Subject: Testimony submitted into the record-Oct 7 - house hearing on DNS PREFACE Ronda Hauben wrote: When I heard that the US House of Representatives Subcommittee on Basic Research was going to have a hearing on Oct. 7 I asked to submit testimony into the record and was told that wasn't possible. Then on Monday the staffer in charge of the hearing told me I could submit something into the record. That left only a day to work on something. Here's what I sent to the House Science Committee Subcommittee on Basic Research via email today and I also handed in a copy when I was there for the hearing. I have made one correction here with regard to what I submitted. The mission of the hearing was to talk about the proposals received by the NTIA on the DNS. Only one proposal was discussed and almost all the witnesses said they hadn't read any of the others. On the notice of the hearing all it mentioned were two proposals. And at the hearing the Committee people and witnesses talked about how to implement the Postel/Sims(?) proposal, with no concern that there were two other proposals submitted and that the Postel/Sims(?) proposal was created thru a closed door and silence everyone process. When the NTIA official was asked about what process there would be for the proposals she said that the NTIA would look at the comments and review the proposals. However the actions of the Committee and the witnesses was to act as if there was only one proposal that was already being implemented. The testimony I submitted follows. I welcome comments. Ronda ronda@panix.com TESTIMONY Testimony before the Subcommittee on Basic Research and Subcommittee on Technology of the Committee on Science on the subject of Internet Domain Names Rayburn House Office Building U.S. House of Representatives Washington, D.C. 20515 by Ronda Hauben researcher, writer, co-author Netizens: On the History and Impact of Usenet and the Internet Published 1997 by IEEE Computer Society Press October 7, 1998 INTRODUCTION I am pleased to be invited to submit testimony to the House Science Subcommittee on Basic Research and Subcommittee on Technology on the subject of whether the Domain Names System and related essential functions of the Internet should be transferred from U.S. Government oversight into a private sector corporate entity. My name is Ronda Hauben. I am co-author of the book Netizens: On the History and Impact of Usenet and the Internet published in May 1997 by the I.E.E.E. Computer Society Press. I am also an editor and writer for the Amateur Computerist newsletter which has covered the history and importance of the Internet since 1988. I have studied and taught computer programming and have participated online since 1988 and on Usenet since 1992. Also I submitted the proposal "The Internet an International Public Treasure" to Ira Magaziner and the U.S. Department of Commerce at the request of Mr. Magaziner based on the concerns I presented to him about the narrow phrasing of the question of the transfer of the Domain Name System to the private sector. I also responded to the Green Paper and submitted comments expressing concern that the general nature of the Internet and its history and traditions, and its nature as a communication medium were being lost sight of in the Framework for Electronic Commerce issued by Mr. Magaziner and his staff and in the Green Paper and subsequent White paper. And I attended the Geneva IFWP meeting in July 1998 and wrote up an account of what happened in an article "Report from the Front: Meeting in Geneva Rushes to Privatize the Internet DNS and Root Server System".(1) The proposal that I wrote and submitted to Mr. Magaziner on September 4, 1998, is now one of the three proposals that has been posted at the U.S. Department of Commerce web site by the NTIA with a request for comments. As you can see from my proposal I have found your hearing process valuable and have referred to testimony given by one of the witnesses in this matter in the Preface to my proposal. I want to commend the committee for both holding these hearings and for putting the testimony received on the committee's web site. I want to make a further recommendation, however. I want to recommend that you explore having an online discussion group. There the public could comment on the issues before the committee and on the testimony received or offer additional information or viewpoints into the public record so that you will have a broader set of information and viewpoints to influence your deliberations, especially when those deliberations concern the operation and future of the Internet. I hope that after you hear the rest of my comments you will understand better why this is so important. HISTORY OF INTERNET First, I would like to offer a bit of history of how the Internet came to be and I will endeavor to show how knowing this history will be helpful in determining how to evaluate the proposals before the NTIA. Then I will provide some recommendations toward the policy decision that this Committee and the NTIA are proposing to make. The Internet is a product of several significant and successful research projects that were conducted under funding from the Advance Projects Research Agency (DoD) in the 1960s and 1970s. One of the earliest of these projects is perhaps one of the most important in its relevance to the problem before this committee today. That project was the creation and support for interactive computing and