only devices that included those elements. Prosecution history estoppel requires that patent claims be interpreted in light of the
proceedings before the Patent and Trademark Office
leading to the issuance of a patent, and precludes the patentee from claiming any interpretations of its patent that were "cancelled
or rejected" during the course of the prosecution
of the patent. Id. at 1838. If a patentee chooses to amend his claims in light
of a patent examiner’s rejection, then the
patented invention does not encompass the original claim that
was rejected by the examiner. Id. (citing Goodyear Dental Vulcanite Co. v.
Davis, 120 U.S. 222, 228 (1880); Wang Labs., Inc.
v. Mitsubishi Elec. Am., Inc., 103 F.3d 1571, 1577-78 (Fed..4 Cir.
1997)).
Festo
argued that prosecution history estoppel applied only when narrowing amendments were made for substantial reasons relating to
patentability – for example, when the applicant made
the amendment to avoid prior art, but not when an amendment was made for the
sake of form. Id. at 1839. In a controversial
decision, the Federal Circuit disagreed, finding that any narrowing
amendments made during the patent prosecution foreclosed the patentee from
arguing that equivalents infringed elements of
their patent, even if the amendment was not made for patentability
reasons. Id. at 1835.
The
Supreme Court disagreed, and held that narrowing amendments made during the prosecution process did not automatically prevent the
patentee from claiming equivalents. However, it
ruled that narrowing amendments create a rebuttable presumption that they were made in order to obtain the patent. The patentee may then
overcome this presumption, but only by
demonstrating that the amendments were made for some reason. If he succeeds in
rebutting the presumption, there is no bar to his
invoking the doctrine of equivalents; if he fails, the doctrine
of equivalents cannot be applied to that claim.
The
Supreme Court's Festo decision re-affirmed the rule announced in
Warner-Jenkinson that prosecution history estoppel
applies only where claims have been avoided for a "substantial reason
related to patentability," such as to avoid prior art or to address a
specific concern such as obviousness. Festo, 122 S.
Ct. at 1839 (quoting Warner-Jenkinson, 520 U.S. at 30-32). Under its ruling,
whether prosecution history bars the patentee from claiming equivalents is to be
examined on a case-by-case basis:
By
amending the application, the inventor is deemed to concede that the patent does
not extend as far as the original claim. It does
not follow, however, that the amended claim becomes
so perfect in its description that no one could devise an equivalent. After amendment, as before, language remains an imperfect fit
for invention. The narrowing amendment may
demonstrate what the claim is not; but it may still fail to capture precisely
what the claim is. There is no reason why a narrowing amendment should be deemed to relinquish equivalents unforeseeable at the time
of the amendment and beyond a fair interpretation
of what was surrendered. Nor is there any call to foreclose claims of equivalence
for aspects of the invention that have only a peripheral relation to the reason the amendment was submitted. The amendment does not show
that the inventor suddenly
had
more foresight in the drafting of claims than an inventor whose application was granted without amendments having been submitted. It shows
only that he was familiar with the broader text and
the difference between the two. As a result, there is no more reason
for holding the patentee to the literal terms of an amended claim that there is
for abolishing the doctrine of equivalents
altogether and holding every patentee to the literal terms
of the patent. Id. at 1840-4l..5
The
Court noted that Warner-Jenkinson struck the appropriate balance between
flexibility and certainty when it placed on the
patentee the burden of proving that the amendment was not made
for the purposes of patentability. If the patentee cannot provide this
explanation, the court will presume that the
amendment was for a substantial reason related to patentability, and prosecution
history will preclude any reliance on equivalents for the amended element. Id.
at 1841-42. So strong is the presumption that,
"[W]hen the court is unable to determine the purpose
underlying a narrowing amendment – and hence a rationale for limiting the
estoppel to the surrender of particular equivalents
– the court should presume that the patentee surrendered all
subject matter between the broader and the narrower language." Id. at 1842.
Contributory
Infringement
Under
35 U.S.C. § 271(c), one who sells a component of a patented invention or a component used in a patented process is liable for
infringement as a contributor if: (1) the component
is a material part of the patented invention, (2) the seller knows the component
is specifically made or adapted for use in an
infringement of the patented invention, and (3) the component
is not a staple article of commerce suitable for substantial noninfringing use. Knowledge that one’s activity causes infringement is
necessary to establish contributory infringement.
Hewlett-Packard Co. v. Bausch & Lomb Inc., 909 F.2d 1464, 1469-70 (Fed. Cir. 1990). Absent proof of direct infringement, there can be
no contributory infringement or inducement of
infringement. Met-Coil Sys. Corp. v. Korners Unltd., Inc., 803 F.2d 684, 687 (Fed. Cir. 1986).
Active
Inducement
A
claim of active inducement under 35 U.S.C. § 271(b) requires the plaintiff to
prove that the accused infringer knowingly aided
and abetted another’s direct infringement of the patent. Rodime
PLC v. Seagate Tech., Inc., 174 F.3d 1294, 1306 (Fed. Cir. 1999). "Proof of
actual intent to cause the acts which constitute
the infringement is a necessary prerequisite to finding active
inducement." Hewlett-Packard Co., 909 F.2d at 1469. "It must be
established that the defendant possessed specific
intent to encourage another’s infringement and not merely that the defendant
had knowledge of the acts alleged to constitute inducement." Manville Sales
Corp. v. Paramount Sys., Inc., 917 F.2d 544, 553 (Fed.
Cir. 1990). The plaintiff must demonstrate that the
alleged infringer induced the infringing acts and that "he knew or should
have known his actions would induce actual
infringements." Id..6
The
Sargent Patent
The
Sargent Patent describes a system in which multiple users, located at remote terminals, can access data stored at a central computer. (Markman
Op. at 2.) The data is received by the remote
terminals via the telephone lines. The information
accessed by the remote terminals is stored on the central computer in the form of blocks, each block identified by a complete
address. The central computer uses the complete
address to retrieve the block identified by that address from storage when a
user requests it. Each block stored on the central
computer is comprised of two parts: a first portion, which
contains textual and graphical data for display (a display page), and a second
portion, not intended for display, which contains
the complete addresses of other blocks of information that are
related to the current display page. (Id.) The two portions of information are
stored together – indeed, next to each other –
yet they can be separated from each other. (Id. at 14-15.) For a given
block of information, the displayed (first) portion references other blocks of
information, while blocks in the second portion of
the block of information contain the complete addresses. (Id.
at 14.)
In
the asserted claims, the entire block is transmitted to the remote terminal
where the first portion is displayed and the second
portion is stored in the local memory. The display page includes
abbreviated addresses for particular blocks of information that can be accessed
from the central computer. When the user selects
one of the displayed abbreviated addresses from the first (displayed)
portion of the block, the terminal accesses the second portion of the block from
its memory to determine the corresponding complete
address. That complete address is then sent via the
modem to the central computer to obtain the next desired block of information.
(Id.)
BT’S
ARGUMENTS IN MACRO
BT
argues that Prodigy infringes the Sargent patent through its business activities
as an Internet Service Provider. BT contends that
Prodigy’s web servers provide access to information in
a manner that literally infringes the Sargent patent.
