`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`________________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`________________
`DROPBOX, INC.,
`Petitioner,
`
`v.
`
`ENTANGLED MEDIA, LLC
`Patent Owner.
`___________
`Case No. IPR2024-00285
`U.S. Patent No. 8,484,260
`___________
`
`
`
`DECLARATION OF PATRICK D. MCDANIEL, PH.D.
`
`U.S. PATENT NO. 8,484,260 (CLAIMS 1-8)
`
`
`
`
`
`
`
`
`
`
`
`
`
`Dropbox Exhibit 1003
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`C.
`
`V.
`
`Contents
`ENGAGEMENT .......................................................................................... 1
`I.
`PROFESSIONAL BACKGROUND............................................................ 1
`II.
`III. MATERIALS REVIEWED ......................................................................... 6
`IV.
`DESCRIPTION OF THE RELEVANT TIMEFRAME, THE
`RELEVANT FIELD, AND A PERSON OF ORDINARY SKILL IN
`THE ART ...................................................................................................... 7
`TECHNOLOGY BACKGROUND .............................................................. 7
`A. What are “File Systems”? .................................................................... 8
`B.
`The Evolution of File Systems ........................................................... 12
`1.
`From Files to Folders ............................................................... 12
`2.
`Remote Storage for File Systems ............................................. 13
`3.
`Sharing Files with Peers ........................................................... 15
`The Fundamentals of File Systems .................................................... 18
`1.
`The File System API ................................................................ 19
`2. Modifying File Systems ........................................................... 21
`D. Distributed File Systems in Peer-to-Peer Networks .......................... 24
`OVERVIEW OF THE ʼ260 PATENT ....................................................... 26
`VI.
`VII. OVERVIEW OF THE PRIOR ART .......................................................... 28
`A. Havewala ............................................................................................ 28
`B. Adams ................................................................................................. 32
`C.
`Saridakis ............................................................................................. 34
`D.
`Rothman ............................................................................................. 35
`VIII. CLAIM CONSTRUCTION ....................................................................... 36
`IX.
`ANTICIPATION ........................................................................................ 37
`X.
`OBVIOUSNESS ......................................................................................... 37
`XI.
`OPINIONS REGARDING PATENTABILITY ........................................ 39
`XII. MY OPINIONS AS TO THE CHALLENGED CLAIMS OF THE
`’260 PATENT ............................................................................................. 41
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`i
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`A. My Opinions as to Claims 1, 4-8 concerning Havewala in view
`of Adams ............................................................................................ 41
`1.
`Claim 1 ..................................................................................... 41
`a.
`1[P]................................................................................. 43
`b.
`1[A] ................................................................................ 53
`c.
`1[B] ................................................................................ 56
`d.
`1[C] ................................................................................ 66
`e.
`1[D] ................................................................................ 70
`f.
`1[E] ................................................................................ 76
`g.
`1[F]................................................................................. 80
`h.
`1[G] ................................................................................ 85
`i.
`1[H] ................................................................................ 90
`Claim 4 ..................................................................................... 93
`2.
`Claim 5 ..................................................................................... 95
`3.
`Claim 6 ..................................................................................... 98
`4.
`5. Motivation to Combine Havewala and Adams ........................ 99
`6.
`Claim 7 ................................................................................... 103
`a.
`7[P]............................................................................... 104
`b.
`7[A]-[H] ....................................................................... 105
`Claim 8 ................................................................................... 107
`a.
`8[P]............................................................................... 108
