Not a Lexis Advance subscriber? Try it out for free.

Pers. Web Techs., LLC v. Apple, Inc.

United States Court of Appeals for the Federal Circuit

March 8, 2019, Decided



 [*1377]  [***1485]   CHEN, Circuit Judge.

PersonalWeb Technologies, LLC (PersonalWeb) appeals the final written decision of the Patent Trial and Appeal Board (Board) in an inter partes review (IPR) proceeding concluding that Apple demonstrated by a preponderance of the evidence that claims 24, 32, 81, 82, and 86 of U.S. Patent No. 7,802,310 (the '310 patent) are unpatentable as obvious in view of two prior art references—U.S. Patent No. 5,649,196 (Woodhill) and U.S. Patent No. 7,359,881 (Stefik). Because one of the Board's key underlying fact findings as to Woodhill's disclosure is not supported by substantial evidence, we reverse.


A. The '310 Patent

The '310 patent explains that in conventional data processing systems, data items such as files are typically identified by their user-created alphanumeric name and/or pathname or location. J.A. 69 at 1:53-2:5. [**2]  Certain problems arise, however, using traditional naming conventions. For example, if one device transfers a data item to a second device using just the name associated with the data item, it is possible that the data item already exists on the second device, and a duplicate of the data item will be created. J.A. 69-70 at 2:63-3:9. The '310 patent contemplates a method and apparatus for resolving this and other concerns by creating a substantially unique identifier for each data item in the data processing system that is independent of the data item's user-defined name, location, etc., but rather is dependent on only the content of the data item itself. J.A. 70 at 3:52-58. The identifier for a particular data item is created  [***1486]  by applying a cryptographic hash function to the  [*1378]  data item. J.A. 74 at 12:21-26. The output of the hash function is the content-based identifier or "True Name," which is "virtually guaranteed" to be unique to the data item. Id. The system uses the content-based identifier alone to determine whether a particular data item is present on the system. J.A. 70 at 3:59-62. When the data item's contents are changed, the content-based identifier of the data item also changes. See [**3]  J.A. 86 at 35:55-63.

The '310 patent explains that content-based identifiers can be used for various purposes in data processing systems, including, for example, to identify data items in a "license table." J.A. 74 at 11:33-43. The patent describes a license table as a two-field database containing a list of content-based identifiers and, for each content-based identifier, a list of users authorized to access the data item associated with the content-based identifier. Id.

Read The Full CaseNot a Lexis Advance subscriber? Try it out for free.

Full case includes Shepard's, Headnotes, Legal Analytics from Lex Machina, and more.

917 F.3d 1376 *; 2019 U.S. App. LEXIS 6919 **; 129 U.S.P.Q.2D (BNA) 1485 ***; 2019 WL 1087004


Prior History:  [**1] Appeal from the United States Patent and Trademark Office, Patent Trial and Appeal Board in No. IPR2013-00596.

Apple, Inc. v. PersonalWeb Techs., LLC, 2015 Pat. App. LEXIS 3518 (Bd. Pat. App. & Interferences, Mar. 25, 2015)

Disposition: REVERSED.


Binary, identifier, backup, content-based, column, content-dependent, plurality, objects, restored, locate, unpatentable, contents, explains, hash, user, Database, combine, digital, files, previous version, comprises, artisan, look-up, skilled

Patent Law, Jurisdiction & Review, Standards of Review, De Novo Review, Substantial Evidence, Nonobviousness, Evidence, Fact & Law Issues, Evidence