BT
also alleges that the Internet infringes the Sargent patent and that Prodigy
facilitates infringement by its subscribers by
providing them with access to the Internet. BT contends that Prodigy
contributorily infringes or actively induces the infringement of the Sargent
patent by providing the necessary software and
encouraging its subscribers to access pages of information from
Web servers maintained by third parties. Therefore, BT argues, even if Prodigy’s
servers do not infringe the Sargent patent as a
matter of law, summary judgment should be denied because
Prodigy infringes the ’662 patent by making and using infringing remote
terminals.
UNDISPUTED
FACTS
The
Internet
The
Internet is a network of computer networks that links millions of public and
private computers to form the largest computer
network in the world. (Prodigy’s Local Rule 56.1 Statement
¶ 1 ("Prodigy’s Undisputed Facts"); BT’s Local Rule 56.1
Statement ¶ 1 ("BT’s Undisputed Facts");
Wah Report ¶¶ 22, 23; Clarke Decl. ¶ 6.) It enables millions of people to obtain and share information electronically. The World
Wide Web is a collection of Web pages that are
linked together on the Internet. (Wah Report at ¶ 22.) The
Internet relies on several mechanisms to make Web pages available to a vast audience. (Wah Report ¶ 25; Clark Decl. ¶ 6.) These
mechanisms include: (1) a uniform naming scheme for
locating resources on the World Wide Web, such as the Uniform Resources Locator Standard ("URL"); (2) protocols for access to
named resources over the World Wide Web, such as
the Hypertext Transfer Protocol ("HTTP"); and (3) HyperText Markup
Language ("HTML"). (Wah Report ¶ 25;
Clark Decl. ¶ 6.) HTML is the language in which
Web pages are formatted. (Wah Report ¶ 24 (citing A Beginner’s
Guide to HTML, NCSA).) This language is a kind of "publishing mother tongue
that all computers may potentially understand."
(Id. ¶ 22). HTML allows authors to mark up a document
by representing structural, presentational, and semantic information adjacent to
the content. (Id. ¶ 23.) In HTML, a mark up
element is called a tag. These tags include, for example,
paragraph separators, text treatments (e.g. bold, italic, etc.), and hyperlinks.
(Id.) HTML files may or may not include
instructions, including a programming language called JavaScript.
(BT’s Undisputed Facts ¶ 15; Wah Report ¶ 169.)
HTML
files are stored in computers called Web servers. (Wah Report ¶ 24.) A Web server "serves" up Web pages to Web browsers
upon request. (Id.) A user looking to access an HTML
file stored in a Web server requires a personal computer ("PC") with
software called a Web browser. (Id.) A
hyperlink points to the URL for a Web page. (Id. ¶ 27.) Hyperlinks often take
the form of a colored text (such as a URL
description), logo or image that is displayed on the screen. When
a user clicks on the highlighted word or icon, she is sent to the URL requested,
or receives more information about what she clicked
on. URLs are identifiers that are used to locate
resources on the Internet, such as HTML files stored
on a Web server. (BT’s Undisputed Facts ¶ 3; Wah Report ¶ 26; Clarke Decl.
¶ 7.) A URL is a compact representation of the
location and access method for a resource that is available
through the Internet, generally consisting of three pieces of information: (1)
the access method, (2) the name of the server where
the Web page is stored, and (3) the name of the page itself
within the server, which may be depicted as a path. (Wah Report ¶¶ 26, 28.) For example, in the URL http://www.prodigy.net/a/b/c.html:
•
"http" is the access method. This identifies the communications
service. These include hypertext transfer protocol
("http"), file transfer protocol ("ftp"), gopher, mail,
etc.
•
www.prodigy.net identifies
the Web server. Names of this form are called DNS names.
The Web system makes use of the Internet Domain Name Service, which translates
the name of a machine on the network into an Internet address
•
a/b/c.html identifies the desired page within the server (Wah
Report ¶ 29; Clarke Decl. ¶ 7.)
When
a user requests an HTML file, the browser sends a request to a Web server via
the telephone lines. (Wah Report ¶ 32.) The Web
server retrieves the file and sends it back to the user’s
browser. (Id.) The browser then reads and interprets the HTML file and displays
the Web page on the user’s PC screen. (Id. ¶
96.)
Web
browsers and Web servers communicate using the TCP/IP protocol. (Id. ¶ 32.)
This is a request/response protocol – when a Web
page is requested, the Web browser sends a request to
the Web server. (Id.) This request includes a portion of the URL for the
requested Web page and the version of the HTTP
protocol being used. The Web server responds to the request by sending
the contents of the requested Web page to the computer on which the Web browser resides. (Id.)
A
Web browser must first establish a connection with a Web server through a
network before it can obtaining a file from the Web
server. (Id. ¶ 33.) Typically, this connection is established
on the Internet via a TCP/IP connection. (Id.) To establish a TCP/IP connection,
the transport-layer protocol software initiates a
request to connect to a special protocol port of the Web
server. (Id. ¶ 34.) If the address of the Web server is specified as an IP
address in the URL for the requested page, then the
computer running the Web browser initiates a request to resolve the
domain name into an IP machine address name. (Id.) Once the TCP connection is
made, the Web browser can send repeated Web page
requests to the same server without making a new connection.
(Id.)
Under
the current Internet Protocol system, each machine connected to an Internet Protocol ("IP") network is addressed using a 32
bit number, the IP address. These addresses are usually
written in "dotted quad" notation, as a series of four 8 bit numbers,
written in decimal and separated by periods. For
example, an IP address might appear as 151.126.95.10. (BT's Counter
to Undisputed Facts ¶ 19.) Many machines have more than one IP address. For example, a machine hosting multiple websites often has an
IP address for each website it hosts. Other times,
a pool of IP addresses is shared between a number of machines – e.g. on a
dynamic IP dialup connection such as the Prodigy
Internet Service, a subscriber’s machine will be allocated
a different IP address each time the subscriber connects. (Clark Decl. ¶ 12.)
The
Prodigy Service
Prodigy
is an Internet Service Provider ("ISP") that has supplied its
customers with access to the Internet since October,
1996. (Prodigy Form 10-K for the period ended Dec. 31, 2000,
Exh. B to Decl. of Benjamin Hershkowitz ("Form 10-K")) Prodigy’s
services include dial-up access, and broadband DSL
access service. (Id.) Prodigy’s networks can provide dial-up access,
using a local telephone call, to approximately 90% of the households in the
United States. (Id.)
According
to Prodigy’s form 10-K, "Prodigy has a variety of arrangements to acquire customers, including . . . contract acquisition programs
with selected major PC manufacturers; bundling [connection
and browser software] with PCs shipped by leading PC manufacturers; . . . includ[ing] Prodigy software with every copy of Windows 98
and Windows 2000 for sale in retail channels; . . .
direct mail and telemarketing. . . ." (Id. at 3.) Prodigy has extensive customer support services to assist members "including
toll-free support, various online support options
and online members helping members program." (Id.) The technical support is
available so that "customers are assisted with
Prodigy software installation," among other things. (Id. at
4.)
