Waxy.org is the sandbox of Andy Baio, an
independent journalist and programmer living in
Portland, Oregon. I created After weeks of criticism from YouTube commenters, the creator of the popular fan-made "Music Is My Hot Hot Sex" video finally pulled it offline. (It's still mirrored here.)
On March 7, YouTube administrators removed it from its #1 spot on the rankings while they investigated it. Apparently, no foul play was detected and it was reinstated. Stephen Hutcheon from the Sydney Morning Herald has more on the story, including a screenshot from the leaderboard on the day it was removed.
It's hard to get a sense of the scale, which roused suspicions in the first place. To put it in perspective, in the seven days from March 7-13, the CSS video gained 17 million new views. That's more views than this week's top 20 videos received, combined.
In one week, the CSS video got nearly as many views as the insanely huge Crank That (Souljah Boy) received in 7 months. It was six times as popular as Mariah Carey's new video, in half the time. More popular in a week than the all-time views for Amy Winehouse's "Rehab," Tay Zonday's "Chocolate Rain" or Chris Crocker's "Leave Britney Alone."
Assuming YouTube's numbers are accurate, what was the mystery source of traffic? Now that the video is gone, I don't think we'll ever know. There have been a number of theories, but none of them really pan out.
- Popular search terms like "hot sex" and "obama." Unlikely, since the video never ranked well with those queries. Searching for "sex" or "hot sex" didn't return the video anywhere in the top 100 results.
- Social network embeds. It's still possible that there's a single source of traffic from an embedded video on an extraordinarily popular website on autoplay. If so, it's managed to evade YouTube's referral tracking, while still getting counted in views.
- Leaderboard traffic. Once in the top 10, could traffic have snowballed from people clicking from the all-time most viewed page? No, since the video gained an additional 25 million views in the week it was removed from the leaderboard. Also, other videos in the top 5 only saw a small fraction of the growth.
- Chinese users. Someone noted that Chinese users watch YouTube, but won't (or can't) sign in to rate/review/comment. Could they be coming from China?
- Buzz from the iPod Touch ad. That might make sense in the days following the original commercial's release last October, but the video's growth was highest in the last two months.
Rajeev Kadam from Divinity Metrics, a company that provides video metrics for media companies, got in touch with me and provided these historical stats for the CSS video for the last five weeks. Here's a chart of that data, or you can see the spreadsheet.
Philip Rogosky asked Clarus Bartel why he removed the video. Clarus reminded Philip that he'd contemplated deleting it before, but his friends advised him to wait to see if it would reappear on the leaderboard, clearing his reputation. At that point, he decided to delete it only because of the critical comments he was receiving on his other videos.
Asked how he felt when he pressed "delete," Bartel responded, "Sad but relieved! If only I'd earned a buck or two or a job offer, I'd feel different today."
11:23 AM
Shame about the negative comments. It’s funny how riled up people can get about apparent video views on a web site. It’s not like the guy was making money, or really even diverting attention from other people.
11:57 AM
Wouldn't driving this type of traffic be possible if you had control over a largish botnet? Maybe there's some dude in Ukraine with too much time on his hands that tried a little experiment on a "random" video, and it just happened to be this one. Then just kept going to see how far he could take it?
12:20 PM
Still highly mysterious. It's as hard to believe this guy cheated YouTube in a way spammers have never managed as it is to believe it legitimately got that many views.
Presumably YouTube have stats on these visitors - their country, browser versions, etc, even if they're failing to gather referral information. And if they've looked at the stats and found them legitimate, it's hard to see how they could be wrong.
I'd love for someone to rise to the visualisation challenge this presents. Your stats there are enlightening, but I'm sure there's someone out there who could just show me a bunch of colourful boxes or dots that would make me go "Wow, okay, that's bullshit."
3:22 PM
I totally agree. Just to be clear, I don't think that there's been cheating and almost certainly not from Clarus Bartel himself. I'm just terribly curious where that kind of traffic windfall could come from without generating the corresponding social activity.
9:27 AM
Makes me wonder how legit all the buzz is about YouTube views for Obama's speech a couple nights ago. NPR carried a segment this morning about it. It may be harmless with a mediocre fan vid, but is tantamount to media manipulation when it occurs in the context of political memes or messages. Artificial virality is the new yellow journalism.
1:07 AM
shame the owners taken the vid down. Some people take comments too seriously. He should not have editted the comments on the vid. He should have just ignored everything... done all the letterman interviews and laughed his way to the bank. What ever he did, you have to respect what he achieved
7:10 AM
Shame about the negative comments. It’s funny how riled up people can get about apparent video views on a web site. It’s not like the guy was making money, or really even diverting attention from other people.
8:42 PM
where is it