AN OVERBROAD PATENT ON LOCATING DATA ON A COMPUTER - This application seeks to patent the idea of... locating data for applications on a computer! 10 minutes of your time can help narrow US patent applications before they become patents. Follow @askpatents on twitter to help.

I am not sure whether "locating for application-specific data" can be patentable. This sounds like inventing a wheel.

  • Patent: US 20130166517 A1
  • Title: Systems and methods for locating application specific data
  • Prior Art Date: Seeking prior Art predating December 12, 2012
  • Open for Challenge at USPTO: Open through Dec 27, 2013

Claim 1.

A system for locating for application-specific data comprising:

  • at least one data storage device, having unidentified data stored therein;

  • at least one processor operatively coupled to the at least one data storage device, the at least one processor configured to:

(i) access unidentified data from at least one data storage device;

(ii) examine the unidentified data to detect at least one application-specific data pattern associated with at least one application;

(iii) for each detected application-specific data pattern, execute an application-specific validation process to determine whether the unidentified data includes valid data associated with a corresponding application; and

(iv) if it is determined that the unidentified data includes valid data associated with the corresponding application, then recover the valid data.

What is good prior art? Please see our FAQ.

Want to help? Please vote or comment on submissions below. We welcome you to post your own request for prior art on other questionable US Patent Applications.

5 Answers 5


In addition to the myriad of programs that already do this (http://pcsupport.about.com/od/filerecovery/tp/free-file-recovery-programs.htm), here is a website explaining the exact process that the patent claims to have invented (which, according to archive.org, was written in 2002): http://www.ntfs.com/disk-scan.htm

It also fails under the "non-obvious" requirement. When searching a filesystem for application-specific deleted files, the first thing a programmer would do is search for data patterns that identify said application, and then validate what is found to determine if it's really what is being sought.

This also appears to already be patented: https://www.google.com/patents/US20100257146


At least the majority of the claims here were covered by Norton Utilities v1.0 in 1982. UnErase/FileFix specifically.



the applicant is a supplier of forensic software, their primary product is a tool to search computer hard disks for artefacts of computer use (internet history primarily).

the functionality in the patent can be seen in many other forensic programs such as Encase, FTK and Netanalysis all these programs have been around for a good while.


That patent talks about recovering data pertaining to an application, like looking for and recovering Microsoft word documents.

Recuva (https://www.piriform.com/recuva) a free tool for recovering data has been doing this for years, as have many others.


The GPL software PhotoRec does just about everything mentioned in the software patent: http://www.cgsecurity.org/wiki/PhotoRec , https://en.wikipedia.org/wiki/PhotoRec

PhotoRec looks at the raw data on disk. For example, if it sees a JPEG (certain application) header, it looks around nearby -- possibly scanning past a comment -- for the end of the header, actual photo data, and footer. It then recovers the files. PhotoRec also does similar things for other files types, has user-selectable options, and can scan free space on a disk (patent: "marked as being available for storing new data").

The public domain software Foremost and its successor Scalpel, are similar: http://foremost.sourceforge.net/

The idea of allowing the data to be slightly off and letting the user determine a threshold of similarity is a very common technique going by many names such as "fuzzy hashing/searching" (http://computer-forensics.sans.org/summit-archives/2010/eu-digital-forensics-incident-response-summit-jesse-kornblum-beyond-fuzzy-hashing.pdf). It is implemented by most virus-scanning software at a low level -- see http://www.google.com/patents/us7284273 which specifically uses the term "number of allowed mismatches." In the virus-scan scenario, "desired applications" are executable files, and vulnerable applications such as Word scripting language interpreters, etc. Virus scanners often let you decide where to search and how strict they operate. They often scan raw memory which is a form of "suitable data storage" which has space "marked as available for new data" as mentioned in the patent, as well as files on disk.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .