Bedrock v. Google Verdict – What It Covers, What It Means

By on Thursday, May 5th, 2011 in Blogs and Opinions, Open Source Licensing | Keywords: ,

In recent years federal courts across the country have been filled with FOSS-related patent claims. Last month, a verdict in the Bedrock v. Google case renewed calls for reform of the US patent system. What exactly were the issues in this case, and what impact is this verdict likely to have on the open source world?

In a two-page jury verdict form returned on April 15 in the Bedrock Computer Technologies, LLC (BCT) suit, a jury held that Google infringed US Patent No. 5,893,120 (held by BCT), determined that the patent is valid, and stated Google must pay $5 million as damages for past infringement. The case is far from over, however, as claims continue to proceed against other defendants and further juries in the case may return additional verdicts that bear watching.

Unfortunately, large portions of the available record in the case have either been redacted or sealed, but BCT in its original seven-page complaint (and in Exhibit A, 15 pages, detailing the patent itself; Amended Complaint; and Corrected First Amended Complaint for Patent Infringement) asserted in its general allegations that it holds “all right, title and interest in and to United State Patent No. 5,893,129, entitled ‘Methods and Apparatus for Information Storage and Retrieval Hashing Technique with External Chaining and On-the-Fly Removal of Expired Data.’” The complaint further alleges that “[d]efendants infringe the Patent-in-Suit directly, contributorily and/or by active inducement by importing, manufacturing, using, marketing, distributing, selling, and/or supporting products and/or services that fall within one or more claims of the ’120 Patent.”

Exactly what code does BCT’s patent cover? BCT’s patent-in-suit, which is attached in full to the Complaint as Exhibit A, and available separately for viewing at the USPTO and elsewhere, notes in the abstract that “a method and apparatus for performing storage and retrieval in an information storage system is disclosed that uses the hashing technique with the external chaining method for collision resolution.” BCT has claimed that the methods used in the Linux kernel itself, apparently starting with the version 2.4.22.x root and impacting updates and version since, utilize BCT’s patent method in on-the-fly garbage collection and other operations.

Jury verdict notwithstanding, how reasonable is BCT’s claim? Frankly, the method and technique described would seem eminently subject to an attack as to novelty, and more particularly the nonobviousness requirement necessary before a patent may be granted. Essentially “nonobviousness” means that the invention would not have been obvious “to one of ordinary skill in the art.” Indeed, BCT’s own patent highlights that “external chaining is discussed in considerable detail … by D[onald] E. Knuth, in Data Structures and Program Design, Second Edition….” For any patent examiner with a computer science background the citation to Knuth’s foundational tomes, veritable bibles in the programming world, should immediately have sent up a red flag leading to further detailed examination. While the patent prosecution that occurred is unknown to us, programming for and utilizing linked lists and hashing functions are fundamental practices. Courts have held a “[a] showing of obviousness requires (1) a motivation or suggestion to combine or modify prior art references, coupled with (2) a reasonable expectation of success.” — Brown & Williamson Tobacco Corp. v. Philip Morris Inc., 229 F.3d 1120, 1124-25, (Fed. Cir. 2000).

Download the Open Source Support Evaluation Kit

Linux stalwart Red Hat immediately recognized the potential downside, and affirmatively injected itself into the case, intervening (see Red Hat Inc.’s Complaint in Intervention) due to the fact that a number of customers utilizing Red Hat Enterprise Linux were named as defendants in BCT’s suit. It sought, and failed, to have BCT’s patent declared invalid. If the past is prologue, what will likely happen as a direct response to the jury verdict is that numerous companies using Linux and variants, and otherwise loathe to either engage in lengthy speculation and costly patent litigation or to await the results of any appeal, may “pay up” and obtain a license to BCT’s patent as an “insurance policy” against any future lawsuit on the patent.

Various commentators have speculated that the patent is more likely to impact Linux on the server side, notably Google’s own server farms, than the client or handheld side of OS operations. However, garbage collection is a basic necessary housekeeping function of every OS, and indeed a means of speeding garbage collection is no doubt even more important on the mobile side, where CPU horsepower and memory are often severely constrained.

An Unfinished Story

While last month’s verdict marks one milestone on the case, litigation is ongoing. BCT placed Google in its patent crosshairs when it filed suit in mid-2009 against the search giant and numerous other parties (Softlayer Technologies, Inc., Citiware Technologies Solutions LLC, Yahoo! Inc., MySpace Inc. Amazon.com Inc., PayPal Inc., Match.com LLC, AOL LLC and CME Group, Inc.). The suit was filed in US Federal court in the Eastern District of Texas – infamously known to patent litigators as a “rocket docket” for its often quick pro-plaintiff results. This suit followed a long line of BCT actions in recent years against other tech giants and companies, primarily claiming patent infringement.

The recent verdict garnered attention not only for the fact that Google lost a patent case, albeit with a relatively minor dollar award attached, but because of the potential broader implications for use and implementations of Linux, given that BCT asserted that a portion of the Linux kernel, as utilized by Google in both its existing server farm and Android mobile OS, infringed its patent.

BCT’s case continues on, and it’s likely that we haven’t seen the end of Google’s response, BCT’s further lawsuits, or the Linux industry’s collective pushback to this and other FOSS-related patents. It’s also likely that the Linux kernel developers will craft non-infringing workarounds to handle the specific chores that are claimed to infringe.

Related posts:

  1. Top Ten Open Source Legal Issues for 2008
  2. Google Web Toolkit and Web Services: The XML Way
  3. Google Web Toolkit and Web Services: The JSON/JSONP Way
  4. Internet Law Part I: Copyright Law
  5. The Open Source Model Really IS a New Software Paradigm

Related Open-Source Packages

Richard L. Santalesa

Richard Santalesa is Senior Counsel at the InfoLawGroup, where he focuses on intellectual property, e-commerce, data security and privacy issues. In a previous incarnation he was a technology editor, journalist, programmer and prolific author.

Leave a Reply

© 2012 OpenLogic, Inc. | Licensing | Privacy Policy | Terms of Use

Bad Behavior has blocked 2283 access attempts in the last 7 days.