YouTube creator’s misrepresentation case over DMCA takedown notices lacked certain key evidence

Plaintiff ran a YouTube channel. Defendant sent takedown notices under the Digital Millennium Copyright Act (DMCA) to YouTube alleging that plaintiff infringed defendant’s copyright. Plaintiff sued defendant under 17 U.S.C. §512(f), which provides for recovery when a person sends a DMCA takedown notice based on “knowingly materially misrepresented” facts.

The district court granted defendant’s summary judgment motion. Plaintiff sought review with the Eleventh Circuit. On appeal, the court affirmed the lower court’s finding that defendant did not knowingly materially misrepresent facts when it sent the DMCA takedown notices to YouTube.

The DMCA’s good faith standard

The DMCA relieves online service providers (such as YouTube) of liability for storing copyright infringing content if, among other things, the service provider properly responds to the copyright holder’s takedown notice. The notice must state that the copyright holder has a “good faith belief” that the identified content infringes. The copyright holder can be liable for damages if it “knowingly materially misrepresents” that the identified content infringes. Copyright law provides that if use of content is fair use, then it is not infringing.

DMCA good faith

In 2016, the Ninth Circuit evaluated the questions of what it means for a DMCA takedown notice sender to have a “good faith belief” and to “knowingly materially misrepresent” that content infringes. In that case (Lenz v. Universal Music Corp., 815 F.3d 1145 (9th Cir. 2016)), the court held that having a “good faith belief” that certain content infringes requires that the copyright holder, before sending a takedown notice, consider if the potential infringement is a fair use. And the court held that failure to consider fair use before issuing a takedown notice is a misrepresentation of copyright infringement.

What diligence must a takedown notice sender undertake?

What kind of belief must the copyright holder have to be in line with the conduct required of the DMCA? The Lenz court held that a copyright holder’s subjective good faith belief that a use is both unauthorized and not fair use serves as a complete defense to a misrepresentation claim.

Counsel’s consideration of fair use was sufficient

In this case, defendant’s attorneys testified in affidavits that they investigated whether plaintiff’s YouTube uploads infringed. And they testified that they evaluated whether the uploads were fair use. So the court found that defendant, by relying on their attorneys’ investigations and opinions, reasonably believed in good faith that plaintiff’s videos infringed and did not constitute fair use, prior to filing the takedown notices.

The court found that plaintiff’s evidence did not show otherwise. Plaintiff testified in her deposition that she “knew” defendant did not consider fair use. She said that if defendant did consider fair use, it would not have sent the DMCA takedown notices. In the court’s view, this proffered evidence was just a bare and speculative statement, not evidence. It found that plaintiff failed to provide any meaningful evidence that defendant did not consider the fair use doctrine and that it did not have good faith when it submitted the takedown notices.

Johnson v. New Destiny Christian Center Church, Inc., 2020 WL 5289881 (11th Cir., September 4, 2020)

See also:

Website operator not liable for copyright infringement despite lack of DMCA safe harbor protection

About the author:

Evan Brown is a Chicago technology and intellectual property attorney. Call Evan at (630) 362-7237, send email to ebrown [at] internetcases.com, or follow him on Twitter @internetcases

Want your online agreements to be enforceable? Keep good transaction data.

Chicago internet attorney Evan Brown

A recent court decision underscores the importance of building online e-commerce platforms with the ability to reliably gather information about transactions. The case also says some troubling things about open source.

Plaintiff loaned money in exchange for the borrower assigning its accounts receivable to plaintiff. As part of plaintiff’s services, it provided a platform for its borrower to generate and send invoices to the borrower’s customers. The borrower began generating fake invoices, and one of its customers — the defendant in this case — refused to pay. There was a dispute over whether defendant had accepted or rejected the invoices using plaintiff’s invoice platform.

After a trial, the judge ruled in favor of defendant. The court found that the digital data showing whether defendant had accepted or rejected the invoices was unreliable. The court found credible the testimony of one of defendant’s employees that he never clicked “I agree” on the fraudulent invoices. And there was no good database evidence that he had.

Plaintiff sought review with the Court of Appeal of California. On appeal, the court affirmed, agreeing that the data was unreliable, and further commenting on the problematic use of open source software in plaintiff’s online invoice platform.