time-sharing. In 1962-3, a computer scientist and engineering researcher, J. C. R. Licklider was invited to join ARPA and to begin the Information Processing Techniques Office (IPTO). At that time the common form of computing available was known as batch processing using large mainframe computers. Someone who wanted to run a program would bring a stack of punch cards to a computer center and return several hours later or the next day to retrieve the printout that the program generated to see if the program achieved the desired aim. Needless to say this was a cumbersome and frustrating means of using a computer. J.C.R. Licklider and the time-sharing projects that ARPA subsequently funded set out to change the form of computing and to make it possible for an individual to be able to type his or her own program into a computer and to achieve the results of the program immediately. This new type of computing that they created was called time-sharing. Relying on the speed of the computer, these computer pioneers were able to set up a series of different terminals for use by users who were all able to utilize the computer at the same time. As a result of time- sharing systems, multiple users were able to interact directly with a computer simultaneously. One of the projects funded by J.C.R. Licklider was called the Compatible Time-Sharing System (CTSS). It was part of the project funded at MIT by ARPA which was known as Project MAC. There were several important surprises that the pioneers of Project MAC reported from their research into time-sharing. 1) They didn't have to rely on professional programmers to do much of the needed programming for their time-sharing system. What they found was that the participants in the project would create programs and tools for their own use and then make them available to others using CTSS. 2) A community of users developed as a result of the ways that people contributed their work to be helpful to each other. 3) CTSS made it possible for users to customize the computing system to their own needs. Thus the general capabilities available provided a way for the individual user to create the diversity of computing applications or programs that this diverse community of users needed. As a result of this project, the researchers realized that once you could connect a remote terminal to a time-sharing system, you could develop a network with people spread out over large geographical distances. The networks that developed as a result of the research in time- sharing provided working prototypes and also a vision that would help to guide the next stage in the development of networking technology. The effort to improve the throughput of data across telephone lines led to ARPA supported research in packet switching and the funding of the ARPANET research to use packet switching to link up the computers that were part of ARPA's research program.(2) The next piece of history that is important to consider is the period during which the early Internet was formed. In 1981/1982 a mailing list was begun on the ARPANET. This mailing list was called the TCP/IP Digest and the moderator was Mike Muuss, a research computer scientist at the U.S. Army Ballistics Research Laboratory (BRL). The BRL during this period was one of the DARPA sites making the transition from an early ARPANET protocol, NCP to TCP/IP, which was to be the protocol suite that would make an Internet possible. By 1983 the cutover from NCP to TCP/IP had occurred and this made possible a particularly relevant event for the matters under consideration by this committee. That event was the separation of MILNET and the ARPANET into two independent networks to create an Internet. This split would allow MILNET to be devoted to the operational activities of the Department of Defense. And those on the ARPANET would be able to continue to pursue network research activities. Gateways between the two networks would provide inter-networking communication.(3) This gets us to a definition utilized in 1974 by Louis Pouzin, who had worked on CTSS at MIT and then returned to France to work on creating a packet switching network that was called Cyclades. Computer science researcher, Louis Pouzin, defined an internet as a network of independent networks. (He called "an aggregate of networks [which would] behave like a single logical network" a CATENET. DARPA adopted his concept as the goal of the research project it was supporting).(4) Each network could determine for itself what it would do internally, but each recognized the need to accept a minimum agreement so that it would be possible to connect with others who were part of the diverse networks that made up the Internet. RECOMMENDATIONS I have taken the time to review these two important developments in internetworking history because these two developments are at the foundation of the design of the current Internet as we know it today. These two developments highlight what is so special and particular about the Internet. The Internet that has grown up and developed is a continuation of the time-sharing interactive communities of users and computers where users contribute to and are in effect the architects of the network that they are part of. Also this understanding leads to another significant aspect. That is that this system of human- computer networking partnerships has a regenerative quality. New connections and programs, and databases or mailing lists are contributed by