Buy Lines: When it's time to change, you'll be glad for IPv6

Internet Protocol version 6 will be worth the pain. Change without warrant is for naught. But in technology, change is not only better, it's inevitable.

Internet Protocol version 6 will be worth the pain. Change without warrant is for naught. But in technology, change is not only better, it's inevitable.

We gained when 8-bit computing gave way to 16-bit, then to 32-bit and now to 64-bit. During this same time, the Internet became ubiquitous without a significant protocol upgrade: We're still using the same 32-bit addressing scheme used when many were not sure that the Internet would be the networking architecture of the future.

But all that is on the cusp of changing as our government, among others such as China, the European Union, Japan and Korea, seek to require the purchase of next-generation IT that can implement a new 128-bit addressing scheme.

IPv6, as defined in the late 1990s, is a much cleaner protocol, promising lower cost administration, higher quality of service and built-in security. Why didn't IPv6 just naturally take off? Because the problem with moving from IPv4 to IPv6 is that until everyone on your network is IPv6-capable, the benefits cannot be fully realized.

This means agencies and companies will be buying IPv6-capable equipment and running dual systems during a transition with only marginal benefit. That's why it takes some initiative from the federal government to get the IPv6 ball rolling.

The government began to take IPv6 seriously when, in February 2003, it was identified as a key component of the National Strategy to Secure Cyberspace. The Defense Department announced in October 2003 that the Global Information Grid would be based on IPv6, stating the protocol is absolutely necessary to fulfill its vision of network-centric enterprise.

The General Accountability Office in May 2005 published an excellent report: "Internet Protocol Version 6: Federal Agencies Need to Plan for Transition and Manage Security Risks," which highlighted the long-term benefits, costs and implementation strategies, along with the major concern that during transition, security will be even more difficult.

The Office of Management and Budget began implementing the GAO report's recommendations with an Aug. 2, 2005, memo to federal CIOs, laying out transition milestones and deadlines. To date, major agencies have complied with the milestones requiring complete inventories of IP devices, as well as an impact analysis of fiscal and operational effects and risks.

The latest move in the transition is a proposed rule to the Federal Acquisition Regulation, released Aug. 24, that requires federal agencies to include IPv6 products in IT procurements to the maximum extent practicable. Comments to the proposed rule are due by Oct. 23.

When I first saw the rule, I imagined that industry would welcome it, but I have since learned that some are expressing concern that the government's timetable for IPv6 transition is too aggressive. I disagree.

Complete conversion by June 30, 2008, seems lofty, but the goal only requires IPv6 at the backbone layer. The reality is that we'll be converting applications to IPv6 for many years to come ? some say a decade ? before we unplug IPv4.

IPv6 is the framework for the next generation of IT innovation. As a nation, we need to lead in this area. For the sake of those commercial companies that have invested to provide IPv6-capable hardware, software and transition tools in response to the government's 2003 promise, I hope the government sticks to its timetable so we can all get on with it.

Steve Charles is a co-founder of immixGroup, a government business-consulting firm in McLean, Va. He welcomes your comments at Steve_Charles@immixgroup.com. immixGroup ranks No. 65 on the Washington Technology 2006 Top 100.

NEXT STORY: Complexity breeds opportunity