Opinion
CIVIL ACTION NO. 02-11324-WGY.
November 8, 2006.
Jeffrey T. Rotella, Kenneth J. Demoura, Adler Pollock Sheehan, P.C., John L. Capone, Martin J. O'Donnell, Kevin Gannon, Michael E. Attaya, Thomas C. O'Konski, Cesari McKenna, LLP, Boston, MA, Kenneth E. Levitt, Peter M. Lancaster, Dorsey Whitney LLP, Minneapolis, MN, W. Robinson H. Clark, Dorsey Whitney LLP, Washington, DC, for Plaintiffs.
Aimee B. Kolz, J. Pieter Van Es, Banner Witcoff, LTD., Chicago, IL, Charles K. Verhoeven, Helen E. Dutton, Quinn Emanuel Urquhart Oliver Hedges, LLP, San Francisco, CA, Dale A. Malone, John P. Iwanicki, Thomas P. McNulty, Banner Witcoff, LTD., Boston, MA, David R. Stewart, RealNetworks, Inc., Seattle, WA, Leah J. Russin, Richard H. Doss, Susan L. Wines, Jonathan A. Land, Quinn Emanuel Urquhart Oliver Hedges, LLP, Los Angeles, CA, Edward J. De Franco, Quinn Emanuel, New York, NY, for Defendant.
MEMORANDUM
I. INTRODUCTION
Ethos Technologies, Inc. ("Ethos") brought this action against RealNetworks, Inc. ("RealNetworks") alleging willful patent infringement by RealNetworks as to Ethos' U.S. Patent No. 6,049,892 (filed Feb. 24, 1997) (issued Apr. 11, 2000) (the "`892 patent"), and U.S. Patent No. 6,381,709 (filed Apr. 3, 2000) (issued April 30, 2002) (the "`709 patent"). See Compl. [Doc. No. 1] ¶¶ 5-14. RealNetworks filed a counterclaim seeking a declaratory judgment of non-infringement and of invalidity as to the two patents. Ans. and Countercl. [Doc. No. 5] at Countercl. ¶¶ 5-30.
The patents teach a process that automatically restarts the download of a file from the internet if the initial download terminates. See `892 patent, Abstract; `709 Patent, Abstract. At trial, Ethos alleged that RealNetworks' products infringed on claims 7-12 of the `892 patent and claims 1, 2, 4, and 5 of the `709 patent.
After three weeks of trial and two and a half days of deliberation, the jury returned a verdict in favor of RealNetworks. The jury found that claims 7, 9 and 12 of the `892 patent and claim 1 of the `709 patent were anticipated and obvious. Jury Verdict [Doc. No. 363] at 1-2. The jury further found that claims 8, 10, and 11 of the `892 patent were not anticipated but were obvious. Id. The jury also concluded that claims 2, 4, and 5 of the `709 patent were neither anticipated nor obvious. Id. at 2. Finally, the jury found that RealNetworks did not infringe any of the claims at issue in the `892 and `709 patents. Id. at 1-3.
The Court employed one procedural innovation in the trial of this patent case, requiring the defendant to present the issues of anticipation and obviousness to the jury followed by the plaintiff's rebuttal. Next, the plaintiff presented its evidence of infringement and the defendant rebutted. Last, the plaintiff proceeded to the damages phase. This approach was eminently satisfying. Presentation of the anticipation defense allows the jury at once to consider the patented advance against the backdrop of the prior art.
The Court intends to continue using this approach to evaluate its fairness and functionality. As the Court explained in The Holmes Group, Inc. v. West Bend Housewares, LLC, Civ. No. 05-11367-WGY [Doc. No. 44], at 46-47 (D. Mass. Sept. 27, 2006) (transcript of Markman hearing):
I've done this once and I liked it so much I'm going to do it again. If anticipation, obviousness are defenses, and they are, I take it prior art . . . is a defense, when we get before the jury[, the defendants are] going first. And the reason I like that, and now I'm talking it up, is that the statistical, the studies by [Kimberly Moore suggest] that these cases go [in favor of] plaintiff because the plaintiff is the patent holder. [Kimberly Moore, Judges, Juries, and Patent Cases-An Empirical Peek Inside the Black Box, 99 Mich. L.Rev. 365, 385, 391 (2000).] I think a case should be tried right down the middle. And the commentators keep saying, well, the jury is in love with inventors and inventors hold the patents and therefore the inventor wins. [Id. at 372 (collecting citations).]