Prodigy provides its customers with a series of products and services containing
everything necessary to access the Web.
When
a new subscriber seeks to use the Prodigy Internet Service, Prodigy provides the subscriber with a CD containing a Client Kit Software
package to install on the subscriber’s PC.1 (Wah
Report ¶ 48 (citing Bhakta Dep. at 138; Exh. F, Prodigy Internet).) This kit
includes software that enables a member to dial
through a modem into Prodigy’s Web server and includes a
copy of the Microsoft Internet Explorer version 5.5 Web browser. (Id. (citing
Bhakta Dep. at 138-39).)
Prodigy’s
subscribers connect their PCs to the Prodigy network using local, toll-free or long distance telephone service. (Wah Report ¶ 51 (citing
Form 10-K, Exh. 51.) Subscribers then have access
to the Internet and to Prodigy’s hosting servers. (Id.) The hosting servers
store some of the data accessed during subscriber
sessions. (Id.) Network connections are made. through
communications hardware such as telephone lines, routers, switches, and modems.
(Id.) When a user dials in to the Prodigy system,
the user’s computer is assigned an IP address dynamically
by Prodigy. (Wah Report ¶ 53 (citing Bhakta Dep. at 34-41.) An IP address is a unique binary number assigned to an interface connection
of a computer to the Internet and is used by the
other computers to send packets using the IP protocol to this computer. The
member can then use the Web browser to retrieve Web
pages from the Prodigy Web site or from other Web
servers connected to the Internet. (Wah Report ¶ 53.)
Prodigy
subscribers receive access to the Internet and Prodigy-branded content and services such as personal Web pages, called "the
Prodigy Personal Web Page." (Form 10-K.) Prodigy
customers may create their own Web pages through this service. (Bhakta Dep. at
14-15.) The network infrastructure used by Prodigy
consists of three primary tiers: local phone access
sites; a middle tier, which connects local phone access sites to regional hubs;
and a backbone tier, which connects the regional
hubs to the Internet or Prodigy’s data hosting center. (Wah
Report ¶ 41 (citing Form 10-K, Exh. 51).) Prodigy’s data-hosting center,
located in New York, houses approximately 400
high-capacity servers to store content and other data. (Id.) Prodigy’s
data center contains Web servers and data banks that process and store the Web
pages for Prodigy’s Web sites, which include www.prodigy.net,
www.prodigy.biz, (for
Prodigy’s business customers) and pages.prodigy.net
(for Prodigy members’ personal web sites). (Bhakta Dep. at 33.)
Prodigy
uses nine servers to help distribute the load of multiple subscribers’
accesses. (Prodigy’s Statement of Undisputed
Facts ¶ 24.) Each of these servers has the same content. The
content is information related to Prodigy’s web pages, for example, www.prodigy.net, www.prodigy.biz. (Bhakta Dep.
at 37.) Prodigy also uses three additional Web servers to service
its subscribers’ Personal Web pages. (Id. at 38.) These servers share a copy
of all of the personal web pages for www.prodigy.net.
(Id.)
When
a Prodigy member makes a request to any URL on the Prodigy portal, such as www.prodigy.net., which is
the URL of the Prodigy main portal page, or "home page," the member’s
browser first makes a request to the DNS service to get the IP address that
corresponds to this URL, either from local
information or from querying servers on the Internet. If found, the IP
address that the DNS server returns to the browser points to a load balancing
server. This load balancing server is a separate
physical computer running the IBM Network Dispatcher® software.
It distributes Prodigy user requests to one of the content servers based on the
traffice on each of these servers. The
load-balancing computer has no data store that is accessed by Prodigy
users. (Bhakta Decl. at ¶¶ 5, 6.)
Prodigy
uses a set of three additional Web server computers to service its members’ personal Web pages located at pages.prodigy.net.
(Wah Report ¶ 45 (citing Bhakta Dep. at 38.) These
Web pages are stored on a network file system. (Id. (citing Bhakta Dep. at
38-39).) Each.Web server accesses the same network file to retrieve a requested
personal Web page. (Id.) Prodigy employs a Netscape
server plug-in to convert or map directly each personal Web page
address into its disk file name address using a hashing algorithm. (Wah Report
¶ 46 (citing Bhakta Dep. at 39-41).) The hashing
algorithm takes member-specific alphanumeric information,
such as the login name of the member, and generates a number using a mathematical
algorithm. (Id. (citing Bhakta Dep. at 42).) This number is then combined with the Web page address to obtain the Web page from the
network file system. (Id.)
BT’s
ARGUMENTS IN MICRO
BT
contends that:
1.
Each Web server on the Internet is a "central computer" as defined in
the Sargent patent because each Web server has its
own centralized data store.
2.
HTML files qualify as "blocks of information" either literally or
under the doctrine of equivalents.
3.
Each URL address is a "complete address" within the meaning of the
Court’s construction of the term, either
literally or under the doctrine of equivalents. BT
makes other arguments 2 but it is not necessary to reach them because its
failure to raise any disputed issue of fact as to
these points means that Prodigy is entitled to judgment as a matter
of law.
DISCUSSION
I. The
Internet Does Not Infringe the Sargent Patent
A. The
Internet Has No "Central Computer"
1.
Literal Infringement
A
"central computer" in the Sargent patent is:
a
single device, in one location. It is referred to as "central" because
it is connected to numerous physically separate
stations, called "remote terminals," by the telephone lines of
a telephone network. So there is one computer, connected to many remote
terminals. The central computer means in this
patent thus serves as the hub of a digital information storage,
retrieval and display system – and all of the remote terminals connect to it. The central computer stores information. The central
computer contains a "main store." In the
context of this patent, the main store is a mass information storage or memory device. An example of a main store is a magnetic disk,
which is a rotating circular plate having a
magnetizable surface on which information may be stored as a pattern of polarized
spots on concentric recording tracks. The central
computer contains an information database, which is "centralized" in
the sense that all of the remote users can access
it by accessing the central computer. (Markman Op.
at 9.)
BT
asserts that, while the Internet is made up of an enormous number of computers,
each individual web server is a central computer in
one location (i.e. central relative to the remote terminals
within the meaning of the patent). (Wah Decl. at ¶ 31.)
The
cornerstone of this argument is BT’s assertion that a central computer is not
limited to a single computer as a matter of law. To
support its position, BT quotes a statement I made in TM
Patents, L.P. v. Int’l Business Machines Corp., 72 F. Supp. 2d 370, 380 (S.D.N.Y.
1999) while construing the claim term "a
multi-unit memory system:" Of course, the fact
that the patent claims ‘a’ system does not mean that IBM or some other
party would escape liability for infringement by constructing two or three or
even more such multi-unit memory systems and
somehow linking them together or causing them to
operate together.
