Friskit, Inc. v. Real Networks, Inc.

BRYSON, Circuit Judge. Friskit, Inc., owns several patents that are directed to the search, retrieval, and playback of multimedia files from a computer network such as the Internet. The overlapping specifications of the patents disclose a system for delivering media content in which a server enables a search for network-accessible media files, creates a playlist from the search results, and causes a media player on the user’s computer to *612 play the files on the playlist sequentially. The use of server-side control of the search and playback functions has application in the field of on-demand digital media services in which users have the ability to access, but not to download, networked media content. Real Networks, Inc., (“Real”) was conceived as a developer of digital media software. After acquiring Listen.com in 2003, however, Real expanded its offerings to include subscriptions to online databases of digital music and video files. Friskit subsequently brought an action in the United States District Court for the Northern District of California, alleging that Real’s multimedia players and subscription services infringed several of Friskit’s patents. 1 Following discovery, Real moved for summary judgment that the asserted claims were invalid for obviousness. Although the district court initially denied the motion, Real later renewed its motion in light of the Supreme Court’s intervening decision in KSR International Co. v. Teleflex Inc., 550 U.S. 398, 127 S.Ct. 1727, 167 L.Ed.2d 705 (2007). The trial court granted the renewed motion based in part on KSR. We affirm. I Friskit challenges the district court’s judgment that claims 35 and 52 of the '467 patent, claims 6 and 16 of the '275 patent, and claim 12 of the '628 patent would have been obvious in view of the prior art. Friskit’s principal argument is that the district court overlooked disputed issues of material fact concerning the content of the prior art. In particular, Friskit argues that three limitations of the asserted claims were absent from the prior art: (1) “programmatic control” of the media player by the server; (2) “direct control” of the media player by the search module; and (3) an integrated client module that controls the media player and includes a user interface to receive search requests. Friskit also contends that the district court failed to address evidence of secondary considerations indicative of nonobviousness. A Friskit contends that the district court erred by failing to construe the term “programmatic control” in claims 35 and 52 of the '467 patent and claim 12 of the '628 patent. Under the proper construction, Friskit argues, the prior art does not teach that limitation. We need not address Friskit’s claim that the case should be remanded to the district court to construe the term “programmatic control,” because we conclude that the prior art technology known as IUMA Radio satisfies that limitation even under Friskit’s definition of the term. According to Friskit, “programmatic control” entails the communication of commands “from one module to another to be executed by the other module in order to control media playback.” The specification of the '467 patent describes how the server module controls the media player by passing instructions and data to the application program interface of the media playback component. One example of a set of instructions and data that might be sent from the server module in that fashion is “play(URL),” which instructs the media player to play the file located on the network at the corresponding uniform resource locator (“URL”). Friskit insists that a “programmatic control” mechanism does not employ the operating system of the user’s computer, because a direct “con *613 duit” is created between the server module and the media player. Real introduced unrebutted evidence that HTML and Javascript commands sent from a server for the internet service IUMA Radio would programmatically control the playback of network-accessible music files. IUMA.com was an online music portal that allowed users to browse, search, and download digital music posted to the website by unsigned artists. The IUMA Radio feature of the website used HTML and Javascript to launch a small browser window containing an embedded media player and a drop-down menu, from which the user would select a particular genre of music. Once the user picked a genre, the server would select a music file of that genre and send the corresponding URL link to the IUMA Radio browser for playback. At the conclusion of each song, the IUMA Radio browser would query the server for more music of the specified genre, and the server would select another music file and return the URL link for that file to the browser. The IUMA browser provided the URLs to the embedded media player, a RealPlayer Internet Explorer plug-in, and it controlled the media player using its application programming interface. The only difference between the method of controlling the media player in the IUMA Radio system and the method described in the specification of the '467 patent was the source of each playback instruction associated with a particular URL. In the '467 patent, the server is responsible for sending a playback command to the media player along with each URL link; in the IUMA Radio system, once the server opened a local browser window, the browser controlled the embedded media player to play back the content from each successive URL passed down from the server. That distinction is immaterial for present purposes because Frisk-it’s proposed construction of “programmatic control” does not require the continuous transmission of instructions from the server. The step of downloading a “program” in the form of HTML and Javascript code necessarily entails the initial transmission of a set of commands executable by the embedded media player. Real argues that the “programmatic control” limitation of the claims