[The defendants are] just going to go first. We're not going to then send the jury out. But [the defendants will] put on prior art, all of those defenses. I will charge as to clear and convincing evidence, then [the plaintiff will] put on infringement, and then we'll deal with damages and [the defendants can] rebut damages. That's how we will try the case.
The great advantage of that is the jury sees what's out there and then they can assess how the patented device advances on the knowledge that's out there generally. I find that very helpful.
The Court is aware that the obviousness defense is for the trial judge to decide. E.g., Alza Corp. v. Mylan Labs., Inc., 464 F.3d 1286, 1289 (Fed. Cir. 2006) (explaining that obviousness is a question of law). As is the Court's wont, the Court desired an advisory jury verdict on the issue. The jury's verdict on anticipation and infringement mooted the need for further consideration of the obviousness issue.
The Court denied Ethos' motion for judgment as matter of law [Doc. No. 374], and writes now only to explain its ruling as to the claim construction of the disputed terms in the patents.
II. DISCUSSION
A. Standard of Review
Claim construction is matter of law with decision exclusively for the Court. Markman v. Westview Instruments, Inc., 517 U.S. 370, 372 (1996) ("We hold that the construction of a patent, including terms of art within its claim, is exclusively within the province of the court."), aff'g 52 F.3d 967, 978-79, 987 (Fed. Cir. 1995) (en banc). Claim construction always begins with a review of the claim language. CAE Screenplates Inc. v. Heinrich Fiedler GmbH Co., 224 F.3d 1308, 1316 (Fed. Cir. 2000); Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996). The words of the claim are generally given their ordinary and customary meaning — the meaning that the term would have to a person of ordinary skill in the art in question at the time of the invention (the effective filing date of the patent). Phillips v. AWH Corp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) (en banc); see also Omega Eng'g, Inc. v.Raytek Corp., 334 F.3d 1314, 1323 (Fed. Cir. 2003) ("We indulge a `heavy presumption' that claim terms carry their full ordinary and customary meaning unless the patentee unequivocally imparted a novel meaning to those terms or expressly relinquished claim scope during prosecution." (internal citations omitted)). In some cases, "the ordinary meaning of claim language as understood by a person of skill in the art is readily apparent" and "involves little more than the application of the widely accepted meaning of commonly understood words." Phillips, 415 F.3d at 1314. Where it is not, the Court looks to other sources that show what a person of skill in the art would have understood that claim language to mean. Id. The Court primarily relies on "intrinsic evidence" of the claim, which includes the claim language itself, the specification, and the prosecution historyId. at 1314-17. The specification "is the single best guide to the meaning of a disputed term." Id. at 1315 (quoting Vitronics, 90 F.3d at 1582). Moreover, although the prosecution history frequently lacks the clarity of the specification, it can "often inform the meaning of the claim language by demonstrating how the inventor understood the invention and whether the inventor limited the invention in the course of prosecution, making the claim scope narrower than it would otherwise be." Id.
Extrinsic evidence — dictionaries and learned treaties — can sometimes be useful in claim construction but is generally considered less reliable than the intrinsic record in determining how to read claim terms. Id. at 1318. The court must be careful to consider extrinsic evidence in light of the intrinsic evidence, and discount any extrinsic evidence that is at clear odds with the intrinsic evidence. See id. at 1318-19.