the users themselves. And thus the Internet grows and spreads and connects an increasingly larger number of computers and users around the world. The second important aspect is that the Internet architecture and design accommodates different needs and capabilities of a diverse set of users and user communities. For example, someone in Ghana with a 386 or 486 computer and a modem can be connected to and send email to someone in a research laboratory in Switzerland which has the most modern computer workstations. That is because the architecture of the Internet requires the least possible equipment and capability to be able to make Internet communication possible. Thus people and computers around the world who are using an extremely diverse set of equipment and computing capability are able to interact and communicate. I have taken the time to describe these general features of the Internet for a few reasons. The first reason is that this is what is so precious about the Internet and this is what I believe needs to be understood and protected when considering any change that may be contemplated in how the Internet is controlled, managed or operated. Any change in the minimal requirement that makes communication possible across the independent networks that make up the Internet can obsolete thousands of computers and many more users around the world and thereby jeopardize the connectivity and global communication that the Internet has achieved. Any change in the ability of users to represent themselves and to utilize the Internet for their diverse purposes and to contribute to what is available to others on the Internet, (as long as this does not put demands on others on the Internet), any such change can deprive millions of users of the Internet of the general form that makes it possible for the Internet to serve the communication needs of so many diverse communities of users. This diversity includes the computer scientists at MIT or the high school student in Sydney, Australia. If there are particular needs of any one group, such as the security needs of DARPA, or the ability to write with Japanese characters of users in Tokyo, the architectural design provides that within an individual network or several networks such needs can be accommodated, without imposing such requirements on the users of other networks. These two principles are important to study and understand because they represent what is being violated by the Framework for Electronic Commerce prepared by Ira Magaziner and his staff. This framework does not treat the Internet as a network of independent networks, but instead as a single network that must be changed to meet the needs of a particular set of users. Thus instead of recommending that an independent commercial network or a few commercial networks be created as part of the Internet to meet the special needs of commercial Internet users, Ira Magaziner's framework document requires that the entire Internet be changed to meet the particular needs of a particular set of users. This is a violation of the concept of an Internet. My recommendation is that the Framework that Mr. Magaziner has created needs to be recast to be a Framework for the Internet as a New Means of International Communication. Within that framework Mr. Magaziner can describe the particular needs of particular communities of users, but these particular needs cannot be allowed to replace the generality of the Internet design so that other users of other independent networks are being imposed on to satisfy the needs of any particular group of users. The second important precaution is that users must be protected to continue to represent themselves and their needs. This is what provides for the diversity of what is available on the Internet and is the continuation of the culture and regenerative quality of the early time-sharing communities. This is what makes it possible for a user in Benin for example, to spread the Internet to other users there, and for a student in Finland to start the linux project that has been developed by thousands of others into an operating system that gives Microsoft competition. Those who might want a different type of network, as I have heard some large corporate entities in the United States explain, as they want to be able to more carefully choose who will do what functions for them, can do so in their corporate network as part of the larger Internet, but they must not be allowed to impose their special demands on the larger Internet community. The reason for this is that then users in MILNET, for example, will be required to do things in their network that do not serve their needs, and the concept of an Internet will be violated, leading not to the further growth and extension of the Internet, but back to a single network, to one that serves only a few commercial entities at the great loss to the many other users on the Internet. The other precaution that follows from understanding these essential characteristics of the Internet is that commercial entities want to carry on certain experiments in how to subject various aspects of the Internet to so called "competition". They must not be allowed to do this in a way that affects the whole Internet, but must be restricted to the particular network that they develop for their commercial purposes. Thus the commercial corporation that is being planned by the U.S. Government to sell off parts of the Internet's essential functions must