Open Text S.A. v. Box, Inc.

Full title: Open Text, S.A., Plaintiff, v. Box, Inc. and Carahsoft Technology…

Court: United States District Court, N.D. California, San Jose Division.

Date published: Jan 1, 2014

36 F. Supp. 3d 885 (N.D. Cal. 2014)

Opinion

Case No.: 5:13–CV–04910–EJD

2014-01-01

Open Text, S.A., Plaintiff, v. Box, Inc. and Carahsoft Technology Corporation, Defendants.

Brian Jeffrey Eutermoser, Britton Frederic Davis, Peter Joseph Sauer, Sara Jane Radke, Sarah Joann Guske, Thomas Joseph Friel, Jr., Wayne Odis Stacy, Angela Leann Campbell, Brian Jeffrey Eutermoser, Frank V. Pietrantonio, Cooley Godward Kronish LLP, Broomfield, CO, Christopher Charles Campbell, Cooley LLP, Reston, VA, Gregory N. Stillman, Brent Lee Vannorman, Hunton & Williams, Norfolk, VA, Thomas J. Friel, Jr., Cooley Godward Kronish LLP, San Francisco, CA, for Plaintiff. Mark Wayne Wasserman, Richard Cyril Sullivan, Jr., Reed Smith LLP, Falls Church, VA, Stuart A. Shanus, Reed Smith Crosby Heafey LLP, Los Angeles, CA, Adaline J. Hilgard, Doyle Johnson, Jesse Luke Miller, John P. Bovich, Jonah Dylan Mitchell, Kirin Kaur Gill, Paulo L. Sousa, Scott David Baker, William Ross Overend, Reed Smith LLP, San Francisco, CA, Jennifer Depriest, Reed Smith LLP, Chicago, IL, Matthew Robertson Sheldon, Reed Smith LLP, Falls Church, VA, for Defendants.

EDWARD J. DAVILA

Motion denied.

Brian Jeffrey Eutermoser, Britton Frederic Davis, Peter Joseph Sauer, Sara Jane Radke, Sarah Joann Guske, Thomas Joseph Friel, Jr., Wayne Odis Stacy, Angela Leann Campbell, Brian Jeffrey Eutermoser, Frank V. Pietrantonio, Cooley Godward Kronish LLP, Broomfield, CO, Christopher Charles Campbell, Cooley LLP, Reston, VA, Gregory N. Stillman, Brent Lee Vannorman, Hunton & Williams, Norfolk, VA, Thomas J. Friel, Jr., Cooley Godward Kronish LLP, San Francisco, CA, for Plaintiff. Mark Wayne Wasserman, Richard Cyril Sullivan, Jr., Reed Smith LLP, Falls Church, VA, Stuart A. Shanus, Reed Smith Crosby Heafey LLP, Los Angeles, CA, Adaline J. Hilgard, Doyle Johnson, Jesse Luke Miller, John P. Bovich, Jonah Dylan Mitchell, Kirin Kaur Gill, Paulo L. Sousa, Scott David Baker, William Ross Overend, Reed Smith LLP, San Francisco, CA, Jennifer Depriest, Reed Smith LLP, Chicago, IL, Matthew Robertson Sheldon, Reed Smith LLP, Falls Church, VA, for Defendants.

[Re: Docket No. 64]

ORDER DENYING PLAINTIFF'S MOTION FOR PRELIMINARY INJUNCTION

EDWARD J. DAVILA, United States District Judge

Open Text S.A. (“Open Text”) brings this motion for a preliminary injunction to enjoin Box, Inc. (“Box”) from “selling, offering to sell, licensing, offering to license, installing, offering for download, offering for installation, providing, or otherwise making available” any of its products or services containing “Box Edit” or “colorably similar features to customers, potential customers, or entities that have or may have 100 or more users.” See Pl.'s Mot. for Prelim. Inj. (“Pl.'s Mot.”) at 24–25, Docket Item No. 65. This motion is limited to four patents from two patent families. The first family, the “Synchronization Patents,” includes U.S. Patent No. 7,062,515 (“the '>515 patent”) and U.S. Patent No. 8,117,152 (“the '152 patent”). The second family, the “Groupware Patents,” includes U.S. Patent No. 7,287,055 (“the '055 patent”) and U.S. Patent No. 7,299,258 (“the '>258 patent”).

The Court held a hearing on this motion on January 24, 2014. Having considered the parties' submissions, argument, and the relevant law, and for the reasons discussed herein, Open Text's motion to preliminarily enjoin Box is DENIED.

I. Background

A. Factual Background

Open Text is a Luxembourg corporation with its registered address at 29 Boulevard Royal, L–2449 Luxembourg. Complaint at ¶ 6, Docket Item No. 1. Open Text S.A. is a subsidiary of Open Text Corporation, a Canadian corporation with its principal place of business at 275 Frank Tompa Drive, Waterloo, Ontario, Canada. Id. Open Text Corporation distributes software products and provides customer support and professional services through a number of subsidiaries, including Open Text, Inc., which sells Open Text software and services in the United States. Id. at ¶ 7. Open Text Public Sector Solutions, Inc. (“OTPSS”), a Virginia corporation with its headquarters in Arlington, Virginia, is a subsidiary of Open Text, Inc. Id. at ¶ 8. One of Open Text's core markets is Enterprise Content Management (“ECM”), which refers to a variety of solutions for managing business content. Id. at ¶ 10. One such solution provides a repository for electronic documents and allows for functions such as organization, display, classification, access and version control, event auditing, rendition, and search. ECM also includes software tools and services for collaboration, records and email management, and archiving. Id. Open Text's ECM provides the foundation for its offerings in a broader market category known as Enterprise Information Management (“EIM”). EIM encompasses capabilities such as Business Process Management (“BPM”), Customer Experience Management (“CEM”), Information Exchange (“IE”), and Discovery. Open Text offers a range of software products and services in each of these areas. Id. at ¶ 11.

Box, Inc. is a Delaware corporation founded in 2005 with its headquarters at 4440 El Camino Real, Los Altos, California 94022. Id. at ¶ 23. Box exclusively offers “cloud-based” storage and content management products and services. Def.'s Opp'n to Mot. for Prelim. Inj. (“Def.'s Opp'n”) at 4, Docket. Item No. 85. The cloud is often referred to as an “off-premises” solution because the end user does not own or control the servers that are used to deploy the features and functionality. Declaration of Grant Shirk (“Shirk Decl.”) ¶ 4, Docket Item No. 88. Box provides five versions of its software: Box Personal, Box Starter, Box Business, Box Enterprise, and Elite Plan. Pl.'s Mot. at 8, Dkt. No. 65. In October 2012, Box introduced a new access feature for all users called Box Edit. Id. Box Edit allows users to access documents stored in the Box database through a web browser and automatically syncs the locally-edited copy to the Box database. Id. at 10. In April 2013, Box released an “auto-updates” feature for Box Edit, which automatically integrates new product upgrades as they are released. Id.

Open Text argues the Elite Plan directly competes with Open Text's Content Server. Pl.'s Mot. at 9–10, Dkt. No. 65. Box released Box Enterprise in 2007 and the Elite Plan in late August 2013. Shirk Decl. ¶ 11, Dkt. No. 88; Pl.'s Mot. at 10, Dkt. No. 65. Box argues the Elite Plan was not Box's entry point into the market for customers with more than 100 users, it was the Enterprise software. Shirk Decl. ¶ 10, Dkt. No. 88. According to Box, the Elite Plan does not add any new functionality and instead provides Box customers with multiple enterprise ID's (commonly referred to as a “sandbox environment”), Box Premier Support, and the ability to build custom internal applications on the Box platform. Id.

The instant preliminary injunction motion filed September 13, 2013 seeks to enjoin Box from providing its Box Edit feature. Def.'s Opp'n at 9, Dkt. No. 85.

B. Technology Background

Four patents from two patent families are at issue in this motion. The Synchronization Patents include the '515 and '258 patents and relate to cached file synchronization. The '515 patent was filed on December 28, 2001. The '152 patent is a continuation of U.S. Patent No. 7,590,655, filed Jan. 10, 2006, which in turn claims priority to the '515 patent. The Groupware Patents include the '055 and '258 patents and relate to web-based groupware systems. The Groupware Patents were filed on July 22, 2004 and claim priority to a U.S. Patent Application No. 08/955,569, filed on October 22, 1997.

1. Synchronization Patents

The Synchronization Patents purport to invent a software program that allows users to retrieve a file from a remote database, edit the file locally, then save the edited file directly back to the remote database. '515 patent, cl. 1. The software, described as a cache manager, stores the file as a local cached copy and determines if the cached file has been modified based on a notification from the file management system of an operating system. Id. This allows the user flexibility to choose which local application to edit the file. Id. at 3:54–61. Moreover, the use of the cache manager does not require specifically programmed applications to send notifications because the cache manager can seamlessly synchronize the cached file based on a notification from the file management system of the operating system. Id. at 3:62–67. Before the patent was filed, prior systems of document management attempted to address the drawbacks of file synchronization through the use of custom-designed tools for editing, secondary synchronization programs, and operating system-level implementation. Id. at 2:13–3:18.

