YouTube is a platform and a community which live off of reuploads. One might even go so far as to say that reuploads have been a key to YouTube’s success and reuploads themselves have been a product of YouTube. With recent events in mind, now is a good time as ever to re-examine the mutually beneficial relationship between YouTube and the practice of reuploading.

In 2005, YouTube started off as a small video-sharing site. At the time few people would have been able to predict that it would grow to be the 2nd most popular site on the web and yet here we are! One factor Co-founder Jawed Karim attributes their success to is timeliness. In particular, he thinks YouTube came at a time when clip sharing became very common. To quote from a talk he gave in 2006:

The “clip culture” you see now is basically this demand that you can find any video at any time and you can share it with other people, or you can share your own videos with other people. […]

There were a couple of events in 2004 that kind of fueled this. One was this [wardrobe malfunction]. So this, of course, happened on television, but it only happened once and never again. And so for anyone who wanted to see it after that, well they had to find it online. The other big event I remember is this [Stewart on Crossfire] interview. And you know this was also shown on television once and not after that. Everyone was talking about it, but people who missed it really wanted to be in on the joke so they would try to find it online…

Jawed Karim, r | p 2006: YouTube: From Concept to Hypergrowth (25:15)

YouTube was able to meet this clip demand, acting as a universal replay button for any clip people could imagine. It’s no coincidence that obscure/lost media fanatics were flocking to the site not soon after its launch. From Sesame Street shorts to TV pilots, old footage quickly piled up! YouTube had an entire subculture of video remixes called YouTube Poops, which were made from recycled clips from old TV shows and games.

Alas this clip culture was both the boon and bane of early YouTube. As users uploaded these clips liberally, some of the owners and rights holders of the original source material of said clips came to view this practice as copyright infringement. This tension led to the infamous
Viacom vs. YouTube case in 2007, where media giant Viacom sued YouTube and Google for $1 billion in alleged damages! If you are looking for a good summary, EmpLemon did a video on it a few years ago.

Viacom did not actually win the case, in fact it came to light that they had taken advantage of clip culture for a stealth marketing campaign of their own. But the whole ordeal had lasting effects on YouTube. In an attempt to appease intellectual property owners, YouTube introduced their content ID system, then called video ID, for automatically detecting copyright infringing videos.

(Video Identification ~ YouTube Advertisers. If the above video is unavailable please use this Wayback Snapshot)

All of a sudden, videos on YouTube became a whole lot more volatile. This automated system did not only take down a lot of infringing material, but it also hit false positives, matching short-length clips, remixes and video reviews as well. At one point you would have been lucky to have had a few of your videos deleted, as opposed to having your whole channel terminated for seemingly having one too many copyright strikes. Yet clip culture on YouTube has somehow been able to endure, even beyond this era.

The term reupload probably first entered the YouTube lexicon when users began uploading new, higher quality versions of videos on their channel as YouTube kept introducing higher caps to video quality. These YouTube upgrades came around the same time as Content ID, so you will find cases where the reupload of a video has survived but the original has been deleted.

It wasn’t just the video makers themselves who were reuploading though, soon other users also began reuploading downloaded copies they had made of their favorite YouTube videos. This was not merely due to fans appreciating content from their fellow YouTubers, but also due to the fact that the frequent channel terminations could deny the original uploader the right to reupload their channel’s videos in the first place.

Ironically, the term reupload soon was associated with degredation in quality as people began reuploading videos over and over again. There’s even a Gizmodo article about it from 2010. There have also been people who have complained about their work being reuploaded without permission or credit, worse yet plagiarised. Clearly, reuploads are a great power that came with great responsibility. Still, many diligent channels are dedicated to preserving the memory of original content through its reuploads.

The fear of such memories being lost through mass-deletions looms over YouTube, even today. Early ContentID was certainly not the last disaster to plague YouTube videos. Hacker pranks, copyright trolls, the Adpocalypse and Elsagate controversies have all taken their toll on many unfortunate channels. Today, we once more find ourselves on the brink of a scene similar to a mass-deletion, with the mass-privating of unlisted videos uploaded prior to 2017.

A few years ago it was discovered that YouTube video IDs were being generated according to a certain pattern and it was thus theoretically possible to predict video links. This presented a problem for unlisted videos, which were meant to be videos that were to be shared by link only.

Unlisted videos are a tricky subject; on one hand, a video might be unlisted, rather than privated, to make it easier to share with friends. On the other hand, many YouTubers also unlist videos such as outtakes, early revisions of videos, stream archives or off-topic content that might not fit their channel’s niche. Such videos are linked to, in video descriptions, pinned comments or Tweets. So while some unlisted videos aren’t meant for everyone’s eyes, other unlisted videos are only hidden from the channel interface and search results. Yet an exploit is an exploit, and URL predictability could be a serious problem for certain videos.

Some action certainly had to be taken here, so in 2017 the video ID formula was changed into something less predictable, that was definitely a step in the right direction. What is happening today, 4 and a half years after, is a security update to set a sizable number of unlisted videos uploaded prior to that date to private. Thus, several million videos have suddenly been virtually deleted, as they are no longer accessible to anyone but the channel owner. While this decision will secure potentially private content for many channels, it is also a great loss for inactive channels who unlisted videos liberally and were not able to opt-out of the decision.

On the bright side, channels which are still active can set their videos to public at a later date. In fact, YouTube goes so far as to encourage these channels to re-upload their own videos to be able to take advantage of the new URL system. Except, it’s just not the original uploaders and video makers who are reuploading. Reuploads from other users who had previously downloaded unlisted videos are starting to also pop up, the same as it ever was.

With tools like youtube-dl or Reddit’s SaveVideo, the YouTube community is pulling together to salvage whatever they can from old unlisted videos. And they are getting only better, Archive Team’s unlisted video project hit over 200TB of data. As videos die off, here are some folks desperately trying to revive them, trying to uphold what one might call their online heritage.

A few days ago one of the oldest videos on YouTube was made public from unlisted. It was originally uploaded on April 29, 2005. Titled Premature Baldness, it too is a reupload and final memento from a chasebrown.com which is no longer recognizable. A whisper to remind us that while invoking the right to be forgotten we ought not to neglect, on the other hand, a right to be remembered…