B. Claim Construction
The parties originally asked the Court to construe twelve terms found in either or both of the `892 and `709 patents. See generally Ethos' Reply Markman Br. [Doc. No. 273] ("Pl. Rep."); RealNetworks' Reply Br. on Claim Construction [Doc. No. 275] ("Def. Rep."). At the Markman hearing, held on January 18, 2006, the parties informed the Court that they agreed to the construction of two of the ten terms — "application protocol" and "transfer protocol." Markman Hr'g, January 18, 2006, Tr. ("Markman Tr.") 3:14-4:2. In addition, the Court declined to construe the term "testing for the existence of a viable connection," considering the parties' constructions so similar that defining the term was unnecessary. See Vivid Technologies, Inc. v. American Science Eng'g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999) (ruling that the Court is required to construe "only those terms . . . that are in controversy, and only to the extent necessary to resolve the controversy").
The parties agreed to construe "application protocol" to mean "[t]he rules or guidelines that control communication by an application (such as a file transfer program). Examples include FTP, file transfer protocol, and HTTP, hypertext transfer protocol." Markman Tr. 3:17-21. The parties agreed to construe "transport protocol" to mean "[t]he rules or guidelines that control the transport of data. Examples include transport control protocol, TCP, and user datagram protocol, UDP." Id. 3:24-4:2.
Ethos construed the term to mean "checking to see if a viable connection exists." Pl. Rep. at 9. RealNetworks construed the term to mean "[d]etermining whether or not a working connection exists." Def. Rep. at 11.
C. The Claims
Claim 7 of the `892 patent reads:
7. A process for downloading a data file from a server computer to a client computer, wherein the data file is stored on a computer readable medium connected to the server computer, wherein the process has an application protocol for sending data as a stream of data using a transport protocol over a computer network connecting the client computer to the server computer, comprising the steps, performed by the client computer, of:
initiating a download of the data file as a stream of data by sending, using the application protocol, a request for the data file to the server computer over the computer network;
monitoring data successfully received by the client computer from the server computer, in response to the request;
storing said data successfully received by the client computer;
detecting a termination of the stream of data being downloaded, where the termination may occur anywhere within said stream of data; and in response to the termination of the download of the data file, testing for the existence of a viable connection from the client computer to the computer network and automatically restarting the download of the data file by initiating a second download of the data file by
sending, using the application protocol, a second request for the data file to the server computer, wherein the request specifies the amount of data successfully received.
`892 Patent, claim 7, col. 13 ll. 65-68, col. 14 ll. 1-26.
Claim 8 of the `892 patent contains similar language as found in claim 7 with the additional reference to a "client program"
8. A process for downloading a data file . . . wherein the process sends data over a computer network connecting the client computer to the server computer, comprising the steps, performed by the client computer, of sending a request for a client program to the server computer over the computer network. . . .
`892 Patent, claim 8, col. 14 ll. 27-35.
Claims 9-12 are dependent claims with no terms to be construed.
Claim 1 of the `709 patent, while containing much of the same language as claim 7 of the `892 patent, contains the additional claimed language:
1. A process for downloading a data file, said data file defining a start of file and an end of file, from a server computer to a client computer . . . comprising the steps, performed by the client computer, of:
. . .
storing said data successfully received by the client computer; detecting a termination of the stream of data being downloaded, where the termination may occur anywhere within said stream of data; and in response to the termination of the download of the data file, automatically restarting the download of the data file by initiating a second download of the data file by sending, using the application protocol, a second request for the data file to the server computer, wherein said download starts at any point within the data file, exclusive of the start of file and end of file, and regardless of the presence of markers, frame boundaries, and other boundaries that may occur within said data.
`709 patent, claim 1, col. 11 ll. 40-48, col. 12 ll. 9-22.
Claim 2 of the `709 patent reads:
2. A process for downloading a first file from a server computer to a client computer over a network connecting the client computer to the server computer, comprising the steps of:
sending a request for said first file to the server computer over the network;
receiving from the server computer, in response to the request for said first file, a client program which when executed by said client computer, performs the step of:
initiating a download of said first file by sending a request for said first file to the server computer over the computer network.
`709 patent, claim 2, col. 12 ll. 23-34.
Claims 4 and 5 of the `709 patent required no construction.