2. Groupware Patents

The Groupware Patents purport to invent a collaborative workspace accessible through a web browser. '055 patent, cl. 1; '258 patent cl. 1. According to the invention, a primary user selects the parameters of the collaborative workspace. Id. Such parameters include a list of secondary users and the secondary users' ability to interact with the workspace based on varying levels of access to the workspace. Id. In response to the primary user's designations, a computer creates the dedicated network site, or collaborative workspace, on a network-connected server. '055 patent cl. 1, '258 patent, cl. 4. Further, users are able to select the types of user applications to be included in the workspace. '055 patent, cl. 4; '258 patent, cl. 1. The invention purports to overcome the prior art disadvantages of having a system administrator controlling the access levels of the system and the need for specialized software to be installed on users' computers. Id. at 1:52–59.

II. Legal Standard

Congress has authorized district courts in patent cases to grant injunctions “in accordance with the principles of equity to prevent the violation of any right secured by patent, on such terms as the court deems reasonable.” 35 U.S.C. § 283 (2006). It is well-established that a preliminary injunction is an extraordinary remedy “not to be routinely granted” and reserved only for those cases where it is clearly warranted. High Tech Med. Instrumentation , Inc. v. New Image Idus., Inc., 49 F.3d 1551, 1554 (Fed.Cir.1995).

Because this motion arises in the context of a patent infringement action, Federal Circuit law applies. See Hybritech Inc. v. Abbott Labs., 849 F.2d 1446, 1450, n. 12 (Fed.Cir.1988). Under Federal Circuit law, the court considers whether the plaintiff seeking a preliminary injunction can establish that: (1) he is likely to succeed on the merits; (2) he is likely to suffer irreparable harm in the absence of preliminary relief; (3) the balance of equities tips in his favor; and (4) that an injunction is in the public interest. Winter v. Natural Res. Def. Council, Inc., 555 U.S. 7, 20, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008). These traditional four factors “apply with equal force to disputes arising under the Patent Act.” eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388, 391, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006).

The Ninth Circuit has clarified that the elements of a preliminary injunction may be balanced such that “a stronger showing of one element may offset a weaker showing of another.” Alliance for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1131 (9th Cir.2011). For example, “serious questions going to the merits” and a hardship balance that tips sharply toward the plaintiff can support issuance of an injunction, assuming the other two elements of the Winter test are also met. Id. at 1132. Accordingly, this “serious questions” approach survives Winter when applied as part of the four-element Winter test. Id.

III. Discussion

A. Likelihood of Success

In order to demonstrate a likelihood of success on the merits, plaintiff must show that, in light of the presumptions and burdens that will inhere at trial on the merits, (1) plaintiff will likely prove that defendant infringes the asserted patents and (2) plaintiff's infringement claim will likely withstand defendant's challenges to the validity and enforceability of those patents. See Genentech, Inc. v. Novo Nordisk, A/S, 108 F.3d 1361, 1364 (Fed.Cir.1997); Titan Tire Corp. v. Case New Holland, Inc., 566 F.3d 1372, 1376 (Fed.Cir.2009).

A patent claim is invalid by anticipation if the invention “was known or used by others in this country, or patented or described in a printed publication in this or a foreign country” before the filing, 35 U.S.C. § 102(a), or “patented or described in a printed publication in this or a foreign country or in public use or on sale in this country” more than a year before the filing date. 35 U.S.C. § 102(b). Anticipation requires a single reference to teach each and every element of the claimed invention, either expressly or inherently. Perricone v. Medicis Pharmaceutical Corp., 432 F.3d 1368, 1375 (Fed.Cir.2005).

A patent claim is invalid for obviousness “if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art.” 35 U.S.C. § 103(a). Under this flexible inquiry, the court considers the scope and content of the prior art, the differences between the prior art and the asserted claims, the level of ordinary skill in the art, and any secondary considerations of nonobviousness. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 426, 127 S.Ct. 1727, 167 L.Ed.2d 705 (2007). A patent claim is obvious if the combination of familiar elements according to known methods yields only predictable results or offers an obvious solution to a known problem. Id. at 415–421, 127 S.Ct. 1727. The Supreme Court in KSR emphasized the role of “common sense” in determining whether a patent is obvious at the time of invention. Id. at 421, 127 S.Ct. 1727.

Because a patent is presumed valid, an alleged infringer must establish invalidity by clear and convincing evidence at trial. Titan Tire, 566 F.3d at 1376. At the preliminary injunction stage, the accused infringer similarly bears the burden to present evidence of invalidity. Id. However, rather than clear and convincing, the accused infringer need only establish a “substantial question” of invalidity. Amazon.com, Inc. v. Barnesandnoble.com, Inc., 239 F.3d 1343, 1358 (Fed.Cir.2001). Once the alleged infringer has raised a substantial question regarding validity, the plaintiff must bring forth evidence to demonstrate that the alleged infringer's invalidity defense “lacks substantial merit.” Titan Tire, 566 F.3d at 1378. If, after weighing the available evidence for and against validity, the court determines that the plaintiff fails to show the invalidity defense lacks substantial merit, it necessarily follows that the plaintiff cannot show a likelihood of success on the merits. Id. at 1379.

Open Text alleges Box infringes claims 1–5 and 7–10 of the '515 patent, claims 8–10 and 12 of the '152 patent, claims 1, 2, 4, and 5 of the '055 patent, and claims 1–3 and 5 of the '258 patent. Box does not dispute any infringement assertions in the instant motion and instead, contends the asserted claims are all invalid under 35 U.S.C. § 102 for anticipation and/or 35 U.S.C. § 103 for obviousness. For the reasons discussed below, the Court finds Box that has raised substantial questions concerning the validity of OpenText's Synchronization and Groupware patents.

1. The Synchronization Patents

Box cites five prior art references as anticipating and/or rendering obvious claim 1 of the '515 patent and corollary claim 8 of the '152 patent: (1) WS_FTP Pro User's Guide, Ver. 5 (“WS_FTP Pro”), (2) Coda Distributed File System client systems (“Coda”),¹ (3) GroupWise 5.5 User's Guide (“GroupWise 5.5”), (4) U.S. Patent No. 5,805,809 (“the '809 patent”, and (5) U.S. Patent No. 6,119,151 (“the '151 patent”). None of the references were considered by the examiner during the prosecution of the patents. The Court has considered all references and finds that Box raises a substantial question as to the validity of claims 1–5 and 7–10 of the '515 patent and claims 8–10 and 12 of the ' 152 patent in light of WS_FTP Pro in combination with the knowledge of a person of ordinary skill in the art.

1 Box relies on two articles to describe the Coda Distributed File System: (1) “The Coda Distributed File System” by Peter J. Braam in Linux Magazine (June 1998); (2) “Coda File System User and System Administrators Manual” (August, 1997).

The '515 and '152 patent share a specification. Claim 1 of the '515 patent relates to a system for synchronizing a cached file while claim 1 of the '152 patent covers the method for synchronizing a cached file. Claim 1 of the '515 patent is exemplary:

A system for synchronizing a cached file with a database:

a computer processor;

a network connection device operable to establish a connection with a database;

a computer readable memory containing a local cache; and

a software program, executable to run in user space on a client computer, stored on a computer medium and executable by the computer processor to:

send a request to the database for a file;

receive the file at the client computer directly from a database;

store the file as a cached file in the local cache;

notify the operating system to open the cached file using a locally running

application associated with the file type for the cached file;

determine if the cached file at the client computer has been modified by a

user using the locally running application based on a notification

from a file management system of an operating system; and

if the cached file has been modified, save the cached file from the cache

directly to the database.

The parties dispute the following claim limitations: (1) software program to be executable in user space; (2) locally cached file that is opened in a locally running application; and (3) determination of a modified cached file based on a notification from a file management system of an operating system. For the reasons set forth below, Box raises a substantial question of validity to claim 1 of the '515 patent and claim 8 of the '258 patent in light of WS_FTP Pro.

a. WS_FTP Pro

WS_FTP Pro is a computer program based on File Transfer Protocol (“FTP”), a method of intercomputer communication developed for the Internet. WS_FTP Pro User's Guide Software Version 5 (“WS_FTP Pro”) at 1, Ex. F, Docket Item No. 89. FTP is based on the client-server model of communication between computers, whereby one computer runs a server program that other computers or systems can request and receive information that the server program retains. Id. The system running the server program is known as an FTP server (or “remote system”). Id. Through a personal computer, WS_FTP Pro allows a user, the FTP client (or “local system”), to communicate with FTP servers around the world in order to transfer files from an FTP server to the user's local system or from the local system to the FTP server. Id. at 2.