was also taught by prior art media players such as Winamp, but Friskit offered evidence that those prior art media players did not teach that limitation. Real’s evidence showed that when a user navigated a web browser to a music portal, such as mp3.com or IUMA.com, and clicked on a link for streaming media, the default media player associated with the web browser, such as Winamp, would automatically play back music in the order designated by the website’s servers. According to Real, that operation reflected “programmatic control” of the media player by a network server. However, Friskit’s expert, Ken Tola, described the playback operation in greater detail, pointing out that the act of clicking on a link to streaming audio resulted in an M3U file being downloaded to the user’s computer. Mr. Tola insisted that the network server in that arrangement was passive because the M3U file specified only the network locations of the audio files to be played and did not provide the “commands” or processing instructions necessary for their playback. Although Real disputes that characterization of M3U files, Mr. Tola’s expert opinion creates a factual issue pertaining to the nature of the information communicated to the media player when it is used to play streaming media content. In addition to pointing to expert testimony relating to streaming media applications, Friskit suggests that Real conceded *614 the absence of the “programmatic control” limitation in the prior art. Real’s expert, Julius Smith, stated that “[i]t is almost unheard of that the server would control the client” and that he could “only think of one example and that would be what’s disclosed in the patents.” However, the context of Dr. Smith’s statements makes clear that he was applying his own definition of the term “programmatic control” to mean “without user intervention,” a definition that was neither advanced by Friskit nor adopted by the district court. For purposes of his obviousness analysis, Dr. Smith applied the construction proposed by Friskit’s expert and found IUMA Radio to be invalidating prior art. Friskit also argues that one of Real’s witnesses admitted that the Javascript in IUMA Radio was used to “relay[] user, input to the media player within the web browser.” As Real points out, however, nothing in the claims or the specification precludes some measure of user control over playback, as long as the server also “programmatically controls” the media player. Indeed, the specification of the '467 patent describes a user interface or terminal that can be used to “control playback settings such as volume, pause, seek and retrieve additional media clip information.” B Claims 6 and 16 of the '275 patent recite “a media player directly controllable by the search module” to receive media resource locators. Friskit challenges the district court’s conclusion that the “direct control” limitation was disclosed by the use of the mini-browser feature in Winamp. There is no serious dispute that, as Friskit suggests, the “directly controllable” limitation is best interpreted to mean that the search module must control the media player “without intervention from other modules or components.” The prior art Winamp program included an integrated “mini-browser” designed to search for and retrieve music files from online music portals, such as mpB.com and IUMA.com. The parties are in agreement that, when properly configured, Winamp was capable of automatically constructing a playlist from any links selected by the mini-browser. Friskit’s expert contended, however, that Winamp could not independently stream audio files without the assistance of the operating system on the user’s computer. He explained that the operating system was responsible for communicating the location of the downloaded M3U file to Winamp and that, if certain file-type associations were disabled in the operating system, Winamp would be incapable of playing back the search results. Real’s expert expressed the contrary opinion that the operating system serves merely as a relay without affecting the server’s control over media playback. The district court accepted Real’s argument because Winamp practiced the same drag-and-drop method that, according to Friskit, infringed the “direct control” limitation. The evidence of Winamp’s reliance on the operating system, however, calls into question whether the control exercised by Winamp’s search module was sufficiently “direct” to satisfy the “direct control” limitation. For that reason, we agree with Friskit that the district court erred in concluding that there was no genuine dispute as to whether the “direct control” limitation was found in the prior art. Accordingly, we disagree with that portion of the district court’s summary judgment analysis. C Friskit next argues that the prior art cited by Real does not disclose the critical element in claim 12 of the '678 *615 patent, “a client module being executable to (i) provide one or more interfaces ... to receive a search request; and (ii) control a media player.” The Winamp interface includes both user controls for the media player (e.g., play, stop, and pause), which satisfy the “executable to ... control a media player” limitation, and a mini-browser pane, which may be used to search websites such as mp3.com or IUMA.com. Indeed, the mini-browser standing alone qualifies as a “client module” because it is operable to search for media files as well as to play back files from any links selected in the browser window. Friskit insists, however, that the mini-browser cannot “control” the media player. It bases that contention on its expert’s assertion that the operating system of the user’s computer acts as an intermediary between the media player and the mini-browser. While that assertion is relevant to the “direct control” limitation, claim 12 requires only that that the “client module ... control a media player”; it does not require “direct control.” In order to give effect