not be allowed to control anything but its own commercenet. Those who are interested in such experimentation should be advised that they will have to form their own network which can be connected to the Internet, but that such experiments can only go on inside their own network, and cannot be imposed on the rest of the users of the Internet. To do otherwise is to jeopardize the fact that only a minimal requirement is necessary for all to connect to the Internet and this is only that which makes the communication across the many independent networks that make up the Internet possible. To do otherwise will mean the obsoleting of many machines and cutting their users off from communication with the rest of those on the Internet. Thus the corporation that IANA and NSI have designed, or that the Boston Group has proposed must not be allowed to take over the essential functions of the entire Internet. Instead such corporate activity needs to be restricted to an independent commercial network that can be part of the Internet but cannot be allowed to impose its special requirements on the others who use the Internet. This might mean that the .com machines will become part of a .com network, and would be able to communicate with others on the Internet, but not impose their "for sale" and speculative practices on the users in the educational or scientific communities who make up much of the Internet. Before there are any plans to change the form or structure or management of the Internet, it is crucial that there be an assessment of the special characteristics and functionality that must be preserved and a plan created for how to be certain that this is done. Since both the IANA/NSI proposal and the Boston Group proposal are for structures that are to be limited to a commercial network, and not imposed on the Internet itself, how then can the essential functions of the Internet be administered in a way that represents the cooperative and international nature of the Internet itself? My proposal provides for a prototype cooperative research program involving researchers in any country or region that agree to participate. These researchers who will be part of this program are to be responsible for carrying out the investigation and inquiry among online users to determine the general characteristics and functions so that they can propose a plan to safeguard these crucial characteristics and functions. There is one final lesson from the history and development of the Internet that it is important to consider when trying to determine how to form a more international system for protecting and administering the essential functions of the Internet represented by the Domain Name System, IP numbers etc. Usenet was begun in the 1979-80 period by graduate students who were part of the Unix community. The invitation to join Usenet which was handed out at the January 1980 Usenix conference explained why it was crucial to develop an online network, not to form committees. They describe why it was crucial for those who were interested in developing Usenet to actually use the network, so that they "will know what the real problems are." It is with this goal in mind that I created the design in my proposal for a prototype where researchers from a diverse set of nations or regions will utilize the Internet to figure out how to create the necessary cooperative, protective forms and processes to administer and support the essential functions of the Internet. Just as adhering to the principle of relying on "using Usenet" made it possible to grow Usenet, so the principle of using the Internet will make it possible to scale the Internet and create a means for a shared oversight of the essential functions and to solve the problems that arise along the way. The Internet is the symbol and manifestation of hope for people around the world. As more and more people communicate on a worldwide basis. the foundation is increasingly set to find peaceful and productive ways to solve the many serious problems that exist in the world today. Also, however, this vision has its enemies. But the U.S. Government has the proud distinction of being the midwife of the achievement of achievement of the 20th Century represented by the development of the Internet. If there are those in the U.S. Government who recognize the importance and respect that comes from giving birth to the communications system that has spread around the world with such amazing tenacity and determination, they must find the means to treat the decisions and changes needed to further develop the Internet with the proper care and concern. ----------------------------------------------------------------- Footnotes: (1) http://www.columbia.edu/~rh120/other/ifwp_july25.txt (2) See chapter 6 "Cybernetics, Time-Sharing, Human-Computer Symbiosis and Online Communities" in Netizens: On the History and Impact of Usenet and the Internet, IEEE Computer Science Press, 1997. A draft is available at http://www.columbia.edu/~hauben/netbook (3) Describing this transition, Vint Cerf wrote: "The basic objective of this project is to establish a model and a set of rules which will allow data networks of varying internal operation to be interconnected, permitting uses to access remote resources and to permit inter-computer communication across the connected networks. (4) Louis Pouzin, "A Proposal for interconnecting packet switching networks," Eurocomp Conference Proceedings, 1974, p. 1023.