`b.
`8[A] .............................................................................. 108
`c.
`8[B] .............................................................................. 109
`d.
`8[C]-[J] ........................................................................ 110
`B. My Opinions as to Claims 2 and 3 based on Havewala in view
`of Adams and Saridakis. ................................................................... 112
`1.
`Claim 2 ................................................................................... 112
`2.
`Claim 3 ................................................................................... 114
`
`7.
`
`
`
`ii
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`3. Motivation to Combine Havewala with Adams and
`Saridakis ................................................................................. 115
`C. My Opinions as to Claims 2 and 3 based on Havewala in view
`of Adams, Saridakis, and Rothman. ................................................. 116
`1.
`Claim 1-8 ................................................................................ 116
`2. Motivation to Combine Havewala, Adams, Saridakis, and
`Rothman ................................................................................. 121
`XIII. CONCLUDING STATEMENTS ............................................................. 123
`
`
`
`iii
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`EXHIBIT LIST
`
`EXHIBITS FILED BY PETITIONER
`U.S. Patent No. 8,484,260 (the “’260 patent”)
`
`Ex-1001
`
`Ex-1002
`
`File History of the ’260 patent
`
`Ex-1003
`
`Declaration of Dr. Patrick McDaniel
`
`Ex-1004
`
`CV of Dr. Patrick McDaniel
`
`Ex-1005
`
`U.S. Patent Application Publication 2007/0016621 (“Havewala”)
`
`Ex-1006
`
`U.S. Patent Application Publication 2002/0046232 (“Adams”)
`
`Ex-1007
`
`U.S. Patent No. 8,874,691 (“Saridakis”)
`
`Ex-1008
`
`U.S. Patent Application Publication 2005/0289218 (“Rothman”)
`
`Ex-1009
`
`Dropbox, Inc.’s Sotera Stipulation
`
`Ex-1010
`The Linux Information Project, “Metadata Definition.” 2006.
`Ex-1011 Wayback printout, “NTFS Master File Table (MFT)”,
`NTFS.com
`
`Ex-1012
`
`Eberspächer, Jörg and Schollmeier, Rüdiger. “Peer-to-Peer
`Systems and Applications.” (2005), Chapter 5 First and Second
`Generation of Peer-to-Peer Systems.
`
`Ex-1013
`Frystyk, Henrik. “The World-Wide Web.” (1994).
`Ex-1014 Wayback printout, “POSIX® 1003.1 Frequently Asked
`Questions (FAQ Version 1.10)”
`
`Ex-1015
`
`Sandberg, Russel. “The Sun Network Filesystem: Design,
`Implementation and Experience.” (2001).
`
`Ex-1016 Wayback printout – “Filesystems”
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`iv
`
`
`
`
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`Ex-1017 Webpage printout – “Multics”
`
`Ex-1018
`
`Sandberg, Russel, et al. “Design and implementation of the Sun
`network filesystem.” Proceedings of the summer 1985 USENIX
`conference. 1985.
`
`Ex-1019
`
`[Reserved]
`
`Ex-1020 Wayback printout – Napster
`
`Ex-1021
`
`Shuler, Rus. “How Does the Internet Work?” (2002).
`
`Ex-1022
`
`Fox, Geoffrey and Pallickara, Shrideep. “Peer-to-Peer
`Interactions in Web Brokering Systems.” (2002).
`Ex-1023 Wayback printout – Introduction to Linux Loadable Kernel
`Modules
`Ex-1024 Webpage printout – Deploying your hardware and software
`systems
`
`Ex-1025 Wayback printout – Invasion of the Data Snatchers
`Ex-1026 Warrick, R. Drake. “File Management: My Computer and
`Explorer.” (2002).
`Ex-1027 MacManus, Richard. “Microsoft has 97% of OS market, says
`OneStat.com.” (2006)
`
`Ex-1028
`
`Alan Freedman, Computer Desktop Encyclopedia, Ninth Edition,
`Osborne/McGraw-Hill, 2001.
`
`
`
`v
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`I, Patrick D. McDaniel, Ph.D., declare as follows:
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`ENGAGEMENT
`I.
`I have been retained by Dropbox, Inc. (“Dropbox” or “Petitioner”) in
`
`1.
`
`connection with the above-captioned petition for Inter Partes Review (“IPR”) of
`
`U.S. Patent No. 8,484,260 to Erik Caso and Michael Abraham (“the ʼ260 Patent,”
`
`Ex1001). I understand the ʼ260 Patent is currently assigned to Entangled Media,
`
`LLC (“Patent Owner” or “PO”).
`
`2.
`
`I have been asked by Petitioner to offer opinions regarding the ’260
`
`Patent, including the unpatentability of claims 1-8 (which I may refer to
`
`subsequently as the “challenged claims”) in view of certain prior art. This
`
`declaration sets forth the opinions I have reached to date regarding these matters.
`
`3.
`
`I am being compensated by Petitioner at my standard hourly
`
`consulting rate for my time spent on this matter. My compensation is not
`
`contingent on the outcome of the IPR or on the substance of my opinions.
`
`4.
`
`I have no financial interest in Petitioner or Patent Owner.
`
`PROFESSIONAL BACKGROUND
`II.
`5. My qualifications are set forth in my curriculum vitae, a copy of
`
`which is attached as an Exhibit 1004. Exhibit 1004 also includes a list of my
`
`publications and the cases in which I have testified at deposition, hearing, or trial
`
`within the past four years. As set forth in my curriculum vitae:
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`1
`
`

