|
Tuesday, July 01, 2003 |
Internet governance: domain name law & policy + the story of ICANNFisher/Zittrain We will talk if internet governance in way that synthesizes all of the sessions thus far. Why are people so interested in ICANN? Who should run the outfit? How the mess came about
I. The origin of the mess. Cerf, et al inherited a mess from · Dave Clark – ITF – nerds thought up protocols and tried to keep rest of us away [ok, I paraphrased that quote mightily] [Z is going a thousand miles an hour here] · Discusses RFC’s · RFC 2555 is a retrospective of 30 years of RFC’s o Talking about numbers § Numbers could change – aren’t very geo portable § So, create a layer of mnemonic names that would match to the numbers – and apply as a universal resource. § 1984 name resource improved with top level and second-level domains. Allowed named zones to be created so name servers weren’t overloaded with details of every name in the universe.(distributed domain names) § Jan Postel’s computer at USC maintained the root. § Created mirror roots to have backups § In 1993 – anchoring the root got to be too difficult. And too routine. So NSF put out a proposal for a fixed fee contract to maintain list of names. § Network Solutions – won the bid § NSI told NSF we can save you money by charging people fees to rent the name… o Problems § NSI making a lot of money – others wanted to do that too § Expand the namespace? § Corporations discover cybersquatting · No longer the province of the nerds – people have expectations of where names II. Legal Intervention against domain name holders o So lawyers looked at trademark law (which is country specific) § Types of trademark infringement · Identical marks on competitive products · Similar marks on competitive products o Test : likely confusion § Similarity of appearance § Of sound § Of meaning § Measured against context of marketing environment · Similar marks on noncompetitive products o Ultimate issue : likelihood of confusion among consumers o Consumer confusion – what does it mearn? § Questions of source § Endorsement (e.g. Rolls Royce Radio Tubes) § Post-sale (e.g. Ferrari – the product looked like a Ferrari sans marks) § Initial interest (e.g. · Dilution o By blurring it (e.g. greatest snow vs greatest show on earth) o By tarnishing it (Amex didn’t like condom mfg using “don’t leave home without it”) § These are the four types GAPP agreement is compelling countries to come closer to IP standards. So, lawyers researching ways to end cybersquatting and landed on dilution test. First cases involving Toeppen – agreed to dilution claim. Int’l standards are a bit tougher in proving dilution. Problems and limitations of this approach
Limitations of dilution doctrine
III. Taking over the domain name system
IV. State of · Org chart had a box for everyone (the history of org charts is hysterical) · Is this a public resource taken from us? · Alternative roots or keywords could become competition for the a root. i. Lessens the impact of what ICANN does · This web of organizations with custody over subset of rules (TM vs Institutional) and has created a system for dealing with DN disputes very different than what was started. TM rules have gained increasing power in this realm. Why Care?
|
Applications: Wires and Wireless
Professor Yachoai Benkler, Professor Lawrence Lessig, Reed Hundt, Former FCC chairman Leslie Vadasz, Director Emeritus, Intel Corp Lessig: After a morning of theory, we have two people who can talk of the practical aspects (Hundt, Vadasz). Hundt: Learned you should talk of way you want to see world, not the way it really is. Want to talk about way that US and So here we go
Are we on an interesting track? Are the only country in the world with a choice of more than one platform for broadband. We’ve encouraged both firms to create a market using their core. So, are we pushing this model with the thought that one of these firms wins and the other goes out of business? Or, cable takes this market and telco that market? So, what is the thought? You won’t find it written down anywhere and no one would share it if they knew it. This is one heck of an experiment. This is a moment of high anxiety for the firms themselves. Telco’s are losing $10 per subscriber. Wireless broadband alternative could create a different environment – gives us 3 cards instead of two. With respect to wireless – FCC is god of policy. Congress doesn’t interfere. Regs are done in jargon that you can’t understand. What can be done? FCC 1) Can sell chunk of wireless spectrum to a manufacturer. That mfg could not go into retail business. 2) Spectrum would be a private park – a proving ground. 