The following text is copyright 2007 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

Do we need a replacement Internet?

By: Scott Bradner

On Friday April 13th the Associated Press published a story with the provocative title "Researchers Explore Scrapping Internet." The article starts out confusing a research approach with a potential result but does come back to reality by the end. But even if it had not, this would have not been the first time that replacing the Internet was seen as the logical thing to do, at least by people wearing the blinders of true belief or challenged logical thinking.

The AP article (http://biz.yahoo.com/ap/070413/rebuilding_the_internet.html?.v=11) focused on the U.S. National Science Foundation Global Environment for Network Innovations (aka GENI) research effort. (http://www.geni.net/) GENI is talking a "clean slate" approach to networking research. A clean slate approach is where one thinks through solutions to problems without being constrained by what exists today. This is a great way to do research but it's often not all that great a way to do product development. In order to be successful, new products generally have to provide a benefit over the cost of purchasing and installing the product.

Today's Internet has a lot of problems, ranging from predictability to security, to the need for a rational business model, that would be good to fix. But there is a lot of Internet out there and working well enough to be very useful. There would have to be a very good reason to scrap and replace it and it does not seem that the development spawned by GENI are likely to be enough to punt the Internet in any large scale way although many of the developments may provide enough benefit to cause an incremental replacement of some key Internet functions.

About the only thing that differentiated this report was who was claiming that the Internet was going to be replaced. In this case it was a reporter trying to get a story published in most previous cases it has been someone tied to a phone company or a phone company supplier.

The first time I heard something like this was in the pre-web Internet in the mid 1990's when some of the phone companies were predicting that ISDN would replace all the silliness of ISPs and enterprise networks (see http://www.sobco.com/nww/1995/13-did.not.do.that.html for an example). About the same time others in the phone business were among the many folks, including much of the press, riding the ATM bandwagon. I recall many times when I was belittled for doubting the vision of ATM as the technology that was going to replace all existing technology. (see, for example, "My last ATM column?" http://www.sobco.com/nww/2001/bradner-2001-07-23.html)

A few years later the song and some of the telco singers were the same but the technology was then the third generation (3G) cellular wireless that was also going to replace enterprise networks and ISPs. Lately some of the telco people have been saying the same about the ITU-T's Next generation network project (http://www.itu.int/ITU-T/ngn/index.phtml) at least for replacing ISPs.

As the folks from the telephone companies realize all too well since it's happening to their own sometimes infrastructures do get replaced but it takes a long time and the new technology has to provide useful new services. At this point the Internet of old (i.e. the one we still have today) works too often to drive quick replacement and none of the new applications I've seen touted for a new Internet cannot be retrofitted to the one we already have.

So, beware salesmen or phone company people selling replacement instead of improvement - keep a good grip on your wallet.

disclaimer: The job of Harvard's development people is to loosen wallet grips but I did not ask them or anyone else at the university their opinion on replacing the 'Net so the above non-belief is mine alone.