The Court now explains its reasoning for construing the nine remaining disputed terms in the manner that it did at the hearing.
1. A process for downloading a data file from a server computer to a client computer
The Court's Construction: Transferring a file capable of being stored from a server computer to a client computer. Markman Tr. 5:12-13.
Ethos found the Court's construction "an appropriate definition." Markman Tr. at 5:9-17. RealNetworks, however, objected to the phrase "capable of being stored," as inconsistent with the term as used in the claim and specification. Id. at 6:11-7:5.
The term "a process for downloading" is common to claims 7 and 8 of the `892 patent and claims 1 and 2 of the `709 patent. The specifications of both patents contain a definition of the term "download" in the following sentence: "In most computer networks it is desirable to have the capability to download, i.e.,transfer, data from one computer to another." `892 patent col. 1 ll. 12-14 (emphasis added); `709 patent col. 1 ll. 19-21. The patents, therefore, appear to equate "downloading" with the transfer of data. Such a definition, however, would encompass every instance of data retrieval on the Internet, including simply clicking on a hypertext link. The specification of the patents eschew such a broad construction of the term by distinguishing between "downloading" and a more basic transfer of data: "Several client programs are available, such as browsers for the World Wide Web which may download files or read and display HTML documents. . . ." `892 patent col. 5 ll. 60-62 (emphasis added); `709 patent col. 6 ll. 1-3. By contrasting "download" and the "read and display" functions of browsers, the patent specifications exclude that function from its characterization of the term "download." Limiting the construction of "download" to "data capable of being stored" is also consistent with the expressed field of the invention as identified by the specifications, which contemplate the storage of the data transferred. Both patents state: "The present invention is related to processes and apparatus for transferring information between computers, particularly between a client computer and a server computer where the information is stored." `892 patent col. 1 ll. 6-9; `709 patent col. 1 ll. 13-16.
The specifications also make several references to the storage of data. See Pl. Br. [Doc. No. 264], Ex. F (citing the numerous references to the storage of data). RealNetworks contends that these references relate only to those claims which contain an additional storage step, citing, by way of example, claim 2 of the `709 patent, which does not have a separate storage step. Markman Tr. 9:1-7. Because it is unclear whether the references to storage relate only to the claims that contain a storage step, the Court does not rely on these references in making its determination. The Court noted that the specifications of the patents describe embodiments of the invention where the executed client program stores data, see `892 patent col. 3 ll. 8-9, col. 6 ll. 29-30, col. 10 ll. 3-4; `709 patent col. 3 ll. 15-16, col. 6 ll. 37-38, col. 10 ll. 11-12, and that claim 8 of the `892 patent and claim 2 of the `709 patent teach the download process involving the execution of a client program,see `892 patent, claim 8, col. 14 ll. 34-38, `709 patent, claim 2, col. 12 ll. 29-31. It is likely, therefore, that some of the "storage" references in the specifications relate to those claims that do not teach a storage step.
2. Stream of data
The Court's Construction: A stream of raw bytes or words with no used marking or framing for sending included by the application program. Markman Tr. 11:12-14.
This construction was taken directly from the prosecution history. In distinguishing between this patent and prior art, the prosecution history states:
The Miller invention [(prior art)] sends a data file as a stream of frames, responds to ACK's and NAK's of frames/blocks received, and may only send entire intact FRAMES to the client. Thus the Miller invention may only restart a terminated download at the boundaries of frames/blocks, not at any point between these boundaries as is true with the present invention. Claim 47 specifically excludes Miller's stream of a file of frames, and specifically emphasizes that the resumption of the terminated transfer starts at any point within the data file, exclusive of the start of file and end of file, regardless of the presence of spurious markers, frame boundaries, and other boundaries that may occur within the data file. These limitations are included to specifically distinguish the Wanderscheid/Miller combination. . . . In effect the "stream of data" is now specifically defined as a stream of raw bytes or words with no used marking or framing for sending included by the application program.
Claim 47 of the `892 patent ultimately issued as claim 1 of the `709 patent. Pl. Br. at 18.
