Pursuant to the Notice of Inquiry ^[1] published January 21, 2004 regarding the Request for Comments on Deployment of Internet Protocol, Version 6, the Electronic Privacy Information Center submits the following comments urging the Department of Commerce to shape IPv6 policy to promote security, privacy, and stability in emerging communication services.

The predecessor to IPv6, IPv4, formed the foundation for the Internet as we know it today. ^[2]   However, weaknesses in security have allowed identity theft, third party surveillance, online fraud, and hacking, to become significant threats to Internet users.  As the reach of the Internet extends with the capabilities of IPv6, further growth of the online community requires strong safeguards for the privacy and safety of persons online.  Such privacy protection can safeguard people online by reducing these threats, and shield people from online exploitation. 

As the replacement network protocol to IPv4, IPv6 provides an important communications framework for the next generation of network applications.  IPv6 will extend the reach of the Internet into new areas and uses in the daily life of consumers; IPv6 has the flexibility to support mobile personal devices in wireless environments to high performance networking devices. ^[3]   In light of the future growth of IPv6 networks, it is absolutely vital that IPv6 incorporate strong privacy protections for end users; these protections will lay a foundation of privacy and security services for use by end user applications. 

Commitment to Privacy from the IPv6 Community

There is already a long-standing commitment within the IPv6 community to promote security and privacy. Historically, the Internet Engineering Task Force (“IETF”) has attempted to increase the reliability, security, and privacy of computer networks.  The Internet Advisory Board and Internet Engineering Steering Group Statement on Cryptographic Technology and the Internet called for the availability and development of stronger tools to protect security and privacy of network users and rejected limitations on computer security based on country requirements for interception. ^[4]   From early in the IPv6 standard development process, the IETF has required support for Internet Protocol Security Architecture (“IPsec”), which provides services such as security, integrity, and confidentiality. ^[5]    Further, as threats to privacy have been identified, IETF has taken steps to address the privacy vulnerabilities through technical privacy protection. ^[6]   The European Commission IPv6 Task Force to the Data Protection Working Group has recognized IPv6 as a “potentially powerful tool to improve the possibilities of user privacy.”

A key feature of IPv6 is IPsec, which provides security, integrity, and confidentiality services at the network, and further includes other features to facilitate the practical, efficient deployment of security technology. ^[7] While IPsec may be used over an IPv4 network, the IPv6 standard requires IPsec capability. ^[8]   IPsec features protect the data flowing over an IPv6 network from interception and surveillance. ^[9]   Because IPsec provides security at the network layer, end user applications are able to, and should, augment the services of IPsec with their own security to ensure robust privacy protection.  These IPsec privacy features, if properly used and complimented by security architecture in end user programs, offer significant advantages over non-IPsec implementations of IPv4. ^[10]   The DOC should strongly encourage the use of the IPsec features by end programs, and ensure that government networks and applications fully utilize the features of IPsec.

The importance of privacy to the IPv6 community is seen through the affirmative actions by the community to eliminate threats to privacy. As an example, early IPv6 implementations used an addressing scheme that threatened user privacy and online anonymity by tying a user’s IPv6 address to the embedded network hardware access address. ^[11] This mechanism would have the effect of creating an unchangeable, unique identifier that could be used to correlate “seemingly unrelated activity” and allow a system and user to be traced across multiple unrelated networks. ^[12]     This behavior is very much like that of an online “cookie,” except while a “cookie” tracks usage on a web site and may be erased, the original IPv6 addressing scheme would have allowed the tracking of all online activity (e.g., email, instant messaging, video conferencing, in addition to web traffic) through an unchangeable identifying number. ^[13]

To address this privacy and security threat, the IETF developed RFC 3041, “Privacy Extensions for Stateless Autoconfiguration in IPv6.” ^[14]   This aspect of the IPv6 standard increases end user privacy by enabling users to periodically randomize their IPv6 address as well as generate temporary addresses, thus preventing the creation of a unique, unchangeable IPv6 address assigned to a specific person. ^[15]

Further, this threat to online privacy also created a threat to network security. The early static addressing scheme that created unchangeable, unique IPv6 addresses could allow malicious users to map the “topography” of IPv6 networks, and locate key infrastructure, such as underlying subnet structures and mapping between networks, to focus their attacks. ^[16]   The feature created to protect end user privacy in this situation also protects network security from malicious attack.  Randomized addressing increases network security by allowing IPv6 systems to “hide” from attacks and threats.  For example, the White House changed IP addresses of www.whitehouse.gov to dodge the “Code Red” denial of service attacks. ^[17]   Thus, strong privacy protections also serve as important security safeguards, which help ensure the safety and stability of the Internet in general.

Recommendations for IPv6 Policy by the DOC