to the adjective “direct,” the use of the noun “control” by itself must encompass indirect control of the type that the mini-browser exercises over the media player. We therefore agree with the district court that the contested limitation was taught by Winamp. II Having addressed these preliminary points, we turn to the ultimate question whether Friskit’s invention would have been obvious to one of ordinary skill in the art. Given the insignificant differences between the prior art media players and the claims, the advanced state of the art at the time, and the lack of persuasive secondary evidence to justify a contrary result, we affirm the district court’s summary judgment that the claims at issue would have been obvious at the time of the invention. A The asserted claims describe a system for delivering streaming media content that permits users to search and play back online media content. All of the essential components of the claimed invention — searching archived digital media files, building playlists from the search results, and sequentially playing back the playlist — pre-dated the patents-in-suit, but were limited in their application to streaming media. At the most rudimentary level, websites such as mp3.com and IUMA.com offered users the ability to search online databases of music and download any of the search results. If a default media player was designated, simply clicking on the search results in the user’s web browser would download the music and create a playlist. Some media players, such as Winamp, offered their own integrated mini-browsers, giving users a single interface with which to navigate to sites such as mp3.com or IUMA.com, create a playlist of desired music files, download the music files, and play them back in their designated order. The inventor of the patents-in-suit proposed to integrate search-and-playbaek capabilities into a streaming media system, which was a new approach as of the priority date of the patents. With the exception of IUMA Radio, streaming media systems in the prior art generally relied on downloaded M3U files to point the media player to the location of the individual media file or stream that was intended for playback. When a user clicked on a link to an M3U file in a standard or mini-browser, the M3U file would be downloaded to the user’s computer. If the media player had been properly configured, the music files located at the network addresses specified by the M3U files would then be automati *616 cally played back in the predetermined sequence. IUMA Radio deviated from that model in that it also automated the process of updating the playlist. Once the user had selected a genre, the local browser window would automatically retrieve more songs and control the media player to play them back. Because the selection and order of the songs was fixed by the server, neither IUMA Radio nor the use of the M3U file format permitted a user to create original playlists from selected media files. Throughout most of the litigation, Frisk-it contended that the inventive step of the patents-in-suit was the combination of preexisting media search and playback technologies to create an on-demand, digital music service. Friskit acknowledged that it did not invent streaming media, playlists or media players, but emphasized that its patents “deliver the glue to put existing technologies together into a single application.” That characterization of the claimed invention did not overcome the showing of obviousness, however, because, as the Supreme Court noted in KSR, the “predictable use of prior art elements according to their established functions” is likely to be within the grasp of one of ordinary skill in the art. KSR, 127 S.Ct. at 1740. By the time of Real’s second motion for summary judgment of invalidity, Friskit’s theory of nonobviousness had shifted. Since then, Friskit has argued that the novel aspect of the claimed invention was the manner in which integration was achieved. According to Friskit, the “fundamentally new feature” of the invention was the “programmatic control” of a target module, such as the media player, through executable code communicated to the target module by a signaling module, such as the network server. Friskit also suggests that “direct control” of the media player by the search module is a critical limitation of the claimed invention. As we have noted, IUMA Radio exhibited “programmatic control.” That is, the server module transmitted a program consisting of Javascript and HTML code, which caused a local browser window to open and controlled the embedded media player. Even if “programmatic control” and “direct control” were not literally present in the IUMA Radio system, the evidence before the district court made it clear that it would have been trivial for one of ordinary skill in the art, starting with the prior art devices, to develop the control mechanisms described in the claims at issue. Although it is true that server-side control, as emphasized in the asserted claims, was contrary to the then-conven.tional computer architectures in which the client controlled the server, server control was plainly the alternative to the client-server model. Moreover, the use of web browsers in combination with media players demonstrated a form of remote operation distinct from client control. And with programming tools such as Java, Javascript, and ActiveX in widespread use, the benefits of network control of local processes and the methods for implementing such control were well known at the time of the invention. Indeed, as early as October 1995, Real touted the expanded functionality of its media player that had been made possible by granting third-party application developers access to the application programming interface. Real specifically noted that media content providers interested in offering “music on demand” over the Internet would be able to “plug their own interface” into the media player, “embed instructions” into