RealNetworks Initial Markman Br. [Doc. No. 267] ("Def. Br."), Ex. 8, Advisory Action, 11/99 ("11/99 Action") at 9 (emphasis added).
Ethos argued that the Court's construction was incorrect only as to the use of the word "used," which Ethos asserted was a typo in the prosecution history and that "user" should replace the word "used." Markman Tr. 11:20-23; 12:13-16. The Court had no evidence suggesting that this was, in fact, a typo.
RealNetworks argued for "stream of data" to be construed as "[a] sequence of generally undifferentiated bytes transferred with little or no formatting or processing, which can include a plurality of markers or frames." Def. Rep. at 6. RealNetworks relied in part on a sentence in the 11/99 Advisory Action in the paragraph preceding the quoted language above, which states: "The Examiner has rightly interpreted the `stream of data' as a file that may contain a plurality of frames." Def. Br. at 9 (quoting 11/99 Action at 9); Markman Tr. 14:2-16. At the same time, however, RealNetworks challenged the Court's reliance on a subsequent passage in the same document, expressing doubt that a patent applicant can be his own lexicographer "in a prosecution history setting as opposed to the specification." Markman Tr. 13:22-25.
Although the prosecution history is less useful for claim construction than the specification, the prosecution history can "often inform the meaning of the claim language by demonstrating how the inventor understood the invention and whether the inventor limited the invention in the course of prosecution."Phillips, 415 F.3d at 1317; Vitronics, 90 F.3d at 1582-83. The language relied on by the Court, though it purports to define "stream of data," actually limits the scope of that term so as to disclaim interpretations that would encroach on prior art. See Chimie v. PPG Industries, Inc., 402 F.3d 1371, 1384 (Fed. Cir. 2005) (explaining that "[t]he purpose of consulting the prosecution history in construing a claim is to exclude any interpretation that was disclaimed during prosecution" (quotation marks omitted)). The phrase relied on in the prosecution history by RealNetworks does not disclaim or limit the scope of the invention and, in fact, appears to broaden its scope, with the use of the permissive auxiliary verb "may."
3. Computer network
The Court's Construction: An Ethernet network, asynchronous transfer mode network, or any other kind of network using two or more computers, including the internet. Markman Tr. 28:22-24. RealNetworks agreed with the Court's construction of this term.Id. 20:12-14. Ethos objected to defining "network" as potentially being comprised of only two computers. According to Ethos, "[t]wo computers connected together by a wire don't constitute a network in the sense of the `892 [patent] or [the] `709 patent[;] . . . [t]he network is something different from either the client computer or the server computer." Id. 19:1-8.
The specification states:
The computer network may be an ethernet, an asynchronous transfer mode network, or any other kind of network.
`892 patent col. 5 ll. 2-4; `709 patent col. 5 ll. 10-12 (emphasis added).
Ethos admits that "[t]here's no question," that two or more computers connected together are "sometimes loosely referred to as a network in the technology." Markman Tr. 18:21-24. The question then is whether the intrinsic evidence revealed that the scope of this term as commonly understood was narrowed in the context of this invention. Ethos argued that the claim language suggests such a narrowing, by teaching the sending of data "over a computer network connecting the client computer to the server computer." Id. 19:4-5 (quoting `892 patent, claim 7, col. 14 ll. 2-4). Ethos asserted that this language requires the court to construe the term "computer network" as "something different from either the client computer or the server computer[,] [s]o that you have a client computer, you have a server computer, and a network in between connecting them." Markman Tr. 19:7-10. The Court did not read this language as requiring such an interpretation. The language relied on by Ethos is altogether ambiguous as to whether the computer network must be independent from the server and client computer. The Court, therefore, rejected Ethos' argument and adopted the commonly understood meaning of the term.
4. Request
The Court's Construction: An act or instance of asking for something. In the context of misdirection, a request is made by the client computer. Markman Tr. 29:18-20.