`

`I earned my Ph.D. in Computer Science from the Computer Science
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`6.
`
`and Engineering Department at University of Michigan, Ann Arbor in 2001. I
`
`earned my Bachelor of Science degree in Computer Science from Ohio University
`
`in 1989, and my Master of Science degree, also in Computer Science, from Ball
`
`State University in 1991.
`
`7.
`
`Since 2022, I have been Tsun-Ming Shih Professor of Computer
`
`Sciences in the School of Computer, Data and Information Sciences at the
`
`University of Wisconsin-Madison. Prior to my move to Wisconsin, I was the
`
`William L. Weiss Professor of Information and Communications Technology in
`
`the School of Electrical Engineering and Computer Science at the Pennsylvania
`
`State University in University Park, PA. I was also the director of the Institute for
`
`Network and Security Research, and founder and co-director of the Systems and
`
`Internet Infrastructure Security Laboratory, a research laboratory focused on the
`
`study of security in diverse network and computer environments. My research
`
`efforts primarily involve computer systems, network management and
`
`authentication, systems security, and technical public policy.
`
`8.
`
`I frequently teach and do research on filesystems and storage,
`
`particularly as it relates to security. For example, Chapter 6 of my PhD
`
`dissertation, “Case Studies: Virtual Private Filesystems in AMirD”, described a
`
`
`
`2
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`peer-to-peer distributed file system that allowed multiple devices to synchronize
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`and share metadata indices and file data blocks amongst devices. I continue to do
`
`research in filesystems to this day, and in my most recent work (appearing in the
`
`technical report Securing Cloud File Systems using Shielded Execution) has
`
`developed techniques for securing filesystem metadata on untrusted hardware. I
`
`also frequently teach filesystems design in my systems programming course.
`
`9.
`
`Prior to 2017, I was an Assistant Professor (2004-2007), Associate
`
`Professor (2007-2011), Full Professor (2011-2015), and Distinguished Professor
`
`(2016-2017) of Computer Science and Engineering and the William L. Weiss
`
`Professor of Information and Communications Technology (2017-2022) at the
`
`Pennsylvania State University. Since 2004, I have taught courses in the field of
`
`computer systems, systems programming, mobile device security, networks, and
`
`network and computer security at both the undergraduate and graduate level. I
`
`created and maintained several of these courses at Penn State University. I have
`
`begun teaching courses at the University of Wisconsin-Madison in the Fall of 2023
`
`teaching network and systems security.
`
`10. From 2003-2009, I was also an Adjunct Professor at the Stern School
`
`of Business at New York University in New York, NY. At the Stern School of
`
`Business, I taught courses in computer and network security and online privacy.
`
`
`
`3
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`I am a Fellow of the Association for Computing Machinery (ACM,
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`11.
`
`the leading professional association for computer science), a Fellow Institute for
`
`Electrical and Electronics Engineering (IEEE, the leading professional association
`
`for engineering), and Fellow of the American Association for the Advancement of
`
`Science (AAAS, the leading professional association for science).
`
`12.
`
`I am the director of the NSF Frontier Center for Trustworthy Machine
`
`Learning (CTML), a research organization exploring security and privacy of
`
`machine learning (and AI in general). Funded by the US National Science
`
`Foundation, the CTML is led by the University of Wisconsin-Madison and has
`
`members from Stanford, the University of California, Berkeley, The University of
`
`California-San Diego, and the University of Virginia.
`
`13.
`
`I was the Program Manager (PM) and lead scientist for the Cyber
`
`Security Collaborative Research Alliance (CRA). The CRA is led by Penn State
`
`University and includes faculty and researchers the Army Research Laboratory,
`
`Carnegie Mellon University, Indiana University, the University of California-
`
`Davis, and the University of California-Riverside. This initiative is a major
`
