Mega-minute of Data — MATH VALUES

Written by Tim Chartier, Davidson College

Illustrated by Ansley Earle

Data analytics is a hot topic in our world and rightly so. With so much data, we need mathematics to analyze and make meaning from the mass of data. As such, we must rely on algorithms to search the web or recommend products or movies among the multitude of options. Mathematics lies at the foundation of such algorithms. It can also give us a sense of how much data there is and why such amounts of data are produced.

First, do we really need analytics and mathematical methods to help process data? The answer is yes. In fact, even one minute of activity underscores the need for mathematics and why the foundational methods we learn in a classroom are relevant far beyond a course. Before we start, let’s get a handle on the unit of time we will discuss. Think about what you have done in the last minute. Probably not a whole lot. It was just a minute. In the digital world, an astounding amount happens in 60 seconds.

In one minute, Amazon users spend an average of $283,000. That's over a million dollars every 4 minutes and over 400,000,000 dollars every day. A dollar bill is 0.0043 inches thick. The Empire State Building is 1,454 feet from the sidewalk to its tip. So, if you piled the number of dollar bills spent in an hour on Amazon, you'd have 283,000 x 60 x 0.0043 inches or just over 4 Empire State Buildings piled one on top of the other. If you pile the number of dollar bills spent in a day, it would reach 27 miles into the sky. The recommendations you see on Amazon are customized for you from an incredible number of purchases made today and over time.

Every minute of every day, about 450,000 hours of Netflix video content are streamed. So, just how long would it take for one person to watch that much content back-to-back? You would need to sit in front of a screen exclusively watching Netflix content for 450,000/(24 x 365) or 51 years. That would lead to some serious eye strain! Looking only at YouTube, almost 700,000 hours of content are streamed every minute. That's almost 80 years. Just think about the last two minutes. More hours of content were streamed than you could watch in your entire lifetime even if someone had started streaming content at the moment of your birth! If you ramp up to an entire day, almost two centuries of content stream on YouTube. Individually, we may be viewing a gentle stream of video content. Collectively, it's more like a massive firehose of content every minute. Mathematical methods enable YouTube and Netflix to infer your likes and dislikes from this voracious velocity.

From data, Netflix knows what you watched, how much of a movie you watched, and your thumbs up or down, if you rate the movie. Amazon's data reflects your purchases and, in some cases, what web pages you visited before purchasing. Some companies look at Instagram posts to predict future style trends. On average 65,000 images are uploaded to Instagram per minute. That's 65,000 x 60 x 24 or almost 94 million per day. The entire Library of Congress has 16 million images, which includes photographs, historical prints, posters, cartoons, documentary drawings, fine prints, and architectural and engineering designs. Each day, the equivalent to 94/16 or almost 6 Library of Congress's image collections are uploaded to Instagram, although the proportion of selfies would be much higher on Instagram!

Now, let’s switch gears and see the central advances that have been made to process such data. Mathematics may help us gain meaning, but computer science, which is also fundamentally important in data science, also lies at the heart of understanding data made at the scale of our modern digital world.

The astounding amount of data produced comes from the number of people online and the amount of time they use it. For example, 4.48 billion people were using social media worldwide in 2021 which more than doubled from 2015. Few of us use all social media platforms but many of us use multiple. The average social media user engages in between 6 and 7 social media platforms. Among all those platforms, the average person spends 2 hours and 24 minutes daily on social media. If you started using social media at age 16 and lived to 70, that amount of time per day equates to 5.7 years of one's life.

How does that time distribute among social media platforms? People spend an average of 40 minutes a day on Facebook. Don't remember sitting that long in front of the computer or on your phone? You probably didn't in one sitting. People check Facebook a number of times a day. On average, people check Facebook 10 to 12 times a day. Twitter is only 17 minutes but people are checking only 3-4 times a day. Instagram is the same as Twitter accounting for 17 minutes a day. TikTok is close to Facebook with an average of 34 minutes a day. How about you? Do you have an account on Facebook and/or TikTok? How often do you check your account and how long do you spend on social media platforms?

So, how does this compare to other ways we spend time? On average, we spend 24 minutes a day waiting on other people. Maybe that's partially when people are checking their social media! Now, 24 minutes a day from age 16 to 70 equates to almost a year of your life. That's a lot of time waiting. You may be on social media for those 24 minutes. Remember, on average, a person spends another two hours a day on social media, which equates to over five years of one's life between age 16 to 70.

How long have you been reading this article? Just think of the magnitude of data produced. Then, the next time you see a recommendation or click a link recommended for a search engine, remember that mathematics played a fundamental role in making meaning from the mega-amounts of data we are producing every minute of every day.

Tim Chartier is the 2022-23 Distinguished Visiting Professor at the National Museum of Mathematics and the Joseph R. Morton Professor of Mathematics and Computer Science at Davidson College.