WS_FTP Pro discloses two distinct modes of remote editing in the “classic” user interface: “Remote Edit uses ShellExecute” is checked and “Remote Edit uses ShellExecute” is not checked. Id. at 32–33. When “Remote Edit uses ShellExecute” is checked, WS_FTP Pro permits a user to edit a file in an associated application on the local system then WS_FTP Pro saves the edited file back to the remote system with the same name. Id. When “Remote Edit uses ShellExecute” is not checked, WS_FTP Pro requires the user to download a file from the remote system to the local system. Id. at 33. After the user edits the file on the local system, the user must upload the file to the remote system with a new file name. Id. The “explorer” interface utilizes the same core engine as the classic interface but integrates with Windows Explorer so this version will not be analyzed as invalidating prior art. See id. at 4.

i. User space

As a threshold issue, the parties the dispute location where WS_FTP Pro runs. Open Text argues WS_FTP Pro does not teach any of the asserted claim limitation because WS_FTP Pro does not run in user space. Declaration of Prof. Ketan Mayer–Patel (“Mayer–Patel Decl.”) ¶ 66, Ex. 4, Docket Item No. 107. Open Text claims that the executable file of WS_FTP Pro, “FTP95Pro.exe,” is not listed as a file installed on a user's computer. Id. In contrast, Box points to a screenshot from the WS_FTP Pro manual that lists the files installed on a user's PC. Declaration of Srinivasan Jaggannathan (“Jaggannathan Decl.”) at 8, Ex. M, Docket Item No. 89. The screenshot demonstrates that when the local computer connects to a remote server in the WS_FTP Pro classic interface, the local and remote systems are divided: local directories and files are displayed under the local system while the remote directories and files are shown under the remote system. See WS_FTP Pro at 3, Ex. F, Dkt. No. 89. From this screenshot, “FTP95Pro.exe” is clearly shown as being installed as a file on the local system. This demonstrates to the Court that WS_FTP Pro is likely executable to run in user space on the user's computer, rather than on the remote system.

ii. Locally cached file

Open Text contends that WS_FTP Pro does not “receive the file at the client computer” and “store the file as a cached file in the local cache” because a user edits the file remotely rather than locally. Mayer–Patel Decl. ¶ 65, Ex. 4, Dkt. No. 107 (quoting the '515 patent, cl. 1). Box argues that WS_ FTP Pro copies the file from the remote system to the local system and stores the file in a local cache based on the exact language in the manual. Jaggannathan Decl. at 16, Ex. M, Dkt. No. 89. The Court finds Box's argument persuasive that WS_FTP Pro copies the file from the remote system to the local system for editing.

When “Remote Edit uses ShellExecute” is checked, the user initiates the editing process by selecting a file on the remote system. WS_FTP Pro at 32, Ex. F, Dkt. No. 89. After the user selects the “Edit File” option from the right-mouse menu, “WS_FTP copies the file to the local system and opens it in its associated application.” Id. (emphasis added).

Moreover, that WS_FTP Pro copies the file from the remote system to the local system is further supported by the manual's description of how a user associates an executable file to a local application. The user is able to select the type of application where the user wishes to edit by associating a file of a particular extension with any application capable of opening that file. Id. at 46. For example, if a user associates .pdf with Adobe Acrobat Reader, when the user opens file myfile.pdf, the file is opened in Adobe Acrobat reader. Id. at 32. The manual teaches that to open (execute) a file in an associated application, the user must choose the remote file then select “Exec” on the remote database. Id. Prior to opening the file, “WS_ FTP Pro downloads the file to the Windows temporary directory.” Id. at 32. Thereafter, the file is transferred from the remote system to the local system and executed by the operating system, Windows. Id. at 46. Based on the association settings, WS_FTP Pro notifies Windows through the command, “ShellExecute,” how to execute the file. Id. at 47.

When “Remote Edit uses ShellExecute” is checked, WS_FTP Pro explicitly teaches the use of the same shell command, “ShellExecute,” to open an associated application. Id. at 32. Accordingly, when a user selects “Edit File” when “Remote Edit uses ShellExecute” is checked, WS_FTP Pro operates as if the user selected “Exec” to open a remote file. Thus, when a user selects “Edit File,” WS_FTP Pro transfers the file from the remote system to the local system then notifies Windows through “ShellExecute” to open the locally cached file in its associated application. After Windows operating system opens the locally cached file in the user's chosen associated application, the user is free to edit the file. Id. at 32. Once the user edits the file and chooses “Save” in the application that opened the file, “[t]he file is saved on the remote host with the original name” and the user can exit the application. Id. at 33.

The crux of both parties' argument is how WS_FTP Pro operates based on the phrase “copies the file to the local system.” Open Text bases its entire analysis on the assumption that WS_FTP Pro does not copy the file from the remote system to the local system. Open Text draws support from a portion of the manual that states “[c]heck ‘Remote Edit uses ShellExecute’ to have WS_FTP Pro open a remote file for editing without downloading it to your system.” Mayer–Patel Decl. ¶ 70, Ex. 4, Dkt. No. 107 (quoting WS_FTP Pro at 36, Ex. F, Dkt. No. 89 (emphasis added)). However, rather than looking at the enumerated steps of how WS_FTP Pro operates when “Remote Edit uses ShellExecute” is checked, Open Text incorrectly focuses on this inconsistency without context. This excerpt appears ambiguous at first blush, but the step that WS_FTP Pro “copies the file to the local system” is not. Further, Open Text offers no argument as to what else “copies” means in this context.

While the WS_FTP manual uses the terms “copy,” “transfer,” and “download” interchangeably to denote that a file is moved from the remote system to the local system, an analysis of different parts of the manual disclose a subtle difference that can account for this supposed inconsistency in terminology. Although the “explorer” interface is not analyzed as prior art, a discussion of its operation is helpful here. To review, WS_FTP teaches two modes of remote editing in the “classic” interface: “Remote Edit uses ShellExecute” checked and “Remote Edit uses ShellExecute” not checked. The “explorer” interface functions similarly to the “classic” interface and permits remote editing but does not teach editing through the use of “Remote Edit uses ShellExecute.” Id. at 4. However, the “explorer” interface provides significantly more detail than the “classic” interface about the precise location of the file moved from the remote system to the local system when editing files.

To edit a file on the remote system in the “explorer” interface, a user must locate the file the user wants to edit. Id. at 68. When the user selects “Open” from the File menu, WS_FTP Pro Explorer copies the file to the “Temporary Directory” “with a unique name, and then opens it in the application associated with its file extension.” Id. at 68. The “Temporary Directory” is defined as the folder on the local system where WS_FTP Pro Explorer copies all files the user opens. Id. at 70. After the user edits the file, the user can choose “Save As” from the application that opened the file. Id. at 69. Thereafter, the user must upload the file to the FTP site and can save it with the original name in order to reflect the change to the file in the remote database. Id.

As evidenced, the “explorer” interface parallels the first enumerated steps when “Remote Edit uses ShellExecute” is checked in the “classic interface”: a user selects the file to be edited, WS_FTP Pro (or Explorer) copies the file to the local system and opens it in its associated application. It follows that when “Remote Edit uses ShellExecute” is checked, like the “explorer” interface, the file that is copied to the local system is stored in a temporary directory. Further, this is consistent with the manual's description in the “classic” interface which articulates that WS_FTP Pro downloads the file to the temporary directory before opening the file. See id. at 32. However, after these steps, the “explorer” interface begins to parallel the mode when “Remote Edit uses ShellExecute” is not checked: both require the user to select “Save As” and manually upload the edited file to the remote system.² In contrast, when “Remote Edit uses ShellExecute” is checked, the user need only select “Save” from the file menu. It is common knowledge that the selection of “Save” in a file menu updates the document in use whereas “Save As” creates a new document. Accordingly, when the user selects “Save” from the file menu when “Remote Edit uses ShellExecute” is checked, the same file in the temporary directory is updated. In contrast, when a user selects “Save As” in the “explorer” interface or when “Remote Edit uses ShellExecute” is not checked in the “classic interface,” another file is created on the user's computer separate from the temporary directory folder.

2 The difference in “explorer” interface remote edit and in the “classic” interface when “Remote Edit uses ShellExecute” is not checked is that in the “explorer” interface, the user has permission to edit the file on the remote system. This is why WS_FTP Pro Explorer changes the name of the file when it copies the file from the remote system to the local system: so that the user can upload the new, edited file to the remote system with the original name. Because the user when “Remote Edit uses ShellExecute” is not checked does not have permission to edit the remote file, the user must upload the edited file to the remote system with a different name.

With reference to the above alleged inconsistency by Open Text, the WS_ FTP Pro manual appears to be making a subtle distinction between how the file is stored after being copied to the local system: either as a temporary or permanent copy on the local system. When “Remote Edit uses ShellExecute” is checked, the temporary copy simply temporary and not stored as a permanent copy on the local system. In this mode, the user may edit the remote file without downloading the file permanently to the local system. Accordingly, Box's argument that WS_FTP Pro “copies the file to the local system” by moving the file from the remote system to the local system and storing the file in a local cache is in accord with the WS_FTP Pro manual.