`research project aimed at developing a new science of cyber-security for military
`
`networks, computers, and installations.
`
`
`
`4
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`I have served as an advisor to several Ph.D. and master’s degree
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`14.
`
`candidates, several of whom have gone on to become professors at various
`
`institutions such as North Carolina State University, Purdue University, the
`
`University of Toronto, the University of Oregon, and the Georgia Institute of
`
`Technology and the University of Florida. I am currently an advisor to seven
`
`Ph.D. students.
`
`15. Before joining Pennsylvania State University as a professor, I was a
`
`software developer and project manager for companies in the networking industry
`
`including Applied Innovation, Inc., and Primary Access Corporation. I was also a
`
`senior researcher at AT&T Research-Labs. As part of my duties in these industrial
`
`positions, I informed, reviewed, and formed corporate policies and practices
`
`relating to the security of mobile devices, networks, and software systems
`
`(including filesystems)
`
`16.
`
`I have published extensively in the field of mobile security, network
`
`and security management, computer systems, authentication, storage and
`
`filesystems systems security, applied cryptography and network security. In
`
`addition to writing several articles for industry journals and conferences, I have
`
`authored portions of numerous books related to computer systems, applied
`
`cryptography and network security. I have served on the editorial boards of
`
`
`
`5
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`several peer-reviewed journals including ACM Transactions on Internet
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`Technology, for which I was the Editor-in-Chief. I was also an Associate Editor
`
`for ACM Transactions on Information and System Security and IEEE Transactions
`
`of Software Engineering, two highly regarded journals in the field.
`
`17. Additional information regarding my background, qualifications,
`
`publications, and presentations is provided in my CV, which I understand has been
`
`submitted as Exhibit 1004.
`
`III. MATERIALS REVIEWED
`In forming my opinions, I have reviewed the ʼ260 Patent and
`
`18.
`
`considered each of the documents listed in the Exhibit List above and all other
`
`materials discussed herein. In reaching my opinions, I have relied upon my
`
`experience in the field and also considered the viewpoint of a person of ordinary
`
`skill in the art as of the time of the earliest claimed priority date of the ʼ260 Patent,
`
`i.e., May 5, 2009. As explained below, I am familiar with the level of a person of
`
`ordinary skill in the art regarding the technology at issue as of that time and all of
`
`the opinions I have formed have been assessed from the perspective of the person
`
`of ordinary skill in the art.
`
`
`
`6
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`IV. DESCRIPTION OF THE RELEVANT TIMEFRAME,
`THE RELEVANT FIELD, AND A PERSON OF ORDINARY SKILL IN
`THE ART
`19.
`
`I understand that the ʼ260 Patent was filed on March 19, 2012, and
`
`claims priority to Provisional Application No. 61/175,489, filed on May 5, 2009.
`
`Thus, for purposes of my analysis, I have treated the effective filing date for the
`
`challenged claims as May 5, 2009. I reserve the right to update my analysis should
`
`Patent Owner assert a different priority date.
`
`20.
`
`I have received and understand the specification, claims, and file
`
`history of the ’260 Patent.
`
`21.
`
`In my opinion, a person having ordinary skill in the art (“PHOSITA”)
`
`at the time of the effective filing date of the ’260 Patent would have had at least a
`
`Bachelors or Masters degree in computer science or electrical engineering or a
`
`related degree and with at least two years training or experience with networking
`
`and file systems. Additional work or research experience can substitute for less or
`
`a different education, and vice-versa. My opinions presented herein are as viewed
`
`through the eyes of a PHOSITA prior to May 5, 2009.
`
`V. TECHNOLOGY BACKGROUND
`22. Before presenting the relevant facts and my opinions relating to my
`
`charge by the court and as a means for providing context and background for my
`
`later discussions, I present an overview of file systems. Specifically, I first define
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`7
`
`