Id.
at 380. The flaw in BT’s cornerstone argument is obvious. Not only was I not
construing the Sargent patent in TM, I was not even construing the term "central
computer" when I wrote those words. I was
construing the word "system."3 A system is not the same thing as a
computer, and I never said that it was. A computer,
according to the dictionary, is "a device that receives, processes
and presents data," Dictionary of Scientific and Technical Terms 342 (Sybil
P. Parker ed., McGraw Hill 3d ed. 1984), while a
system is "a combination of several pieces of equipment integrated
to perform a specific function" or "a group of related structures."
Id. at 1600. Thus, the word "system"
fairly implies multiple devices connected together. The Sargent patent does indeed cover a system, one that includes
a central computer as one of its elements. BT
conflates the system with the computer. But I made
it clear in the Markman opinion (as the Sargent patent claims
make clear) that the computer is but one component of the system. The
patent claims as construed clearly provide that the central
computer is one device,
in one location. (Markman Op. at 10-11.) Just as a
circle has but one center, hub-and-spoke networks
have only a single hub. There may be other circles with other centers, just as
there may be other hub-and-spoke networks with
other central computers or hubs. But each system (network)
of the type claimed in the Sargent patent can have only one central computer. Therefore, viewing the Internet as a system (as BT asks me
to do), it does not literally infringe the Sargent
patent, because it contains no such central computer.
The
central computer in the Sargent patent also contains an information database,
which is "centralized" in the sense that
all remote users can access it by accessing the central computer. (Id.)
Prodigy argues that the Internet has no such "centralized data store,"
but rather, contains an extremely diffuse data
storage architecture. According to Prodigy’s expert, this distribution is the essence of the Internet, because it allows users to
access information stored throughout a global
network of networks of computers and storage devices that are loosely linked
through adherence to an open group of protocols and
standards adopted by the Internet community. (Clark
Decl. at ¶ 6.)
BT
does not dispute that this is the case. It responds, however, that its argument
is not that the Internet has a centralized data
store, but that each Web server (or central computer, as BT would
call it) has its own centralized data store because it has a main storage device
for storing HTML files. (Wah Decl. at ¶¶ 31-33;
Wah Report ¶ 176.) However, the claims of the ’662
patent state that the centralized data store contains all
of the blocks of
information accessible by the remote terminals. (Markman Op. at 14.) BT does not dispute that there is no centralized data store for the
Internet that contains all of the data remote users
might care to access.
The
Internet is a network of computers intertwined with each other in order to allow
users around the world to exchange information. The
whole purpose of the Internet is for the sources of
information to be in many places rather than centralized. Any user can retrieve
information that is stored on a Web server in any
physical location, as long as that server is connected to the Web.
For example, a Prodigy user does not have to rely on Prodigy to gather
information from multiple sources and put it on its
own server in order for a Prodigy customer to have access to the information.
Rather, when Prodigy users connect to the Internet through Prodigy’s system,
users can access blocks of information located on
remote systems, e.g. a computer in Alaska, rather than
Prodigy’s own computers in New York. This "network of networks" or
"system of systems" allows users access
to information from a variety of sources, in any location. BT
cites the general rule that addition to an accused apparatus of one or more
features than the claim requires does not preclude
a finding of infringement. But what BT characterizes as
"additions" are fundamental differences in the nature of the claim
elements. As it did during the Markman hearing, BT
would have me exclude the word "central" from the construction of "central computer." The Court expressly rejected
BT’s interpretation by ruling that the Sargent patent
claims a central computer that is a single computer with a centralized database.
Because the Internet is not a computer network
consisting of a centralized computer that stores all of the data
accessible by remote terminals, Web servers on the Internet cannot literally
infringe the ’662 patent.
2.
Doctrine of Equivalents
BT
may not rely on the doctrine of equivalents to withstand Prodigy's motion for summary judgment. The doctrine of equivalents requires
insubstantial differences between the patented
invention and the accused product, as determined on an element-by-element basis. Warner-Jenkinson, 520 U.S. at 29-30. Whether the
differences are insubstantial can be determined by
the function-way-result test, which requires that the claimed and accused
elements "perform substantially the same
function in substantially the same way to obtain the same result." Id.
at 40; Johnson & Johnston Assocs. Inc. v. R.E. Serv. Co., Inc., 285 F.3d
1046, 1053 (Fed. Cir. 2002) (quoting Graver Tank
& Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 609 (1950)). In this
case, the Internet fails the function-way-result test, because a central
computer containing all of the data accessible by a
remote terminal user operates in substantially different ways from the Internet.
A
central computer, as claimed in the Sargent patent, has a one-to-one, hub and
spoke relationship with numerous physically
separate stations called remote terminals. The remote terminals
are connected to the central computer by the telephone lines of a telephone
network. All of the remote terminals connect to the
central computer that has one centralized main store for
storing information. (Markman Op. at 9.) As a result, a remote terminal in the
Sargent patent does not identify in its
communication protocol a computer with which it would like to communicate,
for it communicates with only one central computer. In contrast, a computer operating on the Internet must at all times identify a
specific computer with which it seeks to communicate,
because not all of the remote users are connected to a single central source of information. If all Internet users were connected to
Prodigy’s Web server alone, BT’s argument might
have some force; but all Internet users are not connected to Prodigy. Instead,
Internet users have access to multiple Web servers
on the Internet that store information in various locations. Indeed,
the Internet is the very antithesis of a digital information storage system
having a central computer. The opposite of a claim
limitation cannot be considered its equivalent. Moore U.S.A.
Inc. v. Standard Register Co., 229 F.3d 1091, 1106 (Fed. Cir. 2000). In Moore,
the Federal Circuit held that a claim limitation
reciting that an adhesive "extend[s] the majority of the
lengths of said longitudinal marginal portions," could not be infringed
either literally or under the doctrine of
equivalents by a product having adhesive extending only a minority of a marginal portion (i.e. less than 50%) – even if it was almost a
majority (e.g., 47.8%) Id. According to the Federal
Circuit, "it would defy logic to conclude that a minority – the very
antithesis of a majority – could be
insubstantially different from a claim limitation requiring a majority."
Id. BT cannot dispute that any user throughout the
world can access information stored in any of the
millions of computers connected to the Internet. In contrast, the Sargent patent
claims revolve around a central computer – a
single device, in one location, with one main data store. The
Internet, is, in short, an entirely different beast from the system described in
the Sargent
patent.
Consequently, the Internet does not infringe the Sargent patent either literally
or under the doctrine of equivalents. Prodigy is
therefore entitled to summary judgment as a matter of law.
B. The
Phrase "Central Computer" Limits Claims 5, 6, and 7
BT
argues in its Opposition Brief that "central computer" does not limit
claims 5, 6, and 7. BT relies on Catalina Marketing
Int’l, Inc. v. Coolsavings.com, Inc., 289 F.3d 801, 808 (Fed. Cir.
