We recently published three videos showcasing a live performance of classical music performed by a Fife musician who is seeking a summer scholarship in the U.S. The footage captured is to be used as part of their audition tapes.
Unfortunately, these videos have been flagged by music publishing companies claiming copyright ownership of the performances, a claim which we are in the process of disputing. One aspect of these claims, is that the videos now have advertisements displayed alongside the video, with the advertising revenue going directly to the music publishers making the copyright claim.
Ignoring the financial aspects for a moment, are classical performances infringing copyright? It really depends on how the recording was obtained.
Under UK copyright laws most work is protected for the life of the author plus an additional 70 years after their death before it is considered Public Domain. Copyright law will differ by nations, with the US stipulating 75 years after death before a work will be considered Public Domain.
Therefore, classical music is Public Domain and can be freely performed, recorded, sold and distributed without infringing copyright. However, there is still copyright attached to classical music on the basis of recordings and composition.
To clarify the differences, if you take a CD recording of a classical composition and republish this on a YouTube video (or other media medium) then you will be infringing the copyright protections of whoever you took the recording from. Basically there is two copyrights that exist in music recordings. Rights can be split into recording and arrangement. The arrangement is basically someone’s interpretation of the composition.
If you decide to sing a piece of music that is in the public domain, then that is your arrangement of the music and you are then the copyright holder of that arrangement. If a record company release a classical CD, they make their own arrangement and recording of that music, and they will own the rights to that specific recording they made.
The three videos we published were recorded live, therefore copyright of the arrangement lies with the performer which we have obtained the rights to display and distribute.
The flagging of videos is done through a system called ContentID, a system which is made available to select creators and rightsholders (generally companies that hold a large amount of rights, such as music publishing companies). The system allows Rights Holders to remove the infringing content or monetize the content, which in October 2014 reached a milestone by paying out $1 billion to right holders from it’s original launch, 7 years ago.
The three videos which were incorrectly flagged by the ContentID system are still available during the dispute period, but these are now being monetized, with any advertising revenue generated going to the companies that have made the claims. Since the copyright claims are incorrect, the adverts will be removed and restrictions on the content will be lifted once the disputes has been fully resolved. However any advertising revenues generated during the flagging and dispute period will be kept by the claimants even though their claims are false.
The three videos in question will not raise any significant amount of cash for the claimants due to lower viewing figures, if they are lucky it will be a few pennies, but with over 1 billion active users each month, watching 9 billion hours of video, ContentID can be a lucrative income stream for companies regardless of whether claims on videos are false positives.
This has led to an interesting scenario whereby the ContentID system can be gamed by fraudulent claims which is summed up by Mona Ibrahim from Gamasutra;
Under the “traditional” DMCA take-down procedure, a rights holder will submit notice of infringement to the website claiming DMCA safe harbor protection. After a (granted, predominantly cursory or non-existent) review the website will remove the user generated content from the site. The user who uploaded the content would then have to submit a counter-notice (aka a “put back up” notice) stating that the content is non-infringing or is fair use. It’s up to the website to decide whether to put the content back up. If they do decide to put the content up, the original content owner would have to take actual legal action by filing a civil claim in an appropriate jurisdiction. The content owner can then submit a copy of the complaint to the website, which may elect to remove the allegedly infringing content once again.
The Content ID Claim system, however, doesn’t follow this procedure. Instead it allows content owners (or even fraudulent claimants) to hijack and monetize user-generated content even if the user’s inclusion of the underlying work constitutes nominal or fair use. This is, obviously, very different from removing potentially infringing content. Ironically, it facilitates a system whereby the original content owner becomes an infringer.
Ultimately our three videos will be un-flagged, but in the meantime the companies that made the claims will be profiting from someone else’s work, and if they make enough claims against the many millions of videos uploaded to YouTube on a daily basis, this can be a very lucrative income stream and will have undoubtedly seen a rise in fraudulent claims scraping pennies from each video that is flagged.
The users that will be most impacted is the wave of YouTubers that derive their main income from their content generation on YouTube, with some of the more famous YouTubers finding that 15% of their content is getting the revenue redirected to third parties, that have very disparate or no ties to the content in question.