`

`what file systems are and their primary functions (as relevant to this report), then I
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`describe the evolution of file systems, covering topic such as flat versus
`
`hierarchical file systems, centralized versus distributed file systems, client-server
`
`versus peer-to-peer network models, and network proxies. I conclude with a
`
`discussion surrounding the fundamentals of file system functions, how are they
`
`extended to support new features, and how distributed filesystems in peer-to-peer
`
`models operate with such extensions. Note that this exposition is by no means
`
`exhaustive, but simply outlines basic terminology in sufficient depth to frame this
`
`report and my opinions.
`
`A. What are “File Systems”?
`23. A file system (commonly abbreviated as “fs”) is the methods and data
`
`structures that an operating system uses to keep track of files on Storage device
`
`(e.g., hard disk, thumb drive). Ex-1016, p.1. There are many kinds of different file
`
`systems, including minix, xia, ext3, ext2, jfs, xfz, zfs, msdos, vfat, among others.
`
`Id., pp.1-3. While the specific features provided by these varying kinds of file
`
`systems is broad (some of which I cover in subsequent sections of this report), they
`
`all perform the same basic function: to organize and keep track of files.
`
`24. While some of the types of file systems above differ in some respects,
`
`there is a set of common structures and features that they all share (and broadly all
`
`modern file systems today). These structures and features are core file system
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`8
`
`

