Posts tagged ‘Compatibility’

IP Address Shortage - A Looming Crisis?

I have spent the better part of the last week preparing for my fall Internet Economics class - a class that applies economic principles to online selling. It is my intention to use this blog to present the laymen’s terms of what we’ll discuss in class.

What better topic on which to begin than the rising “crisis” of the impending shortage of IP addresses? What does it mean and how will it affect selling online? The analysis is a bit like a combination of the oil crisis and Y2K.

First, what is an IP address and why is it important to you? IP stands for Internet Protocol. The oft used analogy is that of a phone number. Every computer connected to the internet has one, and this allows computers to talk specifically to each other. If you want to find out yours, try clicking on http://whatismyipaddress.com/. Sort of surprising, isn’t it? You have an IP address and the map on the site very possibly points near to where you live if you have not taken steps to hide it. The wireless router in my home also assigns dynamic IP addresses (DHCP) to each computer connected to it within a small range of addresses. Nothing drives my kids more crazy than when the router goofs and assigns two computers in the home the same IP address. The latter computer responds with “IP conflict” and will not access the internet. This can and does happen when Internet Service Providers (ISPs) make the same mistakes with customers. Perhaps with the growing shortage of IP addresses, this may occur more often? (In economics, we call that an “empirical question”.) Even more interesting is that the IP address you just looked up is really just a “dotted decimal number” and not your computer’s address in binary form - which is a series of 0s and 1s. These 0s and 1s are represented in groups of four with eight binary numbers in each (called octets). (These are called IPv4 addresses.) These octets can be arranged in just shy of 4.3 billion combinations. (That is NOT an empirical question. It’s a fact. And we have used nearly 85% as of May 2008.*)

So what does this mean? Well, it’s a simple case of supply and demand. It has been known since the beginning that we could potentially run out of IP addresses. Remember, we started with the knowledge that we could create only 4.3 billion. Starting to sound a bit like Y2K? In that case, programmers knew for years that using two digit dates (e.g., 95, 96, 97) would cause a problem when the year 2000 (i.e., 00) rolled around. So we started with a finite supply of IP addresses, but the supply seemed so vast that there seems to have been little concern. Now, as the assignment of IP addresses is growing exponentially as developing countries like China and India come online and expanding technology in developed countries requires more IP addresses, there is a looming shortage. The OECD issued a report about this in May http://www.oecd.org/dataoecd/7/1/40605942.pdf stirring up concern that finally hit the major news media.

The supply/cost factors in this scenario seem to be upgrading to IPv6 compatibility, obtaining and distributing remaining IPv4 addresses, and/or better managing existing IPv4 addresses. The demand factors in this scenario are the growing demand for Internet access by developing countries as they come online and technological innovation in developed countries. (For example, just imagine that people are now clamoring to have internet access via their cell phones. When I first started teaching, none of my students had cell phones. Now I can’t find any that don’t. Soon, I won’t be able to find any that don’t have internet access via their cell phones. It may be as common as microwaves.)

Developers have known that there are ways to get more IP addresses. Conversion to an IPv6 system of allocating internet addresses (having six octets instead of four) is the solution that would provide trillions more addresses. But it is costly and until devices were compatible, demand was low. Now, many of the newest systems and devices are already IPv6 compatible; however, IPv6 devices are not backwards compatible to IPv4-only devices. And imagine the cost of reassigning billions of IP addresses… Without getting over my head in how these things work, the simple economic notion is that eventually everyone may have to upgrade to the new system. Sort of like how your analog tv will quit working next year (http://www.fcc.gov/cgb/consumerfacts/digitaltv.html). And then there are networks effects. Few want to make the investment until they see others getting on board as well. (Read the Cellular-News article below to see how Japan and China are innovating in this area.)

How is this like drilling for oil? Well, it is less profitable to drill for oil in some places until the price is high enough. We have been quite spoiled to have relatively low cost access to the internet and perhaps our demand elasticity (our quantity response to the change in price) will be similar to our demand for gas as we grow more dependent. When the market price of internet access rises, as could be the case when supply decreases due to higher costs and demand increases due to more desire for internet access, some folks will reduce their access, but how many? (Another empirical question.) If the answer is not many, we can expect the price to rise quite a bit.

And then there is supply elasticity. Some reporters (as reported on Fox News on July 7, 2008 titled “The End is Near: Less Than Three Years Until the End of the Net?”) would have us believe that the supply of IP addresses will become perfectly inelastic - i.e., there will be no more and the Internet will just fail. If there were no solution, then there would be no new users at some point and the cost of purchasing access will grow as those who have it are lured into giving it up by the higher price they can get to sell it. Some fear that companies will monopolize the industry by buying up all the IP addresses and selling them at a high price. This does not make sense to me when there is an alternative that could ultimately flood the market with trillions of IP addresses. There are also methods of sharing IP addresses, much like the wireless router in my house, that can, and are, being applied in larger measures to IPv4 addresses. (This is called Network Address Translation.) But again, if this were the case that the OECD fears, should the government regulate the distribution of the remaining IP addresses? Or should the market handle it? Should we “drill” where we currently know to find IPv4 addresses or explore to find a better, more efficient source? Or would drilling be a good short term solution while working on the longer term solution? Whew, the analogies could go on, and I don’t profess to have the answers.

I am going to recommend that my students read the OECD report for discussion in class. And while this blog is still incredibly new, any comments or suggestions posted here would be great. I don’t claim to be a computer expert, so any corrections to technical errors in this post would be appreciated.

Ok, so how many economic terms have we covered here that we will eventually discuss in class? Supply, demand, elasticity, market price, compatibility, monopolize, network effects, government regulation.

Other Sources:

“What Is An IP Address?”, How Stuff Works, http://computer.howstuffworks.com/question549.htm
*”Mobile Phones Could Be Affected by IP Address Shortage”, Cellular-News, http://www.cellular-news.com/story/31184.php, Posted to the site on 15th May 2008
“Internet Address Space: Economic Considerations in the Management of IPv4 and in the Deployment of IPv6, OECD, http://www.oecd.org/dataoecd/7/1/40605942.pdf, May 15, 2008
“IP address shortage to limit Internet access” USA Today, Posted 8/3/2007 12:37 PM, http://www.usatoday.com/tech/webguide/internetlife/2007-08-03-ip-shortage_N.htm