This is, of course, entirely what was hoped for; the world continued to browse as normal. The Internet Society, which organised the world test, considered it a success: “The vast majority of users were able to access services as usual, but in rare cases, users experienced impaired access to participating websites during the trial.” Google recorded that IPv6 traffic reached 0.3%, which is a significant level given that it had barely registered the day before.
The alternative – internet users unable to access internet pages and significantly slowed browsers – would have been a disaster, so it’s encouraging to have such proof that the IPv6 technology works.
But should we be celebrating? The figures involved have been banded about endlessly, so most of us will know that only around 80 million of the 4.3 billion IP address available under IPv4 remain unallocated. As with all figures we see often, it’s easy to nod and move on, but there are only approximately 1.9% of the total IPv4 addresses remaining. Time, surely, for some speedy action.
My first concern is the number of companies that took part in the trial. Many of those that did were major players – Cable&Wireless Worldwide, Telefonica, Level 3, NTT America, Google, Facebook, Yahoo and Akamai, among others. But from a list which included major internet service providers, businesses, educational institutions and government agencies, there were fewer than 400 participants.
Whether small and medium-sized businesses have started to address these issues at all is anybody’s guess. But I for one would guess not.
The other concern is the speed with which most participants reverted back to business as usual. Akamai’s tracking shows how quickly IPv6 was switched off at the end of the 24-hour trial. If there was a hope that the companies testing IPv6 would leave it switched on, running their sites in the “dual-stack mode”, this was clearly misplaced.
There are, of course, a number of early adopters. The Internet Society pages show an interesting case study of the German news site heise online, which tested IPv6 in September 2010 and has since continued to dual run. NTT America, Telefonica and Cable&Wireless Worldwide have all been rolling out IPv6 for a number of years. But many other companies clearly switched off IPv6 functionality as quickly as they switched it on. Given that what this technology needs more than anything now is some momentum, I think that this is a shame.
The Internet Society emphasised that the aim of the trail was to “educate businesses” about IPv6. I’m concerned that given the timelines involved, businesses should already know about the issues. Worse, its smooth running during testing is a double-edged sword; those that didn’t know about it before the trial probably still don’t.
The 24-hour collapse of the internet on 8 June 2011 would have been an unmitigated disaster, but at least it would have raised the profile of the problem. Now that we know it works, it’s time to spread the word and to start taking some longer-lasting action.