`

`technologies; said differently, in the absence of these features, it could be argued
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`that such technology could not be classified as a “file system.” These features
`
`include storage, filenames, directories (i.e., folders), and metadata.
`
`25.
`
`Storage, in the context of file systems, refers to components of a
`
`computer that are used to ensure that the data within a file system (and the
`
`computer broadly) is preserved when the system is shut down and restored upon
`
`start. Computer storage (e.g., hard disk and solid-state drives) is typically
`
`characterized as non-volatile, in that the stored information is retained even after
`
`the system is shut down (and thus, power is removed). This is unlike computer
`
`memory (e.g., RAM) which loses its information when the system shuts down. File
`
`systems are not limited to a single storage medium; they can, for example, be
`
`distributed across multiple storage mediums, wherein portions of the file system
`
`are “mounted” across multiple hard disk drives, for example.
`
`26. The core feature of all modern file systems is the notion of a filename,
`
`a collection of alphanumeric (and some special) symbols used in association with a
`
`collection of data. In most common file systems, filenames follow a name-dot-
`
`extension convention; filenames are preceded with a collection of symbols (often
`
`assigned by the user), a period, and an extension (typically assigned by an
`
`application). This convention allows files to be recognizable by users (or other
`
`
`
`9
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`applications) and linked with a particular application so that when, for example,
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`clicked on by a user, the appropriate application is opened and loads the contents
`
`of the file from storage into memory. For example, the filename “Document.docx”
`
`would be interpreted to be a Microsoft Word document.
`
`27. Ostensibly all modern file systems can be classified as, “hierarchical
`
`file systems”, wherein the file system organizes files into a tree-like structure. This
`
`organization is made possible by directories (sometimes also called “folders”), a
`
`structure which catalogues the contents inside of it. Directories can have
`
`directories inside of them (i.e., subdirectories), sub-subdirectories within
`
`subdirectories, and so on. Often, the terms “parent” and “child” are used to
`
`describe the relationship between a directory (the “parent”) and a file or
`
`subdirectory inside of it (the “child”). In hierarchical file systems, files can be
`
`retrieved by navigating from the “root” parent directory (the directory in a file
`
`system that contains all files and subdirectories), through child subdirectories, and
`
`finally to the desired files. This process, navigating from parents to children to find
`
`files, is called a file path (or path). The path of a file describes both the location of
`
`a file within the file system and how the file should be organized when shown to
`
`the user (e.g., as being “inside” of a folder). Technically speaking, file paths are
`
`
`
`10
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`built from inode numbers, which inform the operating system where the data is
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`physically located in computer storage.
`
`28.
`
`Information such as file names and paths are stored alongside files as
`
`metadata (i.e., data about data). This metadata is used by both the file system and
`
`operating system for a variety of functions, such as how files should appear to
`
`users (i.e., their names), who owns the file (so that unauthorized users cannot
`
`access or modify file contents), their path, how much storage space the file
`
`consumes, etc. File system metadata can be read via the Unix system (a popular
`
`computer operating system whose origins date back to 1969) call1 “stat.” A system
`
`call (detailed in Section V.C) is a method by which applications can
`
`programmatically request a service from the operating system. Given a file name,
`
`stat returns information such as the current storage device the file is located on, its
`
`size, when it was created, among other pieces of information. Notably, this
`
`bookkeeping represents critical pieces of information for file systems; if metadata
`
`for a file were to be deleted, the file would be “lost” (in the sense that the operating
`
`system would not be aware of the existence of the file, which would require
`
`
`1 A system call (detailed in Section 1.4) is a method by which applications can
`programmatically request a service from the operating system.
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`11
`
`

`

`complex data recovery tools to rediscover the file on computer storage, if it were
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`even possible).
`
`B.
`29.
`
`The Evolution of File Systems
`It is challenging to lay out the history of file systems since many of
`
`the earliest file systems were tightly coupled with the operating system (that is, it
`
`would be hard to treat the operating system and file systems as different entities,
`
`which is a paradigm that started in roughly the 1970s). Nonetheless, the evolution
`
`of files systems can be characterized through three epochs: migration from a flat to
`
`hierarchical file organization scheme, the ability to retrieve data remotely, and the
`
`seamless sharing of data across multiple devices.
`
`1. From Files to Folders
`30. Perhaps the earliest file system can be traced back to the DECtape
`
`(1964) and CP/M (1973) operating systems. The file systems inside of these
`
`operating systems organized files in a flat manner, i.e., there were no directories or
`
`folders to organize files within. This design was partially due to the storage
`
`capacities of computers at that time. While the systems did not exhibit any
`
`technical reason to be limited to a flat file organization, storage space was scare,
`
`and thus, as a matter of practical limits, directories would not be particularly useful
`
`given the small number of files that could be saved to storage. However, as storage
`
`
`
`12
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`sizes increased, flat file systems rapidly became unwieldly from the sheer number
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`of files applications, users, and operating systems would create.
`
`31. To better managing an ever-growing number of files, the Multics
`
`operating system (Ex-101), dating back to 1965, introduced hierarchical file
`
`systems. Multics influenced the design of many modern operating system date
`
`(e.g., Unix), including its hierarchical file system design, support for long file
`
`names, storage quotas, etc. After Multics, essentially every popular modern file
`
`system implements a directory hierarchy, filenames, metadata, among other
`
`features. To support this organization schema, file metadata would be augmented
`
`to include where a file existed within the hierarchy, so that users and applications
`
`could focus on a desired set of files.
`
`2. Remote Storage for File Systems
`32. With the rise in popularity of computing systems and networks, file
`
`systems were further extended to allow access to storage over a network. In 1984,
`
`Sun Microsystems introduced the Network File System (NFS), a distributed file
`
`system that allowed users to access files stored remotely. Ex-1018. File systems
`
`using network protocols, such as NFS, are “distributed” in the sense that the
`
`methods to retrieve data are identical for both local and remote storage mediums.
`
`Thus, many distributed file systems are “transparent” in that users are not readily
`
`aware if files are retrieved via local or remote storage mediums. Specifically, such
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`
`
`13
`
`