Moreover, Open Text attempts to characterize WS_FTP Pro as a virtual file system to further support its position that WS_FTP Pro does not copy the file from the remote system to the local system. Mayer–Patel Decl. ¶ 65, Ex. 4, Dkt. No. 107. Virtual file systems permit users to edit files directly from the remote database because the files are stored remotely rather than as a local copy of the file. Id. at ¶ 50. Open Text explains that the remote location of the files are mapped as a remote or virtual drive on the local computer so the remote files appear as if they are locally available. Id. at ¶¶ 39, 47. While Open Text alleges WS_FTP Pro is a virtual file system, Open Text offers no support from the manual, only a bare assertion that WS_FTP Pro “allows users to edit documents stored remotely using an ‘Edit’ feature.” Id. at ¶ 65. Without any concrete evidence, the Court cannot accept Open Text's empty allegation that WS_FTP Pro operates contrary to the plain language of the manual.

Open Text had the opportunity to address an alternative meaning to “copies the file to the local system” when “Remote Edit uses ShellExecute” is checked and how the same mode operates as a virtual file system, but fails to provide any compelling evidence to support its position. Accordingly, Open Text fails to rebut Box's argument that WS_FTP Pro copies the file from the remote system to the local system and stores the file in a local cache.

iii. Determination and Notification

Open Text argues that WS_FTP Pro does not determine whether a file has been edited after receiving a notification from a file management system, and further that there is no need for a notification because the file is stored remotely rather than by a local cache. Mayer–Patel ¶ 78, Ex. 4, Dkt. No. 107. Box counters that even if WS_FTP Pro does not disclose the determination and notification claim limitations, it would be obvious to a person of ordinary skill to use their own knowledge to add this feature. Jaggannathan Decl. at 22, Ex. M, Dkt. No. 89. Box contends there is sufficient motivation to include this feature considering the patent specification described a notification from a file management system of an operating system to be “an inherent feature of file management system to determine when the cached file is saved.” Id. (quoting the '515 patent 7:20–23). For the following reasons, it appears to the Court that the claimed limitation “determine if the cached file at the client computer has been modified by a user using the locally running application based on a notification from a file management system of an operating system” is obvious in light of the knowledge of a person skilled in the art.

The claim limitations of the Synchronization Patents require significant operating system involvement: the software program notifies the operating system to open the locally running application associated with the file type for the cached file and in turn, the file management system of the operating system will notify the software program when a user makes a modification in the locally running application. See '515 patent, cl. 1; '152 patent, cl. 8. Based on this notification from the operating system, the cached file is saved directly to the remote database. Id. The significance of the notification coming from the file management system was described during prosecution of the '515 patent: “that the notification can be sent regardless of the application modifying the cached file.” '515 patent file history at 63, Ex. 8, Docket Item No. 89. Rather than requiring specific programming for the local application to notify a cache manager that a file has been saved, the advantage of an operating system notification is highlighted as the software being “agnostic as to the file type” such that the user can “work on the file type with any locally running application that the user wishes to use to modify the particular file.” Id. at 116. WS_FTP Pro, like the claimed invention, notifies the operating system to open the local application associated with the file the user wishes to edit. WS_FTP Pro at 32, Ex. F, Dkt. No. 89. Moreover, WS_FTP saves the edited file back to the remote host once the user has edited the file. Id. at 33. Accordingly, WS_FTP Pro permits the user to select any type of local application the user desires to edit the local file in and WS_FTP Pro will save the modifications back to the remote database regardless of the application used. The singular difference between WS_FTP Pro and the claimed invention is that WS_FTP Pro is silent on how WS_FTP Pro is notified to save the edited file back to the remote database. Because the manual does not expressly disclose that WS_TP Pro receives a notification from a file management system when a file has been edited, WS_FTP Pro likely does not anticipate the Synchronization Patents.

In an obviousness analysis, courts look to the differences between the prior art and the patent claim to determine if the subject matter as a whole is obvious to a person skilled in the art. See KSR, 550 U.S. at 399, 127 S.Ct. 1727. A combination of familiar elements according to known methods is likely to be obvious. Id. at 401, 127 S.Ct. 1727. “To determine whether there was an apparent reason to combine the known elements in a way a patent claims, it will often be necessary to look to ... the background knowledge possessed by a person having ordinary skill in the art.” Id.

Here, WS_FTP Pro teaches each and every element of the claimed invention except that WS_FTP Pro does not disclose how WS_FTP Pro determines a change has been made by the user. Tellingly, the patent specification described the state of the art at the time of the patent filing: that a notification from a file management system is an inherent feature of the well-known Microsoft Windows operating systems. '515 patent 7:20–23. As such, the Court presumes this background knowledge of a person skilled in the art at the time of the patent. And considering that a notification from a file management system is an inherent feature of the Windows operating system, it appears that a notification of a user's modification from the file management system in WS_ FTP Pro is obvious to a person skilled in the art. Moreover, WS_FTP Pro supports Windows 95 and Windows NT. WS_FTP Pro at 1, Ex. F, Dkt. No. 89. A person skilled in the art would recognize that these versions of the Windows operating system support this inherent notification feature, as these operating systems teach a graphical user interface and are executable to open the application associated with a file type. See Jaggannathan Decl. ¶ ¶ 100–101, Dkt. No. 89. Thus, a person skilled in the art would likely know that WS_FTP Pro, utilizing Microsoft as its operating systems, would support the same notification from a file management system of Windows to inform WS_FTP Pro to save the file back to the remote host.

In contrast, when “Remote Edit uses ShellExecute” is not checked, the determination and notification claim limitations would likely not be obvious to a person skilled in the art. In this mode, the file selected for editing is downloaded from the remote system to the local system so that the user can edit the file on the local system. WS_FTP Pro at 33, Ex. F, Dkt. No. 89. Once the user edits the file, the user must save the file with the new name and upload the renamed file to the remote database. Id. It would not be obvious for the file management system to notify WS_FTP Pro that a change has been made in this mode because WS_FTP Pro treats the downloaded documents differently than when “Remote Edit uses ShellExecute” is checked. When “Remote Edit uses ShellExecute” is checked, WS_FTP Pro automatically saves the modified file back to the remote host, so it is logical to infer that WS_FTP Pro is notified by a file management system. However, this logical inference does not extend to the mode when “Remote Edit uses ShellExecute” is not checked because the user makes the modifications and then manually uploads the modified document to the remote database. In this mode, the only notification that WS_ FTP Pro receives is from the user who notifies WS_FTP Pro that the user will upload the modified file back to the remote database.

a. Secondary Considerations

As a check on hindsight when applying the obviousness doctrine, the court looks to secondary considerations, or objective evidence such as copying, long felt but unsolved needs, failures of others, and commercial success. Power Integrations, Inc. v. Fairchild Semiconductor Int'l, Inc., 711 F.3d 1348, 1368 (Fed.Cir.2013) (citing McGinley v. Franklin Sports , Inc., 262 F.3d 1339, 1351 (Fed.Cir.2001). Open Text argues that the commercial success of Box's Box Edit feature sufficiently demonstrates that Box's validity defense lacks substantial merit. Pl.'s Rebuttal Br. in Supp. Of Mot. For Prelim. Inj. (“Pl.'s Rebuttal Br.”) at 8–10, Docket Item No. 107. As evidence of such commercial success, Open Text points to Box's marketing materials referring to Box Edit as “awesome” and one of its most popular applications. Id. Moreover, Open Text alleges Box Edit's commercial success based on its expert's analysis of customer sales. Id. Box counters that Open Text fails to isolate the alleged success to the feature claimed in the Synchronization Patents. Def.'s Opp'n at 18 n.13, Dkt. No. 85.

Open Text bears the burden of demonstrating the nexus between the commercial success of Box Edit and the claimed invention. See Wyers v. Master Lock Co., 616 F.3d 1231, 1246 (Fed.Cir.2010); see also In re Huang, 100 F.3d 135, 140 (Fed.Cir.1996) (holding the proponent must establish sales were a direct result of the claimed features of the invention). Here, Open Text relies solely on Box's sales charts after Box Edit was released but does not establish a direct nexus to the claimed limitations of the Synchronization Patents. Thus, absent any independent evidence to establish a nexus, Open Text fails to undermine Box's obviousness argument. See e.g., Wyers, 616 F.3d at 1246 (rejecting plaintiff's allegation of commercial success based on sales of the accused product for failing to provide independent evidence to establish a nexus to the claimed feature).

b. Box Raises a Substantial Question

In sum, WS_FTP Pro appears to disclose the same solution perceived to be problematic with prior art document management systems: the multitude of file types, the number of different users, and subjective user preferences. Rather than addressing these problems through custom-designed tools, secondary synchronization programs, or operating system level implementation, WS_FTP Pro operates exactly like the Synchronization Patents: a local cached file is received by the user, the user edits the file in a locally associated application, the cache manager determines if the file has been modified based on a notification from file management system of the operating system, and if the file has been modified, the cache manager saves the cached file back to the remote database. See '515 patent, cl. 1. By running in user space and copying the files to a local cache, WS_FTP Pro avoids the additional programming required by operating system-level implementation. Moreover, WS_ FTP Pro permits users to select the type of locally running application to be used once the remote system copies the file down to the user's computer. Consequently, individual users are able to select their editing “tool of choice” without regard to how other users are accessing and editing the same document. See id. at 2:16–20. Additionally, like the claim limitations, a user edits the local file in a locally running application and WS_FTP Pro saves the user's modification back to the remote database without requiring additional synchronization software.