Both parties agreed with the Court's construction of the term "request." Id. 29:21-30:9. The Court generally adopted the ordinary meaning of the term in its noun form — the form in which it is used in the patents. See Webster's II New College Dictionary 964 (3d ed. 2005) (defining request as "[t]he act of asking"). As to claim 2 of the `709 patent, which teaches a "misdirection" (requesting one object, but receiving another), the prosecution history specifically identified a "request" as an action by the client computer, and not the action of, for example, a human user clicking on a hypertext link to request a file.
In regard to the language that is now claim 2 of patent `709, the prosecution history stated:
It would have been obvious to a person skilled in the art at the time the invention was made to include the step of receiving the client program in response to a request for downloading the data file into the process taught by Wanderscheid, Miller and Taylor [(prior art)] above. This would have been obvious because Taylor clearly teaches that a client can request a download from the server when certain programs and data files are needed, which would suggest to a person skilled in the art that if the client program is needed in order to perform the downloading process then the client program would automatically be sent to the client in response to the request for downloading the data file.Def. Br., Ex. 7, Office Action Summary 5/26/99 ("5/99 Action") at 8 (citation omitted) (emphasis added).
The parent claim of new claim 24 was rejected by the Examiner during the prosecution of the parent application. However, the misdirection, as now claimed in new claim 24, where the client computer requests a data file and is sent instead a client program, is not suggested in the cited prior art.
Def. Br., Ex. 11, 3/28/00 Action at 5 (emphasis added).
Based on the prosecution history of the `709 patent, the Court added that in the context of misdirection, the request is made by the client computer.
5. Data Successfully Received
The Court's Construction: Data from the data file transferred from the server computer to the client computer. Markman Tr. 30:13-15.
RealNetworks did not raise an objection to the Court's construction of this term. Ethos argued "data is successfully received when it is available for storage." Id. 31:14-15. This construction was not supported by the specifications.
At the Markman hearing the Court construed this term as part of its construction of the entire phrase "storing said data successfully received." Markman Tr. 31:22. RealNetworks objected to the Court's construction of "storing" but not the Court's construction of "data successfully received." Id. 32:16-23.
The specifications use this language in the following manner:
Data received from the server computer as a stream of data in response to the request for the data file is monitored to detect termination of the download of the data file and to track an amount of data from the data file successfully received by the client. The amount of data successfully received by the client is stored when an error is detected that terminates the download.
`892 patent col. 3 ll. 43-51; `709 patent col. 3 ll. 51-58.
In one embodiment, the download is monitored by tracking an amount of data from the data file successfully received by the client. The amount of data successfully received is stored when an error is detected that terminates the download.
`892 patent col. 4 ll. 19-23; `709 patent col. 4 ll. 26-30.
The patent specifications indicate that successful receipt of data is equivalent to the successful transfer of data, whether or not it is available for storage. According to the embodiments described, data successfully received is stored only after an error is detected, indicating that data is still considered successfully received even if it is not available for storage because an error had not occurred. Ethos' construction is therefore inconsistent with the specifications' description of the embodiments of the invention and must be rejected.
6. Storing
The Court's Construction: Data written to a nonvolatile form of storage medium (nonvolatile means the data is not lost when the circuit is broken). Markman Tr. 48:9-11; 60:23-61:1, 63:9-10.
Ethos accepted the Court's construction of this term.Id. 48:12. RealNetworks argued that the term "storage" means placing data in either volatile memory (not capable of being retained when the power is turned off) or nonvolatile memory (capable of being retained when the power is turned off). Def. Rep. at 9; see Markman Tr. 32:23-34:25. At the outset, the language of the claims belies RealNetworks' position. Claim 7 of the `892 patent and claim 1 of the `709 patent refer to "storing said data successfully received by the client computer." `892 patent, claim 7, col. 14 ll. 13-14; `709 patent, claim 1, col. 12 ll. 9-10. For data to be received by or transferred to a computer, at a minimum it must be placed in a computer's volatile memory. As Ethos noted at the hearing, all data received by the computer "comes initially into volatile memory." Markman Tr. 39:10-11. RealNetworks, in arguing for its construction of "downloading" as "transferring," suggested that a download of data only involved a transfer into a computer's volatile memory. Def. Rep. at 5 ("To Ethos, downloading must include storing in nonvolatile memory, otherwise the file may not be `retained' after the transfer is complete. . . . Computers, of course, use both volatile and nonvolatile memory. Ethos ignores the fact that files can be transferred without being stored. . . ."). As such, the "storage" of such data successfully received must do something beyond placing data — or leaving data — in the computer's volatile memory.
