Well, if I called the wrong number, why did you answer the phone? -James Thurber
Question #91389 posted on 05/26/2018 12:48 p.m.
Q:

Dear 100 Hour Board,

So people talk about going saving the trees, going paperless and having tablet and computers and stuff instead of printing anything out. But computers and tablets require rare metals and fuel-burning electricity to operate. What has the most environmental impact?

-My Name Here

A:

Dear reader,

The flaw I immediately see in your question is that it assumes that paper replaces a computer. But I can't very well attend alumni week on a piece of paper, can I? Computers come in discrete units; I need to own the same amount of computer regardless of whether I use it for twenty minutes a week or twenty hours a week. From a plastic-and-metal standpoint, once I've bought my computer, it's best for me to use it for as many things as possible.

The question about energy is a good one, though: is it more energy-efficient to use a computer or a piece of paper for a task that could be done equally well on either medium? I was going to try to do some math (*le gasp*), but thankfully the internet has already done it for me!

Math number one: some random Stanford homework from 2010. It focuses strictly on energy consumption, comparing the cost of producing one sheet of paper to the cost of using a computer to read two pages of text. The conclusion? There's almost no difference in the amount of energy it takes to produce a piece of paper and the amount of energy it takes to read two pages on a desktop computer. Replace that desktop computer with a laptop, though, and you can reduce your energy consumption by 90%.

Yes, you're reading that right. Using a laptop is much, much more energy efficient than using a desktop computer. Why? In a word: batteries. The entire point of having a laptop is that it should be small, portable, and usable where electricity isn't available. That puts a premium on energy efficiency.

How about the difference in costs between online publishing and print publishing? A 2012 blog post tackles that question. While there's an energy footprint associated with server space - data centers use nearly 2% of the total energy consumption in the United States - it turns out that by the authors' math it is still 65 times more energy efficient to publish a scholarly journal online than to publish it in print.

And if you're looking specifically at your carbon footprint? Another blog post from 2009 compares the carbon cost of reading print newspapers to reading them online and finds that it all depends on the amount of time you spend reading. Given the post's assumptions, if you spend 10 minutes reading the newspaper, you're better off doing it online, but if you spend 30 minutes, you're better off getting the print version.

All of this information, of course, is six to nine years old. Six to nine years isn't a lot of time for the paper-producing industry, but it's an incredibly long time for the computer industry. As it turns out, according to a theory known as Koomey's Law, processor efficiency (measured in computations per kilowatt-hour) will double every year and a half until reaching a theoretical maximum efficiency around 2050. The law was proposed in 2011; assuming it's held true so far, processor efficiency has probably increased by about 16 times since the most recent article. Now, that doesn't cover other issues such as the efficiency of the lighting in computer screens, but it does imply that the numbers have shifted substantially in computers' favor over the past half-decade.

This is all resting on a few assumptions, of course. First, I'm assuming that you already have a computer (probably a laptop), and you're not buying a new device just to read or to take notes. Bringing a new device into the picture introduces all of the costs of mining, production, and disposal and substantially changes the efficiency balance. Second, I'm assuming that the added wear and tear on the device from using it to read something is negligible. Basically, I assume that your computer will last X number of years regardless of how much you use it, and you will probably retire it because it is outdated technology and not because it breaks. Change that assumption, and you change the equation. Finally, for each of these comparisons, I'm assuming you read the thing in question only once. If you're going to read something twice, that doubles the energy cost of reading it on a computer but leaves the energy cost of reading it on paper exactly the same. So the more you plan on reading something, the more efficient paper gets.

Overall, though, assuming you're reading or writing something just once or twice and assuming you're using a laptop that you would have owned regardless, computers are probably more energy-efficient than paper.

-yayfulness