For the above reasons, the Court finds that Box has raised a substantial question that claim 1 of the '515 patent and claim 8 of the ' 152 patent would be obvious in light of WS_FTP Pro and a person skilled in the art. The remaining asserted claims 2–5 and 6–10 of the '515 patent and claims 9, 10, and 12 of the '152 patent are dependent on claim 1 of the '515 patent and claim 8 of the '152 patent, respectively. Thus, the Court need not reach the question of invalidity for the remaining claims. Accordingly, Box raises a substantial question concerning the validity of the Synchronization Patents in light of WS_FTP Pro and Open Text fails to demonstrate that Box's invalidity challenge lacks substantial merit.

2. Groupware Patents

Box cites eight prior art references as independently anticipating and/or rendering obvious claims 1, 2, 4, and 5 of the '055 patent and claims 1, 2, 3, and 5 of the '258 patent: (1) the Basic Support for Cooperative Work (“BSCW”) articles, which comprise Wolfgang Appelt et. al, “The BSCW System: A WWW–Based Application to Support Cooperation of Distributed Groups” (“Appelt” ); Richard Bentley et. al, “Supporting Collaborative Information Sharing with the World-Wide Web: The BSCW Shared Workspace System” (“Bentley”); and Wolfgang Appelt, “CoopWWW–Interoperable Tools for Cooperation Support using the World–Wide Web” (“CoopWWW”); (2) Jang Ho Lee et al., “Supporting Multi–User, Multi–Applet Workspace in CBE” (“Lee”); (3) Joseph T. Sinclair, et. al, “Intranets vs. Lotus Notes,” (“Sinclair Book”); (4) Lotus Notes, Version 4.5.2; (5) Mark Roseman, et. al, “TeamRooms: Network Places for Collaboration (“TeamRooms”); (6) Steve Ball, “SurfIt! A WWW Web Browser” (“SurfIt”). The Court has considered the above prior art references and finds that Box raises a substantial question of the Groupware Patents' validity in light of Appelt.

The '055 patent and '258 patent share a specification but the claims do not directly correspond. Instead, claim 1 of the ' 055 patent claims similar subject matter to claim 2 of the '258 patent and claim 1 of the '258 patent claims similar subject matter to claim 4 of the '055 patent. Claim 1 of the '055 patent discloses:

A system for providing a collaborative workspace, comprising:

a computer that is configured to create a dedicated network site on a network

connected server in response to instructions received from a primary user,

said dedicated network site defining said collaborative workspace;

memory associated with the computer that stores

i) a list of secondary users nominated by said primary user that have one or more levels of access to said collaborative workspace, and

ii) access control choices selectable by said primary user and defining the one or more levels of access afforded said secondary users of said collaborative workspace; wherein access to said levels of access include access via a web browser.
Claim 1 of the '258 patent provides:

A system for creating a collaborative workspace, comprising: a computer that is configured to:

receive information from a primary user via the primary user's web browser that specifies a number of secondary users and access levels for the secondary users;

create a collaborative workspace that limits access to the collaborative workspace to the primary and secondary users; and

provide a plurality of user applications selected to be included in the collaborative workspace with which the users can interact via a web browser based on their access level.

The parties dispute the following claim limitations present in the independent claims: (1) dedicated network site; (2) primary and secondary users; (3) applications selectable by the users to be included in the workspace.

i. Appelt

The Appelt article describes the BSCW system, a shared workspace in which members of a group coordinate and organize their work. Appelt at 3, Ex. I, Docket Item No. 95. Users are able to upload and download to the shared workspace various types of (electronic) objects including documents, spreadsheets, pictures, or links to internet pages. Id. at 2. Unlike workflow systems that provide the means for process definition and flow control through predefined coordination and cooperation models, shared workspaces provide the medium for information sharing without such pre-defined limitations. Id. The ultimate purpose is efficient information sharing where the shared workspace acts as an information repository. Id.

While existing groupware products provide the implementation of shared workspaces within one organization, the use of different groupware products gave rise to integration issues across organizations. Id. The BSCW System solves this problem by using the World–Wide Web (“WWW”) as the infrastructure for the shared workspace. Id. The cited advantages of using the WWW include its independence from any particular platform, its commonality amongst businesses and organizations, and the fact that it would not require additional installation. Id. at 3.

ii. Dedicated network site

Open Text concedes that Appelt teaches a shared workspace established by members of a group but argues that Appelt fails to identify a computer that “create[s] a dedicated network site on a network-connected server in response to instructions received from a primary user” because Appelt does not disclose how a workspace is created. Declaration of Dr. Nathaniel Polish (“Polish Decl.”) ¶ 43, Ex. 2, Docket Item No. 107 (emphasis added); see '055 patent, cl. 1; '258 patent cl. 2. Box argues the implementation of the BSCW System itself demonstrates how a user instructs the server to create the collaborative workspace. Declaration of Colin White (“White Decl.”) at 2–4, Ex. J–1, Docket Item No. 95. The Court finds Box's argument persuasive.

The BSCW System is a client/server model whereby the server hosts the shared workspace and the client allows users to access the workspace. Appelt at 4, Ex. I, Dkt. No. 95. The client and server interact such that a user sends a request to the server to perform a particular task, e.g., show the objects within a workspace or start the upload of a file into the workspace. Id. Based on the user's request, the server then computes a responsive HTML page to reflect the result of that task. Id. Accordingly, the server operates “on the fly” to create HTML pages based on the current state of the workspace. Id. Appelt emphasizes the dynamic nature of the workspace compared to earlier systems where users accessed static HTML pages. Id. Part of this dynamic nature is the user being able to interact with the server to make changes to the workspace. Even the shared workspace displayed in Figure 1 of the Appelt article is merely exemplary since the server operates “on the fly” and changes based on user interaction. See id. at 6.

This description teaches how users interact with the server based on an already existing workspace. However, user interaction with the server does not appear to be limited to pre-existing workspaces. Appelt also permits users to create individualized workspaces on the server. For example, a BSCW server was created for public access and users were invited to create workspaces. Id. at 6. Within the first six months, the registered users established several hundred distinct workspaces. Id. Accordingly, the BSCW system supports users interacting with the server by creating and updating existing workspaces.

Contrary to Open Text's argument, “how” a workspace is created can only be by the server after the creator of the workspace requests such a task. Open Text appears to argue that because Appelt discloses that “members of a group” establish a shared workspace for coordinating and organizing their work, there can be no primary user to create the workspace. See Polish Decl. at ¶ 43, Ex. 2, Dkt. No. 107. Such a proposition ignores common sense that a workspace would not exist but for the creator of a workspace requesting the server to create the workspace. Further, the later discussion of hierarchical access rights confirms that there is management authority within the workspace rather than an entirely communal collaboration space whereby any and all members may collectively create or interact with a workspace. Moreover, the ultimate purpose of shared workspaces is customizable group collaboration, allowing for the creator of the workspace to select the members and the features most salient to the group collaboration needs. Inasmuch as a server will not generate content for the shared workspace absent the creator's request, Appelt appears to disclose the limitation that a server creates a dedicated network site in response to a primary user's instructions.

a. Primary and Secondary users

Open Text argues that Appelt fails to distinguish between primary and secondary users whereby the primary user nominates a list of secondary users and selects access control choices for the secondary users. Polish Decl. ¶ ¶ 45–46, Ex. 2, Dkt. No. 107. According to Open Text, Appelt allows any member to add or remove other members to or from a workspace. Id. Box counters that members may “be added to or removed from a workspace by those members who have the respective authorisation.” White Decl. at 5, Ex. J–1, Dkt No. 95. Open Text mischaracterizes the plain meaning of the words “respective authorisation.” This phrase denotes a hierarchical structure to users who have the appropriate authority add or remove other members. Rather than concluding that any member would have such authorization, as Open Text opines, this excerpt conveys to the Court that one member, or primary user, is not precluded from authorizing which secondary users may access the workspace.

According to Open Text, the claimed limitation requires that a primary user select the secondary user's level of access to the workspace. Polish Decl. at ¶ 46, Ex. 2, Dkt. No. 107. The specification discloses the purpose of permitting a primary user to select the varying levels of access to secondary users: security. '055 patent 6:3–6. As an example, the patent discloses the varying security levels through the use of a password “which must be entered by both the primary user and the secondary users to gain access to the workgroup.” Id. at 6:6–10. In line with the claim limitation that primary users select one or more levels of access, the specification discloses that the password can be the same for the primary user and all the secondary users or that secondary users may be provided with a unique password. Id. at 6:10–12. The specification further considers the ability of the primary user to other security features such as who has the authority to add new users and/or delete users from the group, who has access to the administrative records of the workgroup, and whether security levels can be changed. Id. at 6:17–22.

Open Text argues that Appelt teaches the assignment of access rights to objects rather than secondary users. Polish Decl. at ¶ 46, Ex. 2, Dkt. No. 107. As support, Open Text focuses on the excerpt that “users can assign different access rights to objects in a workspace” underneath the subheading “Access Rights.” Indeed, Open Text is correct in pointing out that a creator of a document in Appelt may grant varying levels of access to other members with respect to the objects of the workspace such as editing privileges, read access or no access. Examples of objects include documents, tables, graphics, or spreadsheets. Appelt at 1, Ex. I, Dkt. No. 95. However, Open Text considers this excerpt in isolation without consideration to the other teachings of Appelt.