2002), and Rowe v. Dror, 112 F.3d 473, 478 (Fed. Cir. 1997), for the proposition
that "a preamble [is] not limiting ‘where a
patentee defines a structurally complete invention in the claim
body and uses the preamble only to state a purpose or intended use for the
invention.’" However, Catalina and Rowe are
of little help to BT because limitations in the preamble may
in fact limit the claim terms if the "patentee uses the claim preamble to
recite structural limitations of his claimed
invention." Rowe v. Dror, 112 F.3d at 478. In the patent at issue, the phrase "central computer" is not merely a term
that appears in the preamble of the Sargent patent to
describe the purpose of the patent. This term appears in the body of the Sargent
claims as well, and recites structural limitations
of the patent. In Catalina, the Federal Circuit
ruled that:
While
the phrase "located at predesignated sites such as consumer stores"
appears only in the preamble of Claim 1, this
language appears both in the preamble and body of Claim 25.
Hence, the applicants specifically included this language in the claim not once,
but twice. By virtue of its inclusion in the body
of Claim 25, the phrase limits Claim 25. 289 F.3d
at 810-11; see also Karsten Mfg. Corp. v. Cleveland Golf Co., 242 F.3d 1376,
1380 (Fed. Cir. 2001) (finding that preamble term
limited claim because the term was used in the specification
as well as in all of the claims).
The
term "central computer" appears both in the preamble and the body of
claim 5 of the Sargent patent. Indeed, "central
computer" appears three times in the body of claim 5. Since claims
6 and 7 depend upon claim 5, these claims are also limited by anything that
limits claim 5. Indeed, the Court’s Markman
opinion makes clear that the terminals claimed 5, 6, and 7 must be used
in a system that includes a "central computer," as defined in claim 3.
(Markman Op. at 30.) Furthermore, the Sargent
applicant repeatedly distinguished the prior art during the prosecution
of claims 5-7 by pointing to the claimed novelty of the two-part block of
information and the manner in which such blocks
were stored and accessed from the central computer’s main store.
(Applicant’s Appeal Br. dated Jan. 5, 1987.) Thus, unless a Prodigy user’s
PC is used in a system in which a central computer
exists – or at a minimum is specially adapted for use in such a
system – there can be no infringement of the Sargent patent. BT’s
argument that the phrase "central computer" does not limit claims 5,
6, and 7 is thus without merit.
C.
The Internet Does Not Contain Blocks of Information as Required by the Sargent Patent
1.
Literal Infringement
A
key distinction between the ’662 patent and the Prodigy Internet Service is
the requirement of the ’662 patent that
information be stored as blocks of information. These blocks have
special characteristics:
Each
block has a first portion and a second portion. These portions are separable, contiguous and co-stored sub-units. That means that the
portions are stored together, and they are stored
next to each other, yet they can be separated from each other. A block of information may contain very limited programming
information, the purpose of which is to reduce the
complexity of keying required to communicate with the central computer. The
two parts of a block of information are the first portion and the second portion.
The first portion of a block of information is
intended for visual display on a remote terminal. The
second portion is information not intended for display. The second portion
contains the complete address for each of the other
blocks of information referenced in the first portion.
So if a block of information is referenced in the first portion, then the
complete address for that block of information will
be in the second portion. The second portion may
also contain other information as well, such as information to influence the
display or to reduce the complexity of keying
required to communicate with the central computer. But
it never contains information intended for display. (Markman
Op. at 14.)
Unlike
the blocks of information required by the ’662 patent, HTML code, which is the primary language of the World Wide Web and of the Prodigy
Internet Service, does not use blocks. HTML code
does not separate displayed information into a first sub-unit, and non-displayed information in a contiguous, separable second sub-unit.
Rather, HTML code contains information to be
displayed intermingled with other information concerning formatting and linking,
such as URLs and anchors. In HTML, hypertext
references include each URL link adjacent to each phrase or image
for display. For example, in the code, <script>document.write(HTMLCacheArray[34];</script><A href=(http://www.msnbc.com/modules/exports/ct_prodigy.asp?/news/736921.asp(target=(_top(>Yahoo!
profits meet forecasts</A><TD> the information for display is shown in bold, and the URL
is shown in italics. The first linking tag (which
is referred to in HTML as an "anchor tag," and identified with the
"<A"), is the information always for
display associated with each URL. This information appears on the screen.
The user can mouse this link, and click to activate it. The URL for each such
component of displayed information is set forth in
the HTML code immediately prior to the information for display.
Consequently, URLs associated with information for display are not located contiguously and separably with a sub-unit of information
not for display. BT argues that the Court’s
construction of "blocks of information" does not require that the
displayed and non-displayed information be co-stored as separate and discrete
contiguous sub-units, but merely requires that the
portions be separable. (BT Opp. Br. at 20.) It bases this argument
on the Court’s omission of the word "segregated" from its
construction of the term "blocks of
information." But the requirement that the first and second portions not be "intermixed" was very much a part of the Court’s
claim construction: Prodigy asks me to incorporate
the word "segregated" into the construction, relying on the
same passage from the 1986 Amendment and the 1987 Appeal Brief, (Jan. 24, 1986 Appeal Br. at 3; Jan. 5, 1987 Appeal Br. at 3; File
Wrapper Tab 22), to clarify that the portions may
not be intermixed. While the word "segregated" does appear in the specification and the prosecution history, I am not aware
of any rule of patent construction that requires me
to incorporate each and every word ever used in conjunction with
a patent claim in the claim construction. The parties have not suggested any meaning for "segregated" that makes it integral
to the definition of the term "block of information."
The very idea of the block’s two discrete sub-parts – the first and second portions – incorporates the notion of segregation or
separability, and I have already decided to include
separability in the definition. It will unnecessarily confuse the jury to incorporate a second word that means much the same thing. (Markman Op. at 20.)
During
the Markman phase, the Court expressly rejected BTs contention that the terms "co-stored" and "contiguous " in the
Sargent patent are synonomous. (Markman Op. at 15-16.) While
I did not include the word "segregated" into the explanation of "blocks
of information" that I would have given to the
jury, I did this because I had already included the word "separable" into the definition and did not want to confuse the jury
by including a second word that, to me, meant the
same thing. The word "segregated" was omitted from the definition of
"blocks of information" because it
appeared to me to be duplicative of the term "separable," not to
remove from the claim the fact that the
sub-portions are not intermingled. Alternatively,
BT’s expert, Dr. Wah, states that it is possible to create HTML files so that "they have a top portion containing only information
for display and a bottom portion containing only
information not for display but including the complete assigned address for
another HTML file stored on the same server."
(Decl. of Benjamin Wah at ¶ 22.) He illustrated his argument in two
ways. Neither is persuasive.
First,
Doctor Wah created web sites that he alleged were infringing and placed them on Prodigy's system.4 BT cannot claim that Prodigy infringes
its patent, or induces others to infringe its
patent, if it must invent the infringing device itself. "A device does not
infringe simply because it is possible to alter it
in a way that would satisfy all the limitations of a patent claim." High
Tech Medical Instrumentation v. New Images Indus., Inc., 49 F.3d 1551, 1555 (Fed.
