Skip to content

Viacom International Inc., et al. v. YouTube Inc., et al.

In this closely followed action brought against YouTube and Google and testing the limits of the Digital Millennium Copyright Act’s safe harbor provisions, among other things, the District Court for the Southern District of New York issued a ruling on several of the plaintiffs’ discovery requests.

On a motion to compel brought by the plaintiffs, and the defendants’ cross-motion for a protective order, the court ruled that the plaintiffs are not entitled to the computer source code that controls the search function on both the YouTube and Google websites. Finding too speculative the plaintiffs’ contention that the code could support their claim that YouTube and Google purposefully designed the search function to facilitate users’ ability to locate infringing material, the court denied the plaintiffs’ motion to compel and granted the protective order, notwithstanding the fact that a stipulated confidentiality order was already in place. Absent a “plausible showing” that Google’s search code can discriminate between infringing and non-infringing content, the disclosure of such a valuable trade secret was unwarranted, the court held.

Similarly, the court denied the plaintiffs’ motion to compel the production of the source code for YouTube’s new “Video ID” program, which allows YouTube to locate infringing content where content owners furnish YouTube with video reference samples. The court rejected the plaintiffs’ contention that the Video ID code would help them demonstrate that YouTube and Google could be doing more to control infringement, explaining that “the question is what infringement detection operations are possible,” with or without reference to Video ID, “not how the Video ID source code makes it operate as it does.”

The court also made the following discovery rulings:

  • Granting the plaintiffs’ motion to compel production of all videos that YouTube has removed from its website in the past, finding that these videos are necessary to identify instances of infringement of plaintiffs’ copyrights.
  • Granting the plaintiffs’ motion to compel production of all video-related data from YouTube’s “logging” database, which contains information related to each instance a YouTube video is watched, including the video identifier and the user’s IP address. The court found that the data was necessary to compare the attractiveness of infringing versus non-infringing videos, which would bear on the plaintiffs’ vicarious liability claim and the defendants’ “substantial non-infringing use” defense.
  • Denying the plaintiffs’ motion to compel production of other information maintained on YouTube’s databases related mostly to non-infringing videos, finding this request overbroad.
  • Denying the plaintiffs’ motion to compel production of the schema employed by Google in its advertising operations.
  • Granting the plaintiffs’ motion to compel production of the schema for Google’s own video service, Google Video. The court accepted the defendants’ argument that the extent to which Google controls infringement on Google Video relates to its ability to control infringing content on YouTube.
  • Denying the plaintiffs’ motion to compel production of videos designated by uploading users as private, such disclosure being barred by section 2702(a)(2) of the Electronic Communications Privacy Act, but granting discovery on the non-content data associated with these private videos.