The court of appeal found that substantial evidence supported the lower court’s findings. Specifically, it agreed with the lower court’s findings that the defendant’s employee never clicked on the “I agree” button to accept the fraudulent invoices. The court also credited the lower court’s finding that the data was unreliable in part because plaintiff’s website was developed from open source code, and that the developer made untested changes to the software on a weekly basis.

The treatment of the open source aspect is perhaps unfortunate. One unfamiliar with open source would read the court’s opinion as an indictment against open source software’s fundamental reliability:

Open source code is problematic because anonymous people on the internet design it, and “holes” are not fixed by vendor updates. Notifications that there are issues with the code may not go out.

The lack of reliability of the data in this case was not due to the fundamental nature of open source. (We know that open source software, e.g., Linux, powers essential core features of the modern internet.) So it is unfortunate that future litigants may look to this case to argue against vendors who use open source solutions. Fortunately, the case is not citable as precendent (many California Court of Appeal cases are not citable). But the court’s negative treatment of the nature of open source is a troubling example of how a judge may be swayed by a technological red herring.

21st Capital Corp. v. Onodi Tooling & Engineering Co., 2015 WL 5943097 (Not officially published, California Court of Appeal, October 13, 2015)

Evan Brown is a Chicago attorney advising enterprises on important aspects of technology law, including software development, technology and content licensing, and general privacy issues.

Photo by Flickr user bookfinch under this Creative Commons license.

Tweet served as evidence of initial interest confusion in trade dress case

The maker of KIND bars sued the maker of Clif bars alleging that the packaging of the Clif MOJO bar infringes the trade dress used for KIND bars. Plaintiff moved for a preliminary injunction, but the court denied the motion. But in its analysis, the court considered the relevance of a Twitter user’s impression of the products. Plaintiff submitted a tweet as evidence in which the user wrote, “I was about to pick up one of those [Clif MOJO bars] because I thought it was a Kind Bar at the vitamin shop ….” The court found that this type of initial interest confusion was actionable and therefore the tweet supported plaintiff’s argument.

KIND LLC v. Clif Bar & Company, 2014 WL 2619817 (S.D.N.Y. June 12, 2014)

Evan Brown is an attorney in Chicago, advising clients on matters dealing with trademark protection and enforcement, technology, the internet and new media. Contact him.

Another court puts an end to a social media discovery fishing expedition

480px-Old_photo_of_woman_holding_a_fisherman_caught_fishPlaintiff sued a construction company and certain municipal authorities for negligence and loss of parental consortium after her toddler son was seriously injured in front of a construction site. Defendants sought broad discovery from plaintiff’s Facebook account, to which plaintiff objected in part. But the trial court required plaintiff to answer the discovery. So plaintiff sought review with the appellate court. On appeal, the court overturned the trial court.

It held that defendants’ discovery requests were overbroad and compelled the production of personal information that was not relevant to plaintiff’s claims.

Defendants had sought copies of postings on plaintiff’s Facebook account dealing with:

  • Any counseling or psychological care obtained by plaintiff before or after the accident
  • Relationships with [her injured son] or her other children, both prior to, and following, the accident
  • Relationships with all of plaintiff’s children, “boyfriends, husbands, and/or significant others,” both prior to, and following the accident
  • Mental health, stress complaints, alcohol use or other substance use, both prior to and after, the accident
  • Any lawsuits filed after the accident by plaintiff

The court observed that one of the defendants’ arguments to the trial court essentially conceded it was on a fishing expedition. The attorney stated, “These are all things that we would like to look under the hood, so to speak, and figure out whether that’s even a theory worth exploring.” And the magistrate judge in the trial court (though ordering the discovery to be had) acknowledged that “95 percent, or 99 percent of this may not be relevant,” and expressed some misgivings at the possibility that large amounts of material might have to be reviewed in camera.

Finding that the trial court order departed “from the essential requirements of the law” because it was overbroad and required the production of irrelevant personal information, the court quashed the discovery requests.