3) FCC should create license for that spectrum 4) Should provide interconnect to WiFi providers The two big broadband providers should be engaged in inter-connect scheme from the get/go. … Government ought to re-think vertical integration. … Ought to get rid of voice subsidy program and replace it with subsidies to consumers to get physical link to broadband. … Government is adding $200 per tuner to deliver HDTV – imagine that kind of subsidy for broadband access. Opens up possibility for new kinds of devices. Vadasz I don’t like the two firm model. I want to talk a little bit about one company’s experience to move internet activity through wireless fwd. I was originally skeptical as wireless did not seem robust enough to support network activity. But, I am coming around. What happened is the FCC turned loose a bit of spectrum for whatever use. And, look at the innovation that has occurred. Designing, producing and selling a product is not enough to succeed in this market. The problem was, despite 10 million access cards sold – you had to fiddle with your configs at each access point to get to the point that you could communicate. Clearly the first thing you have to do is participate and to engage the community outside that is involved in consuming the product. We used a couple of techniques: 1) Invested money in companies creating this ecosystem. Engaging them this way helped assure some consistency 2) You then engaged some of the other, established companies to encourage them to modify their approach so that a standard emerges. 3) By the time we introduced the centrino model for notebooks there were 15,000 easy access points available At about the time we were doing this - -Mr. Clarke from the government decided to try to help us by declaring wireless insecure. So, to put some reality in the picture, Intel became a major deployer of the system. Today, every Intel office has wireless access. You must run VPN, but this is a reasonable effort to provide security. Last but not least, had to involve the regulatory agencies…. Make sure your voice is heard. Need more unlicensed spectrum. Would be very desirable that spectrum policy deal with capabilities of technology today. We in the past have defined both the allocation of the spectrum and the application that spectrum can employ. … Get to point we buy Internet as transport and not as a service. Entrepreneurs will supply applications. … Lessig: Tomorrow the theme is how the dinosaurs are defeating the innovators. In this context the dinosaurs could not control the internet – but in terms of broadband, they do. And this can be disastrous. You’ve told us a happy story how wireless tech can blow away the dinosaur’s business model. But that assumes the dino’s political savvy is defeated by our own. V: I have no political savvy and you ask a very valid question. I think that the problem, and exposure to those making the rules, is they are really trying to do the right thing. But, they are looking at these issues through the existing infrastructure. Wireless will be chaotic for a while (and big companies not do well in chaos)… Unless municipalities take a role, there is not much alternative to what is there. SO, best bet is wireless and open standards and people brave enough to get into the business. But I think that the chances are also that the verizon’s of the world will get into wireless and there will be a big landgrab. Y: Will the unlicensed national information infrastructure be available? You got an agreement with DOD and some major players as to what the constraints will be. And the DOD gets together with some number of players that will then get together to decide some operating restrictions. So, that is one story of how the band could be expanded – but be within control of big guys. Then there is the idea that sw defined radios could be created to communicate data. Use TV band spectrum otherwise not used. Finally, Verizon has a competitive strategy so that if you buy our dsl – you get access to wireless access attached to our payphone locations. So there are three very distinct areas available – which of these are indicative of opportunities, of control. Is it an opportunity or a possible crash… V: I would make a deal with verizon to make sure easy access applies. The goal here is ubiquity. But there is a lot of tech to take advantage of existing spectrum better. We are damaged by auctioning license of spectrum. Would be better off looking H: Who makes money in the spectrum? Telecom, mfg, retailer? Where did the infrastructure come from? Lucent, etc loaned the money to start the business. All the models for spectrum have been tried. Teledesic got its spectrum for free and it did not succeed. Lessig: The world of logic gives V. lots of faith that problems can be solved, If he can hire you (Hundt) how best to get Washington to understand why they should not sell the future to the dino’s what would you do? Hundt: Dino’s can fly. Elephants can dance. Telcos can be data access company. I don’t think government should manage that. But govt should provide efficient solutions. There are not political obstacles for procuring spectrum. There are obstacles with respect to price. Completely unlicensed spectrum introduces cost in another way. Lessig: In context of unlicensed spectrum model, Intel has been negotiating with those installing the access points. Hundt: I don’t know if that cost is significantly different than spectrum allocation. We do need to supply both buckets and let market tell us what works. Vadasz: It is very hard from within a business to engineer change even though the world is going that way. So, how do you create pressures to force the incumbent to change? … Hundt: Government should continue to adopt minimal efforts to assure change innovation. … Fundamentally we are consolidating to save. We are worried about loss of capital in that sector. Q: Zak: Could be some things done if municipalities work with hobbyists to work with fee access… municipalities really want access in the parks. Dialogues in the groups could potentially get this done. V: It is hard to herd cats. There is more talking going on than we know. Some municipalities are running fiber. Q: Zak: point is that both have an interest in seeing it happen… Q: could you talk more about the vertical layers Hundt: yes. Layered approach is very inciteful way (orginall NYU article in 1996) to view the model. …We should be concerned about one firm monopolist that is encouraged by govt to integrate vertically. comment [] 6:38:57 PM