the data stream, and allow users “to browse, select and play back audio or audio-based multimedia content on demand, in real time.” Facilitating “programmatic” and “direct” control of the media player by networked software mo *617 dules was the intended purpose of exposing the application programming interface. In light of IUMA Radio and the state of the art at the time of the invention, we agree with the district court that a system employing “programmatic control” and “direct control” of the media player by the network-based server would have been obvious to one of ordinary skill in the art. B Friskit argues that the district court failed to address the objective indicia of nonobviousness, the so-called “secondary considerations.” We have frequently stated that “secondary consideration evidence” such as commercial success, copying, and long-felt need does not necessarily overcome a strong showing of obviousness, as was made in the present case. See Leapfrog Enters., Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed.Cir.2007); Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1372 (Fed.Cir.2007); Ryko Mfg. Co. v. Nu-Star, Inc., 950 F.2d 714, 719-20 (Fed.Cir. 1991). In any event, Friskit’s secondary consideration evidence does not provide persuasive support for the nonobviousness of “programmatic” and “direct” control. With respect to commercial success, although Real frequently advertised the benefits of integrating media search and playback functionality in the accused products, Friskit failed to show that the success of those products was attributable to the subject matter that it contends is nonobvious — programmatic control of the media player by the server module and direct control by the search module. See Asyst Techs., Inc. v. Emtrak, Inc., 544 F.3d 1310, 1316 (Fed.Cir.2008); In re Grasselli, 713 F.2d 731, 743 (Fed.Cir.1983). Friskit’s inability to relate the success of Real’s products to the manner in which integration was accomplished is fatal to its claim that the commercial success of the invention is evidence of nonobviousness. Friskit introduced evidence that Real had copied its products after Real had been shown several live demonstrations of the Friskit products’ ability “to create playlists out of media search results” and to prioritize “catalogs of streaming media such as from Real’s content partners ... within search results.” Copying by the accused infringer, however, has limited probative value in the absence of evidence of failed development efforts by the infringer, B.F. Goodrich Co. v. Aircraft Braking Sys. Corp., 72 F.3d 1577, 1583 (Fed.Cir.1996); Pentec, Inc. v. Graphic Controls Corp., 776 F.2d 309, 317 (Fed.Cir.1985), or of “more compelling objective indicia of other secondary considerations,” Ecolochem, Inc. v. S. Cal. Edison Co., 227 F.3d 1361, 1381 (Fed.Cir.2000) (citing cases). As the district court noted, Friskit failed to make such a showing, and it also failed to introduce sufficient evidence to show that the copied technology fell within the scope of the asserted claims. See Amazon.com, Inc. v. Barnesandnoble.com, Inc., 239 F.3d 1343,1366 (Fed.Cir. 2001); In re GPAC Inc., 57 F.3d 1573, 1580-81 (Fed.Cir.1995). The trial court therefore properly gave little weight to Friskit’s evidence of copying. Finally, Friskit sought to show that there was a long-felt need for the invention and that prior art technology taught away from the claimed method. To do so, Friskit pointed to market forces that allegedly discouraged software developers from creating on-demand media services. Some of the concerns affecting developers of streaming media applications included the risk of infringing copyrights in the downloaded media files, the poor sound quality of streaming media over the dial-up modems used by most internet customers at the time of the invention, and a market *618 focus on promoting consumer demand through increased compatibility. But none of Friskit’s evidence suggested that Frisk-it’s approach presented any technical challenge to one of ordinary skill in the art once market forces had created a demand for integrated, streaming media services. See Orthopedic Equip. Co. v. United States, 702 F.2d 1005, 1013 (Fed.Cir.1983) (“[T]he fact that the two disclosed apparatus would not be combined by businessmen for economic reasons is not the same as saying that it could not be done because skilled persons in the art felt that there was some technological incompatibility that prevented their combination. Only the latter fact is telling on the issue of nonobviousness.”); see also KSR, 127 S.Ct. at 1740 (cautioning against rewarding obvious variations precipitated by “design incentives and other market forces”); Joy Tech., Inc. v. Manbeck, 751 F.Supp. 225, 232 (D.D.C.1990), aff'd, 959 F.2d 226 (Fed. Cir.1992) (experts’ expressed skepticism was entitled to little weight as evidence of nonobviousness because it “was directed to economic and commercial factors, not the technical merit of [the claimed invention]”). Streaming media became increasingly viable when copyright owners recognized that digital media content could be sold and when dial-up modem connections with their bandwidth limitations gave way to broadband connections such as cable-modem and DSL. Friskit may have predicted a business trend that would have proved profitable had its commercial embodiments remained competitive in the marketplace. That, however, is not a sufficient basis to overcome the strong prima facie showing of obviousness that was made in this case. Because we have concluded that the asserted claims are invalid for obviousness, we do not address Real’s arguments that the claims are invalid because the term, “substantially automatic” is fatally indefinite or that Real’s products do not infringe Friskit’s patents. Footnotes [1] The patents at issue in this appeal are U.S. Patent Nos. 6,389,467 ("the '467 patent"); 6,725,275 (“the '275 patent"); and 6,735,628 ("the '628 patent”).