Root v. Balfour Beatty Const. LLC, — So.3d —, 2014 WL 444005 (Fla.App. 2 Dist. February 5, 2014)

Does publication on the web give rise to “access” in copyright infringement analysis?

2003lookbackPlaintiff sued defendant for copyright infringement. Defendant moved for judgment on the pleadings (which is essentially the same thing as a motion to dismiss for failure to state a claim except it is after defendant files an answer). Defendant asserted that plaintiff had not pled copyright infringement because under the Seventh Circuit’s “substantial similarity” test to demonstrate infringement, plaintiff had not pled defendant had “access” to the allegedly infringed work.

The court rejected defendant’s argument and denied the motion for judgment on the pleadings on this issue.

In some copyright infringement cases, a plaintiff may not have direct evidence that the defendant committed infringement. In those situations, a finder of fact may infer that infringement has occurred when it is shown that:

  • the defendant had access to the copyrighted work; and
  • the accused work is substantially similar to the copyrighted work.

In this case, defendant argued it never had access to plaintiff’s designs that it was alleged to have infringed. But the court considered the online publication, 11 years ago, of plaintiff’s designs, to find access for purposes of the motion for judgment on the pleadings:

With regard to online publication, in 2003, [plaintiff] first published the [allegedly infringed work] at [its website]. The Internet already was widely used and accessible at that time. Because the non-movant is entitled to reasonable favorable inferences in evaluating a motion for judgment on the pleadings, the online publication is enough to establish access for purposes of denying [defendant’s] motion for judgment on the pleadings.

The court’s decision provides no meaningful analysis as to why publication on the web gives rise to access. It states the finding above in such a conclusory manner as if to indicate it sets forth some per se rule. But one is left to wonder whether other factual nuance would change the answer to the inquiry: What if publication were in 1993 rather than 2003, at a time when many, many fewer people were on the web? What if the publication were behind a paywall for which defendant had no authorization to pass? What if defendant pled it did not utilize the web for this sort of information, or, even more compellingly, not at all?

Skyline Design, Inc. v. McGrory Glass, Inc., 2014 WL 258564 (N.D.Ill. January 23, 2014)

Judge who was Facebook friends with victim’s father did not have to recuse himself

Case provides valuable guidance to judges on how to responsibly handle social media connections and communications.

Judge sent defendant to prison for assaulting defendant’s girlfriend. Defendant appealed his sentence claiming, among other things, that the judge was not impartial, given that the judge was Facebook friends with the girlfriend-victim’s father, and that the two of them had communicated through Facebook’s private message feature. The appellate court held that the judge did not err by not recusing himself.

The appellate court found that no rule prohibited the judge from being Facebook friends with the victim’s father. And the judge followed the proper procedure concerning the private message by:

  • discontinuing reading it once he realized it pertained to the case
  • warning the victim’s father not to communicate ex parte in that manner
  • printing the message out and placing it in the case file
  • notifying counsel for the parties

Moreover, the private message was not adverse to defendant, but actually asked for leniency. On these facts, the court found an insufficient showing of bias to find reversible error.

Youkers v. State, — S.W.3d —, 2013 WL 2077196 (Tex.App. May 15, 2013)

Court considers Yelp posting as evidence of potential consumer confusion in trademark case

Posting by confused consumer was not hearsay.

You Fit, Inc. v. Pleasanton Fitness, LLC, 2013 WL 521784 (M.D.Fla. February 11, 2013)

In a trademark case between competing health clubs, the court considered a Yelp posting in entering a preliminary injunction, finding that while the anonymous posts were not conclusive evidence of actual confusion, they were indicative of potential consumer confusion.

The dispute centered over the use of “You Fit” and “Fit U” for health clubs. A Yelp user posted the following:

I am soo [sic] confused. I was a member at Youfit in [Arizona] and when I moved back to [California] I saw this place by my house and thought great my gym is here! When I went into the gym, I realized it was called Fit U. They use the same basic color scheme on their sign and the motto seemed the same. When I asked the girl at the desk, … [she] said her owner created this brand. I said what are you [ sic ] rates? Seemed very similar to me as when I was a member at Youfit. Very confusing and a big let down.