|
Logical: End to EndProfessor Lawrence Lessig, 1935, Edwin Howard Armstrong set up an experiment in Empire State building where he broadcast organ recital from transmitter in Long Island where he demonstrated FM technology versus AM technology. This was his third extraordinary invention. {discussion of attributes of FM vs AM} At the time RCA owned dominant power and control of radio of that day. The president of RCA was not happy – “I didn’t think he’d start a revolution – start up a whole damn new technology to compete against RCA.” So, he worked the FCC and Congress to protect AM radio against FM competition. He also fought the patents Armstrong had. Because TV used FM tech to broadcast sound. For six years RCA fought those patents. So, after 6 years, Armstrong was broke – RCA settled for less than litigation costs. Armstrong begged wife to let him settle – wife refused – so he stepped outside his house – on the 13th floor. 1) Will never possibly work 2) “Damned if we’ll allow the creation of a competitor to ourself.” So, the Television Well, question came up to stream video on internet. Eliminates channels. 1998 alliance built between AT&T/Excite at home – question raised to allow broadband cable network to allow streaming of video to computers competing against cable broadcast to tv. “We didn’t spend $60 billion for a network to allow the blood sucked from our veins.” Innovations Let’s talk about some key innovations on the internet Kahn/Cerf kids World Wide Web – Tim Berners Lee ICQ – P2P messaging invented by Israeli kid, sold to AOL for $400 million HoTMaiL – invented by Indian immigrant sold to MS - $400 million Napster What’s the common feature here – all done by kids and non-americans. The internet invites outsiders to innovate. Best ideas come from these outsiders. The architectural feature we focus on today is the end-to-end (e2e) character of the logical layer. End-2-end says intelligence at the edge – keep the network simple. Old network model – innovations that benefited holder of monopoly were encouraged, otherwise discouraged. In e2e model, because network itself has no intelligence, then you can’t discriminate. It is simply what the users want that determines how the network develops (sounds like a market model to me). Enter David Isenberg (1997) – working on AT&T telephone technologies – discovered problem with the AT&T network was the intelligence in the network actually created limitations to evolving the network. He wrote and talked about within AT&T and AT&T got upset when a paper “Stupid Networks” got published (he left AT&T then). So, he started selling the ideas of stupid networks. But his ideas were actually rediscoveries of ideas by (Saltzer, Clark, Reed) in 1981 articulated the idea of theory of e2e (V. 1) talks very narrowly about the way a network must be designed. If you have functionality that cannot complete be conformed inside the network then the functionality must be placed at the edge of the network – i.e. implemented only with knowledge and aid of application) e.g. if you want to assure reliability, security of data – place that function at the edge. (Versus at the hops within the net) Point is, e2e is a pragmatic intuition of how to design networks. Preferred Designs – a principle (bias in network design) – if choosing, choose intelligence at edge of network. This is in many ways similar to the first amendment – where we allow some laws to infringe using processes to weigh harms versus justifications anytime freedom of speech is invaded. Policy Everywhere – now, the designers were really engineers figuring out how to get machines to talk to each other… but the architectural principles, affected the social and regulatory behavior of the internet. Building on the simple common denominator of IP you can build and extraordinarily complex network. Technical consequences 1) Flexibility in how it develops a. Don’t have to change things as you develop functions 2) No coordination among network users required to implement changes or applications (e.g. can we agree on this change to allow this app to run?) a. Talk of voice over IP development – what does it take to do this given the architecture – packetize the voice – and spit it out – then sw at other end to recombine (simple, heh?) b. Talks about GOPHER (Univ of Minn 1991) i. 1993 – browser and WWW and $$$’s ii. Univ decided to charge for use of gopher iii. The greed killed it… fast death iv. No one wanted to risk future of company based upon what Univ of Minn wanted to do with Gopher in future v. The network protects itself – new ideas replace old Competitive Consequences of e2e 1) Maximum Competition a. Commons i. Significance of word - think about resource in a commons a resource that everybody has equal right to gain access to… (can cost, not free, but no discrimination) but Lessig is talking free sw, free resource – no one has proprietary control of who