Returning to the words “respective authorisation,” Appelt discloses that one member may have the authority to add or remove other members from the workspace. Id. at 4. In this way, Appelt discloses more than what is claimed by teaching that a user may choose who can access the workspace as well as their type of access to the objects in the workspace, if any. Further, Appelt discloses that the BSCW System was originally conceived for information dissemination but envisions its use for dispersed working groups and organizations. Id. at 2. Accordingly, Appelt discloses the need for security measures, as does the claimed invention, by requiring that each user must “identify himself or herself by a user name and a password” to access the workspace. Id. at 4. Because the BSCW system administers the user names and passwords, it follows that only an authorized user can notify the BSCW system which secondary users may or may not access the workspace. Although Appelt is silent on whether the user name and passwords are the same between the primary and secondary users or if the secondary users have unique passwords, either would nonetheless satisfy the claimed limitation that the primary user select “one or more” levels of access to the workspace.

b. User applications selectable by the users

Open Text further contends that Appelt does not disclose a plurality of user applications selectable by the users to be included in the collaborative workspace because the BSCW System does not support any type of user applications. Polish Decl. ¶ 47, Ex. 2, Dkt. No. 107; see '055 patent cl. 4; '258 patent, cl. 1. Unlike the “user applications” within the meaning of the Groupware patents, Open Text argues the BSCW System discloses a workspace with “various types of objects.” Polish Decl. ¶ 47, Ex. 2, Dkt. No. 107. Box counters that while the Groupware Patent specification uses “application” and the Appelt article discloses “objects,” the distinction is meaningless because both the patent specification and Appelt provide examples of the same features. Hr'g Jan 24, 2014, Def.'s Likelihood of Success—Slides 79–80. It appears that Appelt's disclosure of objects satisfies the claim limitations that users may select applications to be included in the workspace.

The Groupware Patents' specification discloses several examples of workspace applications capable of being added to the workspace such as a bulletin board, real-time chat room, a group discussion application for groups to offer a central meeting place, and a document manager application to act as a central repository for posting and managing files and documents. '055 patent, 6:28–7:17. Figure 1 of the Appelt article displays a sample BSCW Workspace with exemplary objects such as a discussion forum, Microsoft Word documents for access control and member administration, and an Adobe FrameMaker document for document uploading. Appelt at 8–9, Ex. I, Dkt No. 95. Open Text distinguishes between objects and applications based on the actual terms of the patent and references, but fails to provide a definition of the distinction. Absent any guiding definition, there is no reason for the Court to distinguish between the bulletin board and chatroom applications disclosed in the specification from the discussion forum in Appelt. Similarly, the document manager application from the patent is akin to the member administration and document uploading features disclosed by Appelt. Thus, while the Appelt article uses “objects” and the Groupware Patents teach “applications,” both appear to disclose the same features.

Moreover, Figure 1 of the Appelt article demonstrates that the objects were added by various users. Appelt at 5, Ex. I, Dkt. No. 95. The same figure also shows action buttons on the top banner line permitting users to add folders, documents, URLs, and articles. Id. Further, the figure shows a “New” icon to indicate that an object has been added recently to this folder of the workspace. Id. If the applications were integrated into the workspace, as Open Text suggests, the purpose of a collaborative workspace whereby users have discretion to include or exclude functions relevant to the collaboration would be defeated. See Polish Decl. ¶ 48, Ex. 2, Dkt. No. 107.

Based on the foregoing, Box raises a substantial question as to whether the '055 patent and the '258 patent are invalid as anticipated in light of the Appelt article. Consequently, Open Text has not succeeded in showing it is likely to succeed at trial on the merits of the validity issue.

B. Irreparable Harm

Although Open Text has not shown it is likely to be successful on the merits, a strong showing of irreparable harm may offset a failure to show a likelihood of success on the merits. See Alliance for the Wild Rockies, 632 F.3d at 1131 (“[A] stronger showing of one element may offset a weaker showing of another”). Open Text must provide “[s]ome evidence and reasoned analysis” that the remedies available at law, such as monetary damages, are inadequate to compensate for that injury.” eBay, 547 U.S. at 391, 126 S.Ct. 1837. Open Text must also establish that a “sufficiently strong causal nexus” relates the alleged harm to the alleged infringement. Apple , Inc. v. Samsung Electronics Co., Ltd., 678 F.3d 1314, 1324 (Fed.Cir.2012).

1. Alleged Harm

Open Text advances three theories to support its assertion that it will suffer irreparable harm in the absence of a preliminary injunction. First, Open Text argues that it will suffer a loss of goodwill because, if Open Text is successful at trial, it will likely obtain an injunction that will require Box customers to “rip out” the infringing software, thereby alienating future customers. Pl.'s Mot. at 24–25, Dkt. No. 65. Second, Open Text argues it is likely to lose market share in both the Content Server market and from sales of complementary products linked to lost Content Server sales. Id. at 25–27. Finally, Open Text argues it will lose its first-mover advantage in the emerging off-premises market. Id. at 27–28.

a. Loss of Goodwill

Loss of goodwill, as well as damage to reputation, can support a finding of irreparable harm. See Celsis In Vitro, Inc. v. CellzDirect, Inc., 664 F.3d 922, 930 (Fed.Cir.2012) (citing Abbott Labs. v. Sandoz, 544 F.3d 1341, 1362 (Fed.Cir.2008).

Open Text's loss of goodwill reasoning is nuanced. Open Text's requested preliminary injunction seeks only to prevent future sales of products with the infringing feature to large customers. This relief is necessary to prevent a loss of goodwill, Open Text argues, because if Open Text prevails at trial and obtains a permanent injunction against Box, that permanent injunction would likely require Box customers to remove the infringing software from their systems. Pl.'s Mot. at 24–25, Dkt. No. 65. Because the nature of Box's and Open Text's products is such that large-size customers invest a “significant investment of time, money, and other organizational resources” in order to deploy the software onto their systems, Open Text fears that a permanent injunction requiring Box's customers to remove Box's software will cause those customers to be angry at Open Text because it was Open Text who sought that permanent injunction. Id.; Declaration of Tim Spadzinski (“Spadzinski Decl.”) ¶ 22, Docket Item No. 65. In other words, Open Text's argument is that preliminary injunctive relief is necessary to prevent further large-size customers from installing infringing Box software to reduce the number of customers that will need to remove infringing Box software following issuance of a permanent injunction.

Box disagrees, arguing that Open Text's loss of good will argument is convoluted because the irreparable harm alleged “depends upon too many unknown contingencies” and is not caused by the alleged infringement itself, but instead a “hypothetical, mandatory permanent injunction.” Def.'s Opp'n at 24, Dkt. No. 85. Box contends that any irreparable harm to Open Text would be “self-inflicted” by a permanent injunction requiring the forced removal of Box software. Id. Finally, Box argues that Open Text has failed to establish a sufficient causal nexus between the asserted injury and the infringing conduct.

The Court disagrees with Box that the ultimate cause of the alleged harm is the “hypothetical mandatory permanent injunction,” not the alleged infringement. The relevant question for the Court is to what extent can “the harm resulting from selling the accused product can be ascribed to the infringement.” Apple Inc. v. Samsung Electronics Co., Ltd., 695 F.3d 1370, 1375 (Fed.Cir.2012). Since Open Text would not be seeking a permanent injunction but for the alleged infringement, the harm that would flow from this hypothetical permanent injunction is necessarily ascribed to the infringement. Otherwise, Open Text would be left with the Hobson's choice of either angering potential customers or letting infringement continue in the hopes that monetary damages can be adequately calculated. Thus, Open Text plausibly argues that it will suffer at least some loss of goodwill should a future permanent injunction require that alleged Box infringing products be “ripped out.”

b. Irreversible Loss of Market Share and Complementary Product Sales

Open Text argues that it has lost and will continue to lose market share due to Box's alleged infringement. Open Text contends this harm cannot be remedied by monetary damages. Loss of market share to a competitor or the permanent loss of customers as a result of infringing conduct may support a finding of irreparable harm. Apple, Inc. v. Samsung Electronics Co., Ltd., 11–CV–01846–LHK, 2011 WL 7036077 (N.D.Cal. Dec. 2, 2011) aff'd in part, vacated in part, remanded, 678 F.3d 1314 (Fed.Cir.2012). The relationship between Open Text and Box is also important to the Court's analysis for irreversible loss of market share. Courts are more likely to grant an injunction in two-player markets where the parties are direct competitors. See, e.g., Robert Bosch LLC v. Pylon Mfg. Corp. , 659 F.3d 1142, 1153 (Fed.Cir.2011). In a market where only two parties are directly competing, the potential harm in allowing the defendant to continue its infringing conduct is probably at its greatest. See, e.g., i4i Ltd. Partnership v. Microsoft Corp., 598 F.3d 831, 861 (Fed.Cir.2010).