"The amount of data successfully received by the client isstored when an error is detected that terminates the download." `892 patent col. 3 ll. 48-50 (emphases added).
The specifications support the construction of storage as referring only to nonvolatile memory. In both patents, Figure 1 purports to illustrate "a server computer and a client computer which improves the likelihood of a successful download of information." See `892 patent fig. 1 col. 5 ll. 18-20; `709 patent fig. 1 col. 5 ll. 26-27. The figure has a box labeled "client computer" within which there is a smaller box labeled "executed client program." `892 patent fig.1; `709 patent fig.1. Below, but separate from the box representing the client computer, is a cylindrical figure — labeled number 24 — connected to the client computer by a line. `892 patent fig.1; `709 patent fig.1. The specifications discuss figure 24 in one embodiment, stating: "This process monitors the data as it is received and while it is stored on the storage medium 24." `892 patent col. 6 ll. 29-30; `709 patent col. 6 ll. 37-38. This suggests storage removed from the typical workings of the client computer. The server computer side of the diagram has the same cylindrical figure — labeled number 20 and identified as "client program and other files." `892 patent fig.1; `709 patent fig.1. The specifications, in discussing the invention generally, refer to figure 20 as a "storage device." `892 patent col. 5 l. 67, col.6 l. 1; `709 patent col. 6 l. 25. It would seem reasonable to conclude that the same cylindrical figure on the client side is the storage device where the data file is stored after being downloaded.
RealNetworks pointed to the following statement in the specifications to support its construction: "It should also be understood that the invention is not limited to a particular memory system." Def. Rep. at 10 (quoting `892 patent col. 8 ll. 2-3); Markman Tr. 42:20-43:8. This, however, does not appear to refer to the type of memory — nonvolatile versus volatile. In describing the makeup of a computer system, the specifications identify the memory system as one of the components of the computer system. `892 patent fig. 2 col. 7 ll. 6-10; `709 patent fig. 2 col. 7 ll. 14-19. The specifications go on to identify the components of that memory system as typically consisting of a "nonvolatile recording medium" and a "volatile, random access memory such as dynamic random access memory (DRAM) or static memory (SRAM)." See `892 patent fig. 3 col. 7 ll. 46-61; `709 patent fig. 3 col. 7 ll. 54-67, col.8, ll. 1-2. The Court concluded, therefore, that the statement — "the invention is not limited to a particular memory system" — does not relate to the different types of memory available within a memory system, volatile and nonvolatile.
7. Client program
The Court's Construction: A computer program that is executed by the client computer. Markman Tr. 51:24-25; 53:25-54:1; 55:4-6.
Both parties objected to the Court's construction. Ethos argued that the construction of this term should include an explanation that "the client program includes a process that monitors the download." Id. 52:2-7; see also Pl. Br. at 10. In support, Ethos cited the specification of the `892 patent which, in describing an embodiment of the invention, states: "[t]he executed client program also includes a process for monitoring the download operation." Pl. Rep. at 10 (quoting `892 patent col. 6 ll. 27-28); see also `709 patent col. 6 ll. 35-36. This language, however, refers to the client program as executed. Turning again to Figure 1, the diagram shows the "client program and other files" on the side of the server computer and an "executed client program" on the side of the client computer. `892 patent fig.1; `709 patent fig.1. Within the square labeled "executed client program" is a smaller square labeled "download monitoring process." `892 patent fig.1; `709 patent fig.1. This diagram establishes that the download monitoring process is associated with the "executed client program," as the specifications state, and not with the dormant "client program." The Court thus properly excluded Ethos' proposed addition from its definition.
