Jason Harrow
This is the first post in a multi-part series. New posts will appear every Friday. Feedback is welcome to jason.harrow@gmail.com.
Moore’s Law describes how quickly computer technology evolves, and I do not know how many lawyers, policymakers, and law students are familiar with it. I would guess that most are not. This is probably because the rate of change of computer chips seems obscure, irrelevant to law and policy, and, anyway, a little bit made-up. How can it possibly be that the speed with which people can improve computer technology — that is, how much faster your laptop is now than it was five years ago — can obey anything like a “law?” And even if it does, how can that possibly affect how we think about energy policy?
This series of posts is an attempt to answer both of those questions. This first post is an introduction to Moore’s Law and other related laws of technological change. My description will be at a high-level of generality. That’s because I’m not interested in the details of integrated circuits, and, even if I were, the details don’t matter. What matters will be this: it turns out that some technologies change really, really quickly; have behaved like this for a half-century or for more; and don’t appear to be slowing down. In fact, this set of technologies improves faster than any other technologies ever have in the history of humanity. Sadly, though, not all technologies change so quickly — and green technologies are among this slower set. In subsequent posts in this series, I will explore why policymakers need to realize this when they think about energy policy, and why President Obama does not appear to have sufficiently grappled with this reality in setting his new “clean energy” goal.
1. What is Moore’s Law?
Moore’s Law is named after Gordon Moore, the co-founder of Intel, the world’s largest computer chip manufacturer. The Law states that “the number of transistors incorporated in a chip will approximately double every 24 months.” I recognize that this may be meaningless to some readers, but the key implication is that “computing chips shrink by half in size and cost” roughly every two years. Because we’ve now been living in the “Information Age” for a few decades, we’re getting used to upgrading to amazing new devices every two years, but if you take a step back, this doubling-ever-two-years is simply astounding.
Other aspects of the information technology world exhibit similarly accelerating growth, and some trends even have their own names. Kryder’s Law, for instance, says that hard drive capacity works roughly the same as computer chips, so that every few years hard drives will get substantially better and cheaper. According to this post by technologist Kevin Kelly — whose recent book What Technology Wants inspired this series — and a nice Wikipedia page on Moore’s Law, things work the same way when it comes to the pixels in your digital camera or the ability to transmit information over a fiber optic network. These things don’t just get better. They get a lot better, really quickly. Counterintuitively, when they do, they also get cheaper. This is what has been driving the miraculous Information Age that we are currently living through.
To understand how dramatic the change has been in this cluster of technologies, consider a few illustrative examples. In 2000, one gigabyte of hard drive space cost $44.56. In 2010, it cost $.07. In 2000, it cost $193 per gigabyte to stream video. In 2010, it cost $.03. (Both of those comparisons are from this Wiredinfographic.) In 2000, Apple’s best laptop had a 400 MHz chip, 64 MB of RAM, and a 6 GB harddrive. It cost over $3,100 in today’s dollars. Today’s high-end Apple laptop has a 2.53 GHz processer, 4 GB of RAM, and a 500 GB harddrive. It is exponentially better than the version of a decade ago, yet it costs $1,000 less — $2,199. Today, the miniscule and inexpensive iPhone is easily more powerful than the fastest laptops of 2000. Because things change so quickly, so many technologies that are now critical to our lives, from Facebook to tablet computers to YouTube, did not even exist a decade ago.
Change has proceeded at this pace in these industries for the last half-century or so, and it shows no signs of slowing. There are rumblings that we may be reaching some theoretical physical limitations — just how tiny can you make a computer chip, after all? — but engineers have blown through all predicted limitations so far. There is no particular reason to think they will stop now, and even if they do slow down a little, and improvements double only every three years, change still accelerates exponentially.
This is exciting on its own, but it’s also important for subsequent posts to realize how seductive this kind of growth is. It is tempting to think that everything “high-tech” gets better and cheaper on a two-year scale. But, sadly, not everything follows this curve — in fact, Moore’s Law is rare and amazing, and not at all the norm. For instance, I suppose that cleaning supplies have gotten better over my lifetime, but I cannot clean my kitchen twice as quickly as I could two years ago due to amazing new cleaning products. The same is true of screws, tube socks, ceiling fans, and legal argumentation; none of those is twice as good as it was in 2009. Airplanes are not twice as fast as they were two years ago, polar fleece is not twice as warm, and frozen dinners are not twice as delicious.
In my next post, I’ll show that one other sector that doesn’t obey Moore’s Law is energy technology, like battery capacity and solar panel efficiency. Sadly, and perhaps disastrously, these technologies are not among the special few that show this “off-the-charts” growth.
With that as groundwork, my goal in the next coming posts will be to show that policymakers need to understand that, although they cannot predict the future, they can estimate with some accuracy how quickly technologies will improve. When we look carefully at the curve for renewable energy technologies, it is clear that it will be nearly impossible to get to the President’s goal of producing 80 percent “clean” energy by 2035 without some seriously fuzzy accounting (and, as I’ll show, this fuzzy accounting has already begun). It’s a sad but true reality that the technology curve is just not steep enough, and no amount of wishing or invoking “American innovation” will make it so. At the end, I’ll try to outline what an energy strategy that does not rely on wishful thinking about technology’s ability to easily solve the problem of climate change might look like.