Wheels Down on YouTube

15 years ago this week, I posted my first video on YouTube. It was this 60-second video of landing gear seen through an airplane window as my flight from Aspen to Phoenix came in for a landing:

I shot it with a Canon s400, a point-and-shoot camera that shot video clips up to three minutes long at 320x240 pixels and 15 frames per second. It doesn’t sound like much, but I loved the limitations of this camera. The low frame rate and noisy image made everything feel slightly other-worldly.

In this case, the low frame rate and slow shutter speed resulted in a beautiful smeared effect as the ground gets closer and closer to the wheel in the last few seconds before touchdown.

Commenters on the video identified not only the model airplane I was in, but also what airport I was landing at. Pretty impressive.

Other Minutes

I carried that point-and-shoot camera around with me and captured a lot of little minute-long slices of life for a series I called “60 Seconds in the Life of…” which I posted on my blog. Each is a little moment of zen.

Here are some short excerpts as gifs:

Steam:

Laundry:

An Escalator:

A Bear and Sea Captain:

A Taxi Meter:

A Weeble:

A Subway Window:

A Puddle:

I have so many more of these, but I don’t want to bog this email down too much with heavy gifs (too late).

Okay, just a few more.

Windows:

A Wild Tail Pet Toy:

Apple Store Stairs:

Landing Gear Stans

At some point I noticed that the landing gear video had amassed hundreds of thousands of views. As I write this, there are 476,781 views. How did that happen? I decided to look at the data and see what I could find out.

According to the analytics, 75% of the views come from YouTube’s algorithm recommending the video to people who are watching other videos about landing gear. And there are a ton of videos on YouTube about landing gear. In fact, the #1 search term that led people to my video was, you guessed it, “landing gear".

This graph shows views over the past 15 years:

Looks like the video had steady growth for years, perhaps because of YouTube’s growth. And then all of a sudden it drops off in 2011, gasps for air again around 2013, and looks dead for the next few years. What happened? I assume there was some change in YouTube’s algorithm and it stopped getting recommended. The content viewers expect from YouTube also changed.

Then in 2017, there’s a resurgence! Those views are entirely because YouTube’s algorithm decided that my landing gear video should be recommended after people watch just one video called Fictionalization DO NOT ATTEMPT :: Airplane is being stopped by a Nissan Frontier !!! that was already six years old at the time.

After that, the views more or less flatlined.

Algorithms are weird, man.

Here’s another interesting graph. This one shows viewer duration. Normally you would see a steady decline as people stop watching, or at best a flat line if people watch the whole thing. But here we have a different shape:

It shows that a lot of people skipped to the end to watch the landing. I guess there are two kinds of people in this world: the kind that enjoys the anticipation as the ground gets closer and closer, and people who just want to skip to the end and see what happens.

Which kind are you?

Before YouTube Was Obvious

I uploaded that landing gear to YouTube when the platform was just about a year old. But it wasn’t the first or only video hosting option, and it wasn’t clear back then that it would soon become the go-to place to post video. So when I posted these 60 second videos on my blog, I actually did not use YouTube as my video host.

Instead, I used a site called Revver that offered something YouTube did not: revenue. (Oh, hey, I get it now!). Their video player ran post-roll ads, a precursor to the YouTube ad model that everyone’s familiar with today.

I found the video quality with Revver was actually better than YouTube at the time, which compressed video too much for my taste. But the Revver player was very buggy and sometimes didn’t even load, and YouTube got better. After a while, I switched over to YouTube.

Revver shut down in 2011. In my year or so using Revver, I made $378.49.

I don’t know what I spent that $378.49 on, but thank you for spending your time reading this weeks newsletter. (Obviously I did not spend the money on good segues.)

See you next week!

David