How bootleggers are beating content ID system of sites like Youtube

Using Unicode text to fool text searches is the coolest method.

When ordinary Youtubers play some songs in the background or use music to demonstrate something, they get a copyright strike. How then are bootleggers uploading entire lengths of movies to YouTube and other sites? How does their content remains online for years without any problem?

The film and music industry encodes certain indiscernible fingerprints to their content released to theatres, discs and Internet. Media-hosting sites can trawl their site and identify copyrighted material using these fingerprints. They also provide industry officials a login and search system to identify this content and take them down using DMCA and other legal provisions.

Some content owners have instead opted to profit from bootlegger rather than taking down the bootleg material. Sites like Youtube allow copyright owners to run ads on identified bootleg content and take whatever revenue that would have gone to the channel owner.

Many old movies and TV serials are too politically incorrect for today’s snowflakes. They cannot be hosted on video streaming sites like NetFlix. Even something as recent as Friends is “problematic” for snowflakes. If a movie has words like ‘faggot’, then it is unlikely that Netflix would want to host or advertise it. So, content owners are unlikely to profit from it. With every year, kids are getting more stupid (PC) and more of the old stuff is getting removed from popular video streaming sites.

In such cases, bootleg copies made from old VHS tapes uploaded to Youtube provide content owners an alternative revenue stream. This is outside the content owners’ official channel where they may host videos for free or payment. Money made from selling dogs does not bark. All money is good money. Everyone’s is happy. That is why I do not feel bad about watching old 80s movies from Youtube.

In my journeys to find old 80s gold and trash, I have encountered some pretty smart techniques used to defeat the content ID system, which is both automatic and manual. Here they are:

  • Picture-in-picture: They play the movie inside a rectangle smaller than the whole frame. If you see the HD version, then the rectangle is still big enough to be watchable.
  • Flipped video: They horizontally flip the video. This is not a problem unless you try to read text.
  • Noise: Some low-volume noise is mixed with the audio stream.
  • Slowed audio and video: This is discernable initially but not a big impediment to watch the content.
  • Unicode titles: Unicode has several codepoints where the English alphabet is repeated. It looks like ordinary text but a text search by a copyright owner will not find them. For example, ‘HELLO’ is not the same as ‘ᴴᴱᴸᴸᴼ’. No, it is not superscript font style. These letters in Unicode are like that.
  • Blurring and shadowing: Some videos are deliberately blurred. Some others have radial shadow. The content in the middle is still clear but on the sides it is dark.
  • Cropping: Video around the edges are cut. This and the previous ones are the worst forms. Why do they even bother?

All of these techniques seem to destroy the ID fingerprint or make manual identification difficult. Those channel owners continue to make money.

In a lot of cases, the content owners do not care. Old VHS tapes degrade easily. Going after those bootleggers is a waste of their time. The market for old content is not increasing anyway.

Comments on this blog are moderated by government agents. The Moral Volcano Essential Reading List will explain why.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s