RealNetworks wished to add that the computer program can also be compiled and then executed on the client computer. See Markman Tr. 54:6-10. In support of this construction, RealNetworks relied on language in the specification of the `892 patent stating that the client program may be "compiled for execution on the desired platform, such as a machine with a Windows95 operating system." Def. Br. at 14 (quoting `892 patent col. 8 ll. 10-14); see also `709 patent col. 8 ll. 18-22. The Court found this language ambiguous and unable to support RealNetworks' construction.
The remaining two terms are found only in the `709 patent. Both terms relate to the sending of a second request for a data file to the server computer after the termination of the initial download. `709 patent, claim 1, col. 12 ll. 17-21.
8. Wherein said download starts at any point within the data file, exclusive of the start of file and end of file
The Court's Construction: The download cannot resume at the "start of file" or "end of file" indicators/markers. Markman Tr. 55:25-56:2.
Ethos argued that the Court ought adopt the construction: "[t]he download need not resume at the start of file or end of file." Markman Tr. 56:7-25. RealNetworks agreed with the Court's construction and challenged Ethos' construction as contrary to the plain meaning of the term "exclusive of." Id., 57:12-58:11; Def. Br. at 14-15. Ethos contended that "exclusive of" means "not limited to." Pl. Br. at 19. In support of this construction, Ethos referred to an ad for a New Year's Eve dinner at the Ritz Carlton Hotel which states the rate as "$95 per guest, exclusive of tax and gratuity," and argued that the use of the term in the ad clearly does not suggest that taxes and gratuities will not be added or expected. Id.; see also Markman Tr. 58:24-59:20 (making a similar argument).
The plain meaning of the phrase "exclusive of" is "[n]ot including: besides." Webster's II New College Dictionary 399 (3d ed. 2005). There is no indication in the specification or the prosecution history that the plain meaning of this term was altered as Ethos suggested. The prosecution history confirmed the propriety of adopting the commonly understood meaning of "exclusive of."
The prosecution history, in distinguishing this invention from prior art, states:
Thus the Miller invention [(the prior art)] may only restart a terminated download at the boundaries of frames/blocks, not at any point between these boundaries as is true with the present invention [this language is cited by Ethos]. Claim 47 specifically excludes Miller's stream of a file of frames, and specifically emphasizes that the resumption of the terminated transfer starts at any point within the data file, exclusive of the start of file and end of file, regardless of the presence of spurious markers, frame boundaries, and other boundaries that may occur within the data file.
. . . When an unintended termination of transmission occurs, the client may indicate to the server to resume the transfer starting at any point within the data file. The client may indicate to the server the last byte or word successfully received, or the client may indicate to the server from which byte to resume the download. . . . "Start of file" and "end of file" indicators are now included for the data file being downloaded. These indicators are included in new claim 47 so that the claim language can specifically exclude those indicators as locations from which the resumption of the download can occur. This is to specifically distinguish prior art, e.g. the Miller reference.
Def. Br., Ex. 8, 11/99 Action at 9-10 (emphasis added); see Pl. Br., Ex. 11; 3/00 Action at 6 (containing similar language).
It is clear from the prosecution history that the inventors intended for that claim specifically to exclude the start of file and end of file indicators as restart locations.
9. Regardless of the presence of markers, frame boundaries, and other boundaries that may occur within said data
The Court's Construction: The download restarts at any point within the data file heedless of markers, frame boundaries, or other boundaries that may occur within the data. Markman Tr. 56:2-5.
Neither party objected to the Court's construction of this phrase. Markman Tr. 56:7-8; 57:12-13. The Court simply adopted the ordinary meaning of the term "regardless of." See Webster's II New College Dictionary 955 (3d ed. 2005) (defining "regardless of" as "[i]n spite of").
III. CONCLUSION
Having ruled on the claim construction of the disputed terms, the Court conducted the trial and all other proceedings in this action consistent with the claims as defined above.