In attempting to show the Court that it is permanently losing market share, Open Text relies on Robert Bosch for the proposition that it need only make a prima facie showing of lost market share and Butamax Advanced Biofuels LLC v. Gevo, Inc., 868 F.Supp.2d 359, 373 (D.Del.2012), for the proposition that evidence of actual lost sales is not needed before seeking a preliminary injunction. Open Text argues that the threat of an infringing competitor's entry into a market, the length of its customer relationships, and the magnitude of the harm caused by every lost sale is more than sufficient in light of these decisions. Pl.'s Mot at 25–26, Dkt. No. 65. With this standard in mind, Open Text argues that it has made at least a prima facie showing of lost market share.

The existence of a two-player market may well serve as a substantial ground for granting an injunction because it creates an inference that an infringing sale amounts to a lost sale for patent holder. However, the converse is not automatically true. The Court agrees with Box that Open Text's reliance on Butamax is inapt, not because a duopoly is required, but because the landscape of the relevant market is notably distinguishable from the two-player market in Butamax and Open Text is not a first-mover. The court in Butamax addressed a two-competitor market and found the plaintiff held a first-mover advantage. The relevant market may be filled with as many as twenty-six competitors and the July 2013 Forrester report evidences at least sixteen. Def.'s Opp'n at 25, Dkt. No. 85; Declaration of Joely Urton (“Urton Decl.”) at 3, Docket Item No. 90.

The Court also finds Open Text's reliance on Robert Bosch to be misplaced. Unlike in Robert Bosch, where the plaintiff directly competed with defendant in a two-competitor market and sought a permanent injunction following a jury's determination the defendant infringed, no jury has determined infringement and the cloud market is filled with at least sixteen competitors. Although the “quantum of evidence” required to prove irreparable harm is unclear, case law is clear that the potential for loss of market share is insufficient. MercExchange, L.L.C. v. eBay, Inc., 500 F.Supp.2d 556, 577 (E.D.Va.2007) (emphasis added). Open Text has not provided to the Court, with any level of specificity, what sales have been lost to Box, what Open Text's market share is in relevant market, or any evidence of actual lost customers going to Box. In Praxair, Inc. v. ATMI, Inc., 479 F.Supp.2d 440, 442 (D.Del.2007), the district court denied the plaintiff's motion for an injunction notwithstanding the fact that the litigants were direct competitors because the plaintiff “ha[d] not provided or described any specific sales or market data to assist the court, nor ha[d] it identified precisely what market share, revenues, and customers [it] has lost to [the defendant].” Moreover, Open Text's reliance upon conveniently timed sales growth showing Box's revenue growing in the four quarters following the release of Box Edit is insufficient to carry Plaintiff's burden to establish irreparable harm. The Court puts minimal credence in this circumstantial evidence without more because: (1) it is unclear for what portion of the market Open Text competes with Box and (2) Box has presented evidence indicating that the growth in sales can be attributed to demand drivers other than the patented processes.

Without some evidence showing Open Text's actual market share in the relevant market, and some evidence showing loss of sales to Box, the Court cannot determine if the Open Text is losing market share or even whether the parties meaningfully compete. See Illinois Tool Works, Inc. v. Grip–Pak, Inc., 906 F.2d 679, 683 (Fed.Cir.1990) (rejecting the plaintiff's claim that the potential for lost sales alone demonstrates irreparable harm as “acceptance of that position would require a finding of irreparable harm to every manufacturer/patentee, regardless of circumstances”). Thus, the Court finds that Open Text's theory of harm resulting from lost market share adds little to the total “quantum of evidence” required to show irreparable harm.

The Court also agrees with Box that that Open Text has not shown that the patented feature drove demand for the allegedly infringing Box software. The Federal Circuit requires Open Text as the plaintiff to establish a causal nexus in order to show that lost sales would not be lost but for the offending feature in the accused product. Apple, 678 F.3d at 1324–25. Sales lost to an infringing product cannot irreparably harm Open Text if customers are buying Box products for reasons other than the alleged patented features. Id. Open Text makes a case for nexus circumstantially, grounding its argument on Box's sales growth following the Box Edit release. Hr'g Jan 24, 2014, Pl.'s Slide 21 (showing [Redacted] percent growth in sales to customers with 100+ users in the four quarters after Box Edit is released). Open Text opines that the growth in sales following the release of Box Edit establishes the causal nexus and evidences that Box Edit is driving customer demand in the market. Id. Neither in its Motion nor at oral argument did Open Text present any evidence of consumer surveys showing Box Edit to be the demand driver or any evidence showing sales being lost to Box because of Box Edit. This is not a sufficient showing that the harm flows from Box's alleged infringement. Box sufficiently rebutted Open Text's attempt to show a causal nexus by presenting evidence that its business grew at “a record-setting pace in the first half of 2012” due to Box's expansion of its operations in Europe. Declaration of John Bovich (“Bovich Decl.”) at 2, Ex. 4, Docket Item No. 86. Box Enterprise sales also increased over 200 percent from the previous year. Id. The allegedly infringing Box Edit software that Open Text argues is driving consumer demand was not even released until October 2012, well after these large increases in sales were reported. The evidence Open Text has provided is speculative, especially when Box's sales growth is analyzed on a longer timeline, and leaves the Court to guess whether the increase in sales in the four quarters following the release of Box Edit is the demand driver.

Neither the “difficulty of calculating losses in market share, nor speculation that such losses might occur,” without more, warrants a finding of irreparable harm and the issuance of a preliminary injunction. MicroAire Surgical Instruments, LLC v. Arthrex, Inc., 726 F.Supp.2d 604, 635 (W.D.Va.2010). Regardless of the extent to which Open Text may be injured by the sales of Box products, there is not a sufficient showing that the harm flows from Box's alleged infringement. For all the above reasons, the Court finds that Open Text has not made a sufficient showing of irreparable harm based on irreversible loss of market share.

c. Loss of First–Mover Advantage

Open Text argues that, as a market leader in on-premises enterprise document management systems for the last decade, it has positioned itself to be a first-mover in the emerging off-premises, (e.g., cloud-based) document management systems for large customers. Pl.'s Mot. at 28, Dkt. No. 65. Open Text argues it is well-positioned to become a market leader in cloud-based document management and already hosts 2 billion annual transactions, 1.5 million end users, 25,000 customers, and data infrastructure in 25 countries. Id.The existence of a at 27–28. Open Text contends the allegedly infringing Box Edit, along with Box's other new offerings, allow Box to enter this “new” market for large corporate customers and in so doing deprives Open Text of its first-mover advantage. Id.

Box counters that it was Box, not Open Text, who was the “first-mover” into the cloud market. Box presented evidence that from July 2007 through September 15, 2013, more than [redacted] companies have elected the Box Enterprise tier for use by 100 or more individual users. Def.'s Opp'n at 5. And in the over [redacted] sales calls and presentations to potential large customers that expressed an interest in deploying Box for 100 seats or more, Open Text was never identified as a primary competitor in any of the [redacted] calls. Shirk Decl. ¶ 17, Dkt. No. 114. Moreover, Box presented evidence that for Fiscal Years 2008 through the present, Box has, on average, convinced more than [redacted] of potential customers to deploy Box for 100 seats or more. Id. While the Elite Plan does not add any new functionality, Open Text argues that it is the new services combined with the allegedly infringing feature, Box Edit, that brings Box directly into competition with Open Text. Pl.'s Mot. at 9–10, Dkt. No. 65.

Open Text falls flat of convincing the Court Open Text was likely a “first-mover” into the market. Open Text did not report any revenue from cloud services in 2011, and for most of 2012, while Box showed its Enterprise sales increased over [redacted] percent from the same year. Bovich Decl. at 2, Exs. 4, 7, Dkt. No. 86. Open Text reported revenues in the cloud sector for the first time in the third quarter of 2012. Id.; see also Bovich Decl. Ex. 1. The lack of cloud-based revenue contradicts Open Text's own assertion that it is establishing a “new market” for the off-premises document management market when considerable countervailing evidence in the record indicates Box likely entered the relevant market and was competing for large customers as early as 2007. Pl.'s Mot. at 27, Dkt. No. 65. Also instructive to the Court is the July 2013 Forrester report concluding that Box was the leader in the cloud market among the “16 most significant solution providers,” and Open Text failing to even make the list. Urton Decl. at 3, Dkt. No. 90. Accordingly, the Court finds the evidence does not support Open Text's argument that it is a first-mover into market.

2. Delay

A defendant can rebut a plaintiff's argument of irreparable harm by a showing that the moving party delayed in bringing its infringement action. See Apple, 2011 WL 7036077, at *22. Delay in seeking a remedy is an important factor bearing on the need for a preliminary injunction, the rationale being that absent a good explanation, a patent holder would promptly have sought an injunction if it were being irreparably harmed.