Cir. 1995) (finding that a camera that was "rigidly
coupled to its housing" did not infringe a device that
claimed a rotating camera, since the screws on the accused infringing camera
would have to be loosened in order for it to rotate
like the patented device). An example of this
well-established rule of patent law is found in Bionx Implants, Inc. v. Linvatec
Corp., 99 F. Supp. 2d 396, 396-99 (S.D.N.Y. 2000). In that case, the plaintiff
held a patent on a small plastic suture used to
repair tears in the menisci of the knee. The accused infringer
marketed a suture that was similar to the patented device but was made of a
flexible material and had a hole running through
the interior of the suture. The court found that the claims
of the patent described a device that was rigid. The applicant distinguished his
suture from prior art because the other suture was
made of a flexible material that could not "by itself be pushed
into a body tissue without the use of a needle." Id.
In
support of its position that the allegedly infringing suture was similar to the
patented device because it could be pushed into the
body without using a needle, the plaintiff submitted a videotape
to the court. In the video, they showed that the infringing device could be
pushed into the meniscal tissue without first
cutting the tissue. The court noted, however, that the insertion was
only made possible by using a special insertion rod designed specifically for
the video demonstration. The court found that this
tactic violated the principle established in High Tech Medical
that a device does not infringe because it can be made to infringe. "[T]he
question is not what a device might have been made
to do, but what it was intended to do and did do." Bionx Implants,
99 F. Supp. 2d at 398 (citing High Tech Medical, 49 F.3d at 1555). Second,
Dr. Wah identified the "sheilagenealogy" page on Prodigy’s system as
containing a "block of information" as
defined in the Sargent patent. (Wah Dep. at 41-42.) Unfortunately, for
BT, this HTML file does not look remotely like the "blocks of information"
claimed in the ’662 patent. The Sargent patent
requires that second portion contains the complete address for each
of the other blocks of information referenced in the first portion. So if a
block of information is referenced in the first
portion, then the complete address for that block of information
will be in the second portion. Dr. Wah acknowledged in his deposition that the
parts that he identified as "second portions"
did not contain multiple complete addresses. (Id. at 50- 51.)
Dr.
Wah further testified that, in order to create a "block of information"
containing at least two complete addresses from
HTML, one would have to link conceptually two distinct, bracketed
"sub-blocks." Even after performing this conceptual reconfiguration of
the HTML file, all of the HTML paragraphs
containing URLs would have to be somehow connected in order for
them to be configured as described in the Sargent patent. This is a far cry from
the "neatly segregated block" the
applicant discussed during the prosecution of the Sargent patent. (Applicant's
Appeal Br. dated Jan. 5, 1987 at 12-13.) Consequently, I find, as a matter of
law, that the HTML file that Dr. Wah presents as
evidence of an infringing "block of information" according
to the Sargent patent does not literally infringe the claims of the Sargent
patent. BT is correct that there is no requirement
that its expert analyze every available web page that
it claims qualifies as a block of information. However, in order to raise a
disputed issue of material fact, BT must identify
at least one web
page other than one it invented for the purposes of
this lawsuit. It has not done so. Other than BT's general assertion that blocks
of information as defined in the Sargent patent
exist on the Internet, BT has not presented any other evidence of an
infringing Web page, on Prodigy's system or anywhere else on the Internet. Since
the one example that Dr. Wah identified as an
infringing Web page (other than the ones he created) does not
literally infringe the Sargent patent, BT's expert's conclusory statement that
there are infringing HTML files on the Internet
does not raise a genuine issue of material fact precluding summary
judgment. Phillips Petroleum Co. v. Huntsman Polymers Corp., 157 F.3d 866, 876 (Fed.
Cir. 1998).
2.
Doctrine of Equivalents
BT
cannot rely on the doctrine of equivalents to withstand Prodigy's motion for
summary judgment.
a.
Application of the Doctrine of Equivalents is Barred with Respect to
"Blocks of Information" Because the Applicant Made Unmistakable
Assertions to Avoid the Prior Art
The
doctrine of prosecution history estoppel prevents a patent owner from relying
upon the doctrine of equivalents when the patent
applicant relinquishes coverage of subject matter during
the prosecution of the patent, either by amendment or argument. Pharmacia &
Upjohn Co. v. Maylan Pharms., Inc., 170 F.3d 1373,
1376-77 (Fed. Cir. 1999); Southwall Techs. Inc. v. Cardinal
IG Co., 54 F.3d 1570, 1583 (Fed. Cir. 1995). A patentee cannot invoke the
doctrine of equivalents to "embrace a
structure that was specifically excluded from the claims." Dolly, Inc. v. Spalding & Evenflo Cos., Inc., 16 F.3d 394, 400 (Fed.
Cir. 1994). During the prosecution of the Sargent
patent, a new limitation was added to the phrase "blocks
of information" in order to distinguish the Sargent patent from the Quinn,5
Fedida,6 and Cramer 7 references.8 To distinguish
the Quinn reference, for example, the Applicant cited to several
of its narrowed claim limitations: For example,
there is absolutely no suggestion anywhere in Quinn that blocks of stored data should include a first portion containing information
for display and a second portion containing
information not for display but including the complete address for each of plural other blocks of information. Nor does Quinn teach .
. . manual entry of keyed digital data of less
extent than any one complete address but nevertheless uniquely indicative of one
of the complete addresses contained in the second portion of the block of
information (which contains the first portion then being displayed). (Jan.
26, 1983 Amendment at 9-10.) Because BT relied on
narrowing amendments to overcome prior art, equivalents are unavailable
with respect to these claim limitations. Warner Jenkinson, 520 U.S. at 33. Thus,
BT is barred from asserting that Prodigy’s
Internet service meets the above claim elements under the doctrine
of equivalents.
Additionally,
in order to overcome rejection for obviousness under 35 U.S.C. § 103, the Sargent applicant argued that the first and second
portions of a block of information must be contiguous,
co-stored, and co-addressed: And it must never be
forgotten that none of the references specifically teach or suggest a system
wherein each "screen" of displayed data is specifically associated
with its own segregated but yet contiguously
located and co-stored and co-addressed index for assisting
in the address selection of the next screen of data to be displayed. (Applicant’s
Appeal Brief dated Jan. 5, 1987 at 12-13 (emphasis in original).) Therefore, BT cannot now rely on equivalents for "blocks of
information" to sustain a claim of infringement. This,
too, was a narrowing of the original claim in order to obtain a patent.9
b.
Even if Prosecution History Estoppel did not Prevent BT from Claiming
that Equivalents Infringed the Sargent Patent, Prodigy Would
Still Be Entitled to Summary Judgment as a Matter of Law Infringement
under the doctrine of equivalents requires that the accused device perform substantially the same function to achieve substantially
the same result in a substantially similar manner
as the patented device. Dolly, 16 F.3d at 398. "The doctrine of equivalents
is not a license to ignore claim limitations."
Id. (citing Pennwalt Corp., 833 F.2d at 935). "The accused device
must . . . contain every limitation
or its equivalent." Id. (citing Intel, 946 F.2d at 832). HTML
requires that displayed and non-displayed information be interspersed in a
fashion that is entirely different from the
separate sub-units claimed in the Sargent patent. This intermixing
of displayed and non-displayed information not only differs substantially from
the blocks of information claimed in the Sargent
patent, it is the antithesis of the "neat" Sargent method.