The court rejected defendant’s hearsay argument. It noted that affidavits and hearsay materials which would not be admissible evidence for a permanent injunction may be considered if the evidence is appropriate given the character and objectives of the injunctive proceeding. With no analysis as to why, the court found the Yelp posting appropriate to consider at this stage of the case.

Moreover, the court observed in a footnote that the Yelp post was not hearsay to begin with. It was not being offered to prove the truth of the matter asserted, but to demonstrate the consumer’s confusion — a then-existing mental state of the declarant, which is an exception to the hearsay rule. This is an interesting finding. The hearsay and non-hearsay uses of the post both turn on the same content, particularly the statement “I am soo [sic] confused.” That statement is the matter asserted (and in such capacity, excludable hearsay). And it is also the mindset of the declarant (and in such capacity, subject to an exception to the hearsay rule).

The court’s opinion does not address what one might see as the real problem with the Yelp evidence — its authenticity. Perhaps the parties did not bring that up. But one does not have to venture far in imagination to see how a crafty plaintiff could generate, or direct the generation, of self-serving social media content that would be helpful as evidence in a litigated matter.

See also: Customer reviews on social media provide important evidence in trademark dispute

MySpace evidence was inadmissible hearsay

Musgrove v. Helms, 2011 WL 1225672 (Ohio App. 2 Dist. April 1, 2011)

An Ohio domestic relations court ordered an ex-wife to pay her ex-husband child support. Based on evidence that the ex-wife’s income had increased, the court increased the amount of support she had to pay. One of the pieces of evidence the court relied on was information from the ex-wife’s MySpace page where she had stated her income was “less than $30,000.” (This comported with other evidence suggesting her income was around $29,000).

The ex-wife sought review of the order increasing child support with the appellate court. On appeal, the court found the MySpace page to be inadmissible hearsay, and vacated that portion of the order.

The finding turned on a nuance of the rules of evidence pertaining to hearsay. Generally, hearsay is inadmissible as evidence, but there are exceptions. One of the exceptions is statements made by the declarant that are against her interest. The court found that although the MySpace information was used in a way adverse to the ex-wife’s interest (i.e., to increase her support obligation), as a declaration it was not adverse to her interest because it was not an assertion of fact which was by its nature contrary to her interest.

So this case is a reminder that notwithstanding any increased interest in the discoverability of social media evidence, the rules in place may serve to render the information discovered ultimately useless later in the litigation.

Court says law firm did not eavesdrop on employee phone calls

Bowden v. Kirkland & Ellis, 2011 WL 1211555 (7th Cir. April 1, 2011)

Two former employees of a law firm sued the firm for violation of the Electronic Communications Privacy Act, 18 USC 2510 et seq. and for violation of the Illinois Eavesdropping Act, 720 ILCS 5/14-2. The district court granted summary judgment in favor of the law firm. The former employees sought review with the Seventh Circuit. On appeal, the court affirmed the grant of summary judgment.

The court held that the former employees’ evidence of eavesdropping raised no more than a “theoretical possibility” of a violation. Even one of the strongest experts in the case triple hedged his testimony, saying the records “could indicate the potential that interception may have occurred.” So the grant of summary judgment was proper.

The plaintiffs had also raised an electronic discovery issue, namely a claim that the law firm spoliated evidence by destroying a server that contained phone records relevant to the case. The court rejected that argument, finding no credible evidence that the destruction was undertaken in bad faith.

Plaintiff failed to show that Facebook pics supported hostile workplace claim

Jabbar v. Travel Services, Inc., 2010 WL 3563112, (D.Puerto Rico September 10, 2010)

Plaintiff sued her former employer for racial discrimination. The court granted summary judgment in favor of the employer, finding there was not enough evidence to go to trial on plaintiff’s claim. Plaintiff asked the court to reconsider the judgment against her. The court held its ground.

One of the assertions that plaintiff made was that someone from work had posted a discriminatory comment on a Facebook photo taken at a company outing.

The court found there was no evidence apart from plaintiff’s own deposition testimony that the company’s official policy was to upload photos to Facebook. And there was no evidence as to who owned the Facebook account in question.

So the court found no basis to overturn its earlier determination that plaintiff failed to establish a prima facie case of employment discrimination.

Scroll to top