In this case, Box argues that Open Text unduly delayed in bringing its preliminary injunction motion eighteen months after the release of Box Edit, and as a result, demonstrates Open Text is not being irreparably harmed by the alleged infringement of the asserted patents. Open Text explains its filing date of September 13, 2013 as timely because it was only after the release of Box's Elite Plan in August of 2013 that Open Text discovered Box was directly competing with them. Pl.'s Mot. at 9–10, Dkt. No. 65. While the Elite Plan does not add any new functionality, Open Text argues that it is the new services combined with the allegedly infringing feature, Box Edit, that brings Box directly into competition with Open Text. Id.

The Court agrees with Box that the Elite Plan was not likely Box's entry point into the market for customers with more than 100 users. Box released its Enterprise software in 2007 and had a significant number of large customers by early 2012. Shirk Decl. ¶ 10–11, Dkt. No. 114; Bovich Decl. at 2, Ex. 4, Dkt. No. 86; Def.'s Opp'n at 5 (these large customers include [redacted]. Furthermore, Open Text's current assertion that Box is “only now poised” to sell to large customers is opposite of its statement in its pre-motion court filings, including its June 2013 Complaint, that Box was in “direct competition” warranting a preliminary injunction. See Complaint at ¶¶ 29, 34, 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, Dkt. No. 1. Box correctly points out Open Text failed to move at that time. Ample evidence in the record suggests Box was already competing in the relevant market since at least 2012, and arguably as early as 2007 with Box's release of the Enterprise Plan. Accordingly, Open Text's delay in pursuing its infringement claim against Box weighs in Box's favor.

3. Inadequate Remedies Available at Law

The burden is on Open Text to iterate the specific reasons why the infringement cannot be adequately compensated with monetary damages. eBay, 547 U.S. at 391, 126 S.Ct. 1837. The availability of damages is particularly significant when the plaintiff falls short of convincing the court that protection through interim equitable relief is required. See Illinois Tool Works, Inc. v. Grip–Pak, Inc., 906 F.2d 679, 683 (Fed.Cir.1990).

Open Text argues monetary damages cannot compensate Open Text for the harm it has suffered and it will continue to suffer because the nature of the document management industry and Open Text's business model make full lost profits difficult, if not impossible, to capture with a monetary award. Pl.'s Mot. at 28, Dkt. No. 65; Pl.'s Rebuttal Br. at 8, Dkt. No. 120. Difficulty in estimating monetary damages is evidence that remedies at law are inadequate. See Broadcom Corp. v. Qualcomm Inc., 543 F.3d 683, 703–04 (Fed.Cir.2008). Open Text argues that it is difficult to measure the total amount of lost profits attributable to each lost sale of Content Server Products because its business model typically collects revenues from sales of its both its Content Server Products and Complementary Products. Pl.'s Mot. at 28, Dkt. No. 65. Open Text contends this is because the majority of the revenues stemming from any single Content Server sale likely extends over many years, varies significantly based on the options and add-on products selected by the client, and generates unspecified sales of complementary Open Text products. Pl.'s Rebuttal Br. at 8 –9, Dkt. No. 120.

The Court acknowledges Open Text has provided specific reasons why monetary damages would be inadequate that Box fails to rebut. The Court overrules Box's evidentiary objection to Tim Spadzinski's declaration. Nonetheless, Open Text's showing as to why monetary damages would be inadequate fails to establish irreparable harm in light of the factors discussed herein.

4. Summary

On balance, Open Text's potential loss of goodwill and weak showing of inadequate remedies at law fail to offset Open Text's arguments establishing irreparable harm. Open Text did not show it was entitled to a first-mover advantage, did not make a sufficiently strong causal nexus between the alleged harm and the allegedly infringing Box Edit, and did not adequately establish lost market share. Moreover, Open Text's delay in filing its preliminary injunction undercuts its argument for irreparable harm. In consideration of all the evidence presented, and in light of the Court's finding that Box raises a substantial question to the validity of the patents, the Court finds that Open Text has not established that it is likely to be irreparably harmed in the absence of an injunction.

C. Balance of Hardships

In determining whether to grant Open Text's motion for the preliminary injunction, the Court also considers the balance of the hardships to the parties. Abbott Labs., 452 F.3d at 1334. “In each case, courts ‘must balance the competing claims of injury and must consider the effect on each party of the granting or withholding of the requested relief.’ ” Winter, 555 U.S. at 24, 129 S.Ct. 365 (quoting Amoco Prod. Co. v. Gambell, 480 U.S. 531, 542, 107 S.Ct. 1396, 94 L.Ed.2d 542 (1987)). A preliminary injunction is a drastic remedy and the hardship on a preliminarily enjoined party who is required to withdraw its product from the market before trial can be devastating. See Illinois Tool Works, Inc. v. Grip–Pak, Inc., 906 F.2d 679, 683 (Fed.Cir.1990).

Open Text argues that, even though Box infringes claims of twelve asserted patents from two different patent families, its preliminary injunction is limited in scope. Pl.'s Mot. at 29–30, Dkt. No. 65. By targeting only a single feature, Box Edit, and only customers with 100 users or more, the preliminary injunction still allows Box to continue selling its document management software without Box Edit to large customers and Box's software with Box Edit to any customer whose user base is less than 100. Id. Moreover, the injunction would not force any Box customer to stop using Box Edit, it only would restrict sales between its issuance and trial. Pl's Rebuttal Br. at 30, Dkt. No. 107. Therefore, Open Text argues that this factor weighs in its favor because the irreparable harm to Open Text outweighs any potential harm to Box.

Box argues that Open Text is in a “much better” position to withstand the improper denial of the injunction than Box is to withstand an improper granting of the injunction. Def.'s Opp'n at 28, Dkt. No. 85. Box argues that an injunction preventing sales to large customers is an undue hardship because the majority of Box's revenues are derived from its large customer base. Id. Unlike Open Text whose revenues are not primarily derived from its “cloud” based offerings, Box's entire business is cloud-based. Id.; Bovich Decl. at 2, Ex. 7, Dkt. No. 86 (Cloud computing services accounted for only [redacted] million of Open Text's total revenue of [redacted] million in the first quarter of 2013). The proposed injunction according to Box would prevent Box from providing a feature of its cloud product to the bulk of its actual and potential customer base—any customer with over 100 users. Furthermore, Box argues that although Open Text claims that it only wants “narrow” and “prospective” relief, the injunction would ensure a substantial hardship on Box's existing and prospective customer relationships because current and prospective “smaller” customers often need scalability to potentially accommodate more than 100 users. Shirk Decl. ¶ 17, Dkt. No. 114. Box's current “small” customers may elect to “rip out” and replace Box now rather than wait for the litigation to conclude, and prospective “small” customers may decide not to purchase from Box at the outset because “decision makers typically consider whether Box is scalable” and will be able to meet the needs of the company as it grows. Id.

After careful consideration, the Court agrees with Box that Open Text's proposed injunction is likely to cause a substantial financial hardship to Box pending the outcome of this litigation. Box presented evidence that since 2012, it has more than [redacted] businesses, including large customers such as [redacted]. using Box software. Bovich Decl. at 2, Ex. 2, Dkt. No. 86. Box operates exclusively in the cloud market and these companies constitute the majority of Box's revenue. Declaration of Sebastien Lemenager (“Lemenager Decl.”) at ¶ 5, Docket Item No. 86 (approximately [redacted] of Box's TAV [Total Account Value] was derived from customers having 100 seats or more). Open Text's showing of loss of market share is speculative, whereas an injunction would exclude Box from selling in the market where a majority of its revenue is derived. Moreover, the hardship to Open Text if the preliminary injunction is denied is severely undercut by the fact that Box has raised a substantial question to the validity of the patents. After weighing all the evidence presented, the Court finds an injunction precluding Box from selling Box Edit to large customers tips the balance of hardships in Box's favor.

D. Public Interest

The filial prong the court must consider is whether an injunction weighs in the public interest. Open Text correctly cites in its motion, ‘the public interest factor weighs in favor of the patent holder when it establishes a likelihood of success on the merits.‘ See, e.g., Abbott Labs . v. Andix Pharms., Inc., 452 F.3d 1331, 1348 (Fed.Cir.2006). Unfortunately for Open Text, the Court finds Box has raised a substantial question concerning the validity of the asserted patents. Open Text's total reliance on its likelihood of success, which Open Text failed to persuade the Court was likely, likewise fails to persuade the Court that the public interest factor weighs in Open Text's favor. Where, as here, the Court has found that Box has raised a substantial question concerning the validity of the '515, ' 152, '055, and '258 patents, the public interest is best served by the denial of a preliminary injunction. See, e.g., Dillon Jeneric/Pentron, Inc. v. Dillon Co., Inc., 1999 WL 66537, at *14 (D.Conn. February 3, 1999), aff'd, 205 F.3d 1377 (Fed.Cir.2000).

IV. Conclusion

Open Text has not demonstrated a reasonable likelihood of success on the merits, failed to make a strong showing of irreparable harm in order to compensate for prong one of Winters, and the balance of hardships does not tip sharply toward Open Text. Additionally, this Court recognizes the public interest in protecting patent rights, but also that a preliminary injunction is an extraordinary remedy reserved for limited circumstances which clearly demand it. Finding this case not be one that clearly demands it, the Court hereby DENIES the Motion for a Preliminary Injunction.

IT IS SO ORDERED.