BT therefore cannot claim infringement of the Sargent patent under the doctrine
of equivalents. Moore U.S.A. Inc., 229 F.3d at
1106..22
D. The
Phrase "Blocks of Information" Limits Claims 5, 6, and 7
BT
argues in its Opposition Brief that the phrase "blocks of information"
does not limit claims 5, 6, and 7. "Blocks of
information" applies to the patent for the same reasons that "central
computer" applies. "Blocks of information" is not merely a term
that appears in the preamble of the Sargent patent
to describe the purpose of the patent. This phrase appears in the body
of the Sargent claims as well, and recites structural limitations of the patent.
"Blocks of information" appears five
times in the preamble and five times in the body of claims. Since claims
6 and 7 depend upon claim 5, these claims are also limited by anything that
limits claim 5. Furthermore, the Sargent applicant
repeatedly distinguished the prior art during the prosecution
of claims 5-7 by pointing to the claimed novelty of the two-part block of
information and the manner in which such blocks
were stored and accessed from the central computer’s main store.
(Applicant’s Appeal Br. dated Jan. 5, 1987.) Thus, unless a Prodigy user’s
PC is used in a system in which blocks of
information are downloaded – or at a minimum is specially adapted for
use in such a system – there can be no infringement of the Sargent patent. BT’s argument that the phrase "blocks of
information" does not limit claims 5, 6, and 7 is therefore
without merit.
E.
Neither a URL Nor the Path Component of a URL is a Complete Address
1.
Literal Infringement
All
of the asserted claims of the ’662 patent require that the complete address
for each associated block of information that is
referenced in the displayed portion be stored in the second portion: the complete address is the entire number or name that
uniquely identifies, without reference to other
information, a location on the central computer’s main store where a block
of data will be found. (Markman Op. at 14.)
As
explained in the Markman opinion, a complete address is not a virtual address.
(Id. at 19-20.) Virtual addresses are not complete
addresses because they require reference to other information.
"A complete address is . . . more than simply the address by which the
central computer is able to retrieve a block of
information. The complete address calls up the block of information
directly, without reference to another address." (Id. at 20.) Prodigy
argues that a URL is not a complete address under the meaning of the Sargent patent because a URL does not convey the location of
information on a particular computer. Instead, a
URL requires additional information in order to pull up a web page from a server. When a browser accesses any content page on the World Wide
Web, the browser first obtains the IP address of
the server where the content resides before it can access the information. The browser makes a request to DNS to get the IP address of
the web server which hosts the desired content. The
user’s PC will access an external DNS server, or, if available, DNS
information in cache.
This
multi-step method is used for two reasons. First, numerical IP addresses are not
as easy to remember as URLs with alphabetic names.
(Clark Decl. at ¶ 13.) Second, this system avoids
having to change a URL every time the physical location of a web page is changed. (Clark Decl. at ¶ 14.) For example, a user seeking to
access a Prodigy web page would type www.prodigy.net
to reach Prodigy, whether Prodigy’s servers are
housed in New York or in Alaska.
An
example of how the computer must access additional information is demonstrated
by an extremely frustrating experience that
Internet users commonly experience. If the DNS system is
unavailable, the user cannot access the desired information. This is because the
URL is not a complete address. Rather, it points to
other information that must be accessed in order to pull up the
requested Web page.
Both
Sargent, the sole named inventor of the ’662 patent, and Clarke, the BT Head
of Division who monitored the overall system design
of Viewdata, concurred that the complete address
required by the ’662 patent requires a physical address, including a tract and
sector number of the desired block on the main
store on the central computer. (Sargent Dep. at 198 ("the
track and sector number is the complete address of where that page is");
Clarke Dep. at 11 ("It [a complete address]
would certainly include a sector number, because if you do not have a sector
number you do not have anything like a complete address.").) A
URL contains names – or virtual addresses. It then points to several other
sources of information that must be obtained to
determine a complete address:
1.
The user’s computer must first attempt to translate the URL server name into
an IP address, by reference to other information in
the form of either the external DNS service or
locally-cached DNS information.10
2.
When communication with a content server is achieved, the relative path contained
in the URL must be translated using other information in the form of the
configuration file of the content server to identify an actual path; and
3.
The actual path must be referred to other information in the form of a lookup
table on the operating system’s file system to
determine a physical address for the requested information.
BT
responds to Prodigy’s argument by arguing that the complete address is the
path component of the URL, not the entire URL. (Wah
Report ¶¶ 202, 205.) BT therefore argues that DNS
resolution is irrelevant, because the path component of the URL only comes into
play after the domain name resolution process has
been established. BT notes that during DNS resolution
of the Web server name, the path component is not modified. Hence, BT asserts
that what Prodigy is doing is attempting to avoid
infringement by arguing that the Internet has additional
elements that the patented invention does not require. However,
BT’s alleged "complete address" – the path name – is by
definition incomplete on a network of networks like
the Internet. BT’s own expert concedes as much. (Wah Report ¶ 32;
BTO at 25.) In the absence of the server information, the user’s PC could not
know which one of the millions of computers on the
Internet is supposed to give it the desired information. This
problem does not arise in the context of the Sargent patent, of course, since it
requires a central computer. But in the context of
the Internet, a path name cannot be a complete address. BT
argues that the phrase "without referring to other information," as
construed by the Court, requires only that the
"complete address" be a unique reference to a memory location rather
than a virtual address. BT asserts that regardless of the type of address used,
a computer will translate and reformat the address
into various forms and formats before the computer accesses
the memory location referenced by the address. Therefore, plaintiff’s argument
is that the construction provided by the Court –
that a complete address is a "name or number" – necessarily requires
some translation, since names must be translated by the computer into a digital
or binary form before they can be processed by the computer. (Markman Op. at
14.) BT contends that the Sargent patent
contemplates that the computer will perform routing tasks such as
translating the complete address for a file name into its equivalent binary
number representation which the computer uses to
access the referenced memory location. But BT’s
argument would render the term "complete address" meaningless. The testimony of its own expert, Dr. Wah, illustrates the
point. Dr. Wah takes the position that, since virtual
addressing is excluded from the scope of "complete address," every
identifier is a complete address unless it is a
virtual address:
Q.
What if I typed an address and it was displayed on the screen and a runner
looked at the screen, went through a phone book,
and determined the physical address; would that be
other information?
A.
According to the judge’s ruling, anything that is not virtual addressing does
not – it is not information that would be relevant. Meaning –
Q.
You are saying any other information, in your opinion, is possible, as long as
it is not virtual addressing?.
A. That
is correct.
(Wah
Dep. at 108.) Dr. Wah was then asked to assume that the Court’s construction
prohibiting "other information"
prohibited any other information, not just "virtual addresses." When
asked if URLs are complete addresses applying that
construction, Dr. Wah admitted that they are not. (Id.)
Dr.
Wah is right about this, and unfortunately for BT, the term "complete
address" as used in the Sargent patent means
that the address really is complete, in
that it does not require reference to additional
information in order to retrieve the requested information. BT's interpretation
would eliminate the word "complete" from the patented claim language,
something this Court cannot allow. Thus, as a
matter of law, a URL or even a subset of a URL is not a "complete
address" as defined by the Sargent patent.