`

`remote files are called virtual files, files whose content is stored remotely, with
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`metadata such as file names, path, size, etc. stored locally. See, e.g., Ex-1018,
`
`12:32-13:11. These files support the transparency features provided by distributed
`
`file systems, in that virtual files appear structurally identical to all other local files
`
`in the file system.
`
`33. While distributed file systems using simple network protocols offer
`
`many advantages, they are limited by: (1) files must be transferred over the
`
`network when a user or application uses them, which may be slow, and (2) the
`
`accessibility of files stored remotely is dependent on the connection to the remote
`
`computing device; a loss in connectivity implies users will no longer have access
`
`to the remote files (in other words, a single point of failure).
`
`34. As networking technology improved, many computing technologies
`
`became increasingly reliant on accessing data remotely. The sheer amount of
`
`available data was problematic for modern computers for at least two reasons: (1)
`
`storage mediums simply did not scale to the vast amount of desirable data (thus
`
`necessitating remote access via protocols such as NFS), and (2) the traditional
`
`client-server model, wherein multiple computers (clients) connected to a central
`
`computer (the server) for a particular service, became cost-prohibitive, as serving
`
`many clients mandated powerful (and expensive) computers to act as servers. One
`
`
`
`14
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`

`

`solution was to distribute work evenly (as much as possible) across clients,
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`wherein clients would also serve the requests of others.
`
`Client
`
`Peer
`
`Client
`
`Server
`
`Client
`
`Peer
`
`Peer
`
`Client
`
`Peer
`
`Figure 1 - Client-Server vs. Peer-to-Peer Models
`
`
`
`Sharing Files with Peers
`3.
`35. This solution became known as peer-to-peer (P2P) computing. As the
`
`name implies, instead of “clients” and “servers”, all participating computers are
`
`considered “peers”, as shown in the Figure above. P2P protocols have been used in
`
`many applications, with the earliest popular application being Napster, a platform
`
`for sharing music among peers, in 1999. Ex-1020. Many file-system-like
`
`applications (such as PAST and IgorFs) rapidly emerged in the early 2000s with
`
`the popularity of P2P, offering the ability to store, retrieve, and use data much like
`
`classical distributed network protocols like NSF that relied on the client-server
`
`model.
`
`
`
`Dropbox Exhibit 1003
`Dropbox v. Entangled Media
`
`15
`
`

`

`36. While P2P applications offered many benefits over the traditional
`
`Case No. IPR2024-00285
`Patent No. 8,484,260
`
`
`client-server model, they were not without limitation. Perhaps the largest
`
`assumption made in any P2P application was that all peers are accessible to all
`
`other peers (that is, even if two peers are across the planet, their traffic will
`
`eventually be routed to one another). As the Internet matured, this assumption was
`
`found to rarely hold true. New techniques for providing security to computers
`
`connected to the Internet, such as firewalls, network address translation (NAT),
`
`dynamic IP addresses, among others, implied an asymmetry broadly incompatible
`
`with P2P: a host that can connect to others does not necessarily mean that others
`
`can connect to the host. In this context, network address translation maps multiple
`
`private IP addresses into a single public IP address, allowing multiple devices in,
`
`e.g., a home network to use a single IP address to access the Internet. Dynamic IP
`
`addresses change from time to time, which is problematic for P2P networks since
`
`IP addresses were used to identify peers (and were largely assumed to be s