2.
Doctrine of Equivalents
a.
Application of the Doctrine of Equivalents is Barred with Respect to
the Complete Address element because the Applicant Made Unmistakable
Assertions to Avoid Prior Art.
During
the prosecution of the Sargent patent, the applicant replaced the original
claims with claims 19-22, 11 narrowing the literal
scope of the claims. Because the applicant relied on narrowing
amendments to overcome the prior art, equivalents are unavailable with respect
to these claim limitations. Warner Jenkinson, 520
U.S. at 33. Thus, BT is barred from asserting that
Prodigy’s Internet Service meets the complete address element under the
doctrine of equivalents.
In
response to an obviousness rejection over the Cramer reference, the Applicant
argued that the ’662 patent does not rely on
pointers to the complete address. (Applicant’s Amendment dated
August 29, 1983 at 7; Applicant’s Amendment dated March 28, 1985 at 6;
Applicant’s Amendment dated Dec. 4, 1985 at 9;
Applicant’s Appeal Brief dated Jan. 24, 1986 at 9; Applicant’s
Appeal Brief dated Jan. 5, 1987 at 15.) As the Court noted in the Markman
opinion, during the course of the patent
prosecution, the applicant distinguished the NTZ article by stating that
his simple addressing system allows the "abbreviated keyed-in data to
directly read out the complete address of the next
block to be fetched from the main store." (Dec. 4, 1985 Amend., File
Wrapper Tab 16 at 9-10.)
In
distinguishing the Sargent patent from Tsuda the applicant stated:
Indeed,
the execution of a "Judge Order" (col. 6, line 40 et seq.) [as
described in Tsuda]. involves a fairly complex
algorithm to determine what next should be displayed. By contrast,
the applicant’s novel arrangement permits one to merely directly read out from
a ‘further’ memory the next complete address
– as addressed by abbreviated input keyed-data. (Id.
at 7.)
These
arguments were relied upon by the Board of Patent Appeals:
We
find no indication, however, that the "Request Order" [taught by Tsuda]
includes the complete addresses of the next block
of information which is to be retrieved and utilized for
display purposes as claimed. At most it would appear to us from a review of the [prior art] that the data included in the request order [taught
by Tsuda] would be used by the computer (1) for
determining such an address. (Appeals Decision
dated May 9, 1983 at 4.) The Applicant made these
arguments in order to distinguish the complete address from the
virtual address and branching programs used in prior art. Consequently, BT is
precluded from asserting that a virtual address
infringes under the doctrine of equivalents because here, too, it
made narrowing amendments during prosecution that limited the scope of the term
"complete address."
b.
Even if Prosecution History Estoppel did not Prevent BT from Claiming
that Equivalents Infringed the Sargent Patent, Prodigy Would
Still Be Entitled to Summary Judgment as a Matter of Law. URLs
function in substantially different ways from complete addresses. URLs identify
a computer and page by name, instead of by address.
Indeed, a URL is the antithesis of a complete address,
because it cannot call up a Web page without accessing additional information.
In other words, a URL is incomplete. Since an
incomplete address is the antithesis of a complete address, there
can be no infringement under the doctrine of equivalents.
II.
Contributory Infringement and Active Inducement
Because
the Internet itself does not infringe the Sargent patent, Prodigy can not be
liable for contributory infringement or active
inducement for providing its users with access to the Internet.
I therefore need not address BT's arguments concerning contributory infringement
and active inducement in any detail.
III.
Prodigy’s System Does Not Directly Infringe the ’662 Patent
BT’s
argument that Prodigy’s Web servers directly infringe the Sargent patent also
fails, because Web pages stored on Prodigy’s web
servers do not contain "blocks of information" or "complete
addresses" as claimed in the Sargent patent. Therefore, Prodigy’s System
does not infringe the ’662 patent as a matter of
law, and I need not address Prodigy’s other non-infringement arguments.
CONCLUSION
In
contrast to what BT would have us believe, there are no disputed issues of
material fact in this case. Instead, the two sides
reach vastly different conclusions based on the same set of facts.
I find that, as a matter of law, no jury could find that Prodigy infringes the
Sargent patent, whether directly or contributorily,
either as part of the Internet or on its Web server viewed separate
and apart from the Internet. Prodigy’s motion for summary judgment is
therefore granted. The Clerk is directed to close
the file. This constitutes the decision and order
of the Court.
Dated:
August 22, 2002
Footnotes
.1
The minimum system requirements suggested for a PC to run Microsoft Internet Explorer version 5.5 SP2 under Microsoft Windows 95
includes a 66 megahertz (MHz) Intel 486 processor,
16 MB of RAM, a Microsoft or compatible mouse, and a modem or Internet connection.
(Wah report ¶ 50 (citing Internet Explorer 5.5 Service Pack 2 and Internet
Tools. Microsoft Windows: URL:http//www.microsoft.com/windows/ie/downloads/recommended/ie55sp2/default.
asp.).)
2
BT also argues that in PCs, the first and second portions of blocks of
information are separated and stored in separate
memories (or the separation and storage of HTML files is substantially
similar to the separation and storage of blocks of information); a computer
mouse is a keypad means and it provides for manual
entry of keyed digital data; and a URL is not intended for
display (or the URLs in the second portion of an HTML file are substantially
similar to the second portion of a block of
information).
3
Moreover, I was doing so in the context of a patent that is not even remotely
like the Sargent
patent. The quote is taken from the Court’s construction of the ’342 patent.
That patent is for a computer system that is able
to detect and correct errors in data stored in the computer. The
system uses an error correction code with the capacity not just to detect errors
in data but to correct them as well, and spare disk
drives to back up the corrected data, so that there are always two
copies in the system.
4
There is some dispute about who created these files. However Dr. Wah testified
that he loaded them onto Prodigy's system with the
assistance of BT's counsel. (See Prodigy's Supp. Mem.
in Supp. of Summ. J. at 1-2.)
5
The Quinn reference, U.S. Pat. No. 3,688,276, is a computer controlled vending
and reservation system. In Quinn, a central
computer controls remote vending machine terminals, and
the computer acts as a memory storage for keeping inventory and accounts.
6
The Fedida ‘Viewdata’ paper describes BT’s Viewdata system.
7
The Cramer reference, U.S. Pat. No. 4,065,810 describes a data transfer system
allowing a terminal to use a modem to access data
stored in a memory store.
8
Claim 19, which became issued claim 1 after subsequent amendments, introduced
the limitation that the blocks of information
contain the complete address for each of plural other blocks
of information in the second portion of the block. This claim also introduced
the "further memory means" limitation and
the use of keyed digital data to select the next complete address.
9
The Court did not need additional briefing from BT on this issue because
Warner-Jenkinson remained the law, even before the
Supreme Court re-affirmed its principles in Festo. See
Festo, 122 S. Ct. 1831.
10
This may require reference to many DNS servers.
11
These claims became claims 1-4 of the issued patent.
back
to the overview