gets access. ii. Language (e.g.) everyone has in common iii. iv. This one idea captures how everyone thinks of resources from that moment on, but 1. You cannot have a tragedy if resource is rival-less. 2. e.g. Language (intangible resource) – the more people use a language the more valuable it becomes – so no tragedy here 3. So, need to think is this the kind of a resource that can experience a tragedy v. So, researchers said of internet: 1. It will die – resources exhausted (96, 97, 98, 99) as they adopted the model of the tragedy of the commons. But reality conflicted with the theory. More users actually added more capacity – made it stronger 2. The tragedy model failerd b. E2e builds an innovation commons i. No one is in position to decide who can innovate and who cannot ii. Economists again say – oh no tragedy will occur iii. But, the more people that innovate – there is no destruction of opportunities for others to innovate. iv. So creating property rights or applying that model in this context makes no sense.(Imagine selling the English language – wanna buy a vowel?) 1. Imagine property system in music – you have to buy rights to use a440 (a sharp) 2. But this would do harm… v. Competitive advantage of innovation commons 1. rather than have one innovator (i.e. AT&T or RCA) you have many innovators. Competition to innovate is radically increased 2) Minimize strategic threat a. Strategic behavior i. Monopolists behavior to clear the market of competitors that harm them (defensive monopolization) you protect yourself against great new idea that will destroy you (Core of govt case against Microsoft) ii. Talks of Netscape/Java idea to change competition model by building a platform that enabled people to write applications that were general to all OS platforms iii. Gates saw the threat, he believed the idea would work. Let’s displace Netscape with IE! To control the outcome, the chokepoint of the new strategy. iv. So govt said MS was protecting platform from competition and US Court of appeals agreed v. You can say that MS is a protectionist vi. E2e destroys opportunity for that kind of protectionism 1. This lowers cost of innovation 3) Consumer financed growth a. As utility, you look at cost of innovation and weigh how it will benefit you before embarking upon the path towards innovating b. But, this approach was not the way that developed the internet – it was because the consumers financed the development – it was what the consumers wanted that drove innovation. c. 3G vs. 802.11 i. 3G conceived in old model – 1. plans based on tech that is now wildly out of date 2. when it came time to deploy, 802.11 techs were cheaper and easier to deploy 3. No one was in position with 802.11 nets to decide which innovations would be allowed But, the policy implications of e2e is it is heaven – it induces competitive development of infrastructure and applications. The dark side is that “we are on our way to hell”. What we are seeing is pressure on e2e layer placed by owners of physical layer and content layer. They are acting to corrupt e2e layer. How? 1) Policy based routing (physical) a. Represents layering of new tech to make it increasingly possible to let physical owner to decide how to treat a packet based on a number of selected criteria. 2) xBox and cable a. MS now believes in e2e based on conversations held between MS and cable companies as a result of introduction of xBox tech. i. Cable conversed about whether to deploy xBox on “their” network. ii. So cable picks what they like to run on their net. 3) Content eats the conduit a. Media concentration i. Recent FCC decision – monopolizing news sources ii. Turner says he couldn’t have been competitor if media was as concentrated as it is now iii. FCC says the Internet will solve everything 1. But if same companies that own cable, tv, radio, newspapers change the internet architecture (say choose which packets go fast) then the internet becomes part of the problem. 2. This claim is thus circular. FCC isn’t going to allow internet to compete against cable… Solutions Increasingly the internet is becoming similar to the cable television network. It picks what runs on the medium. People trying to resist in each layer 1) Open Access debate (physical) a. Access to pipes (ISP’s) b. Secure competition and assure e2e neutrality c. In US failed – d. And this requirement is about to be repealed by FCC 2) Neutral Network (logical) a. FCC to regulate to guarantee neutrality b. Prohibit preference based on content 3) Free culture movement (content) a. Minimize power of content layer Abolish the free in any of the three layers and you will destroy the opportunity for competition. Is there any recognition of the problem in any government? No. The basic claim in US is there is no need to do anything – leave it to the market to take care of itself. Questions Q: If you look at RCA and AT&T they had world class innovation labs. Does the notion of innovations commons ignore scales of innovations that are actually barriers. Intel and MS are building size and complexities of research units, and not just anyone can enter that space as you need a level of complexity, a scale to do the job. A: Absolutely correct on great companies having great innovation labs. But, there was constant struggle between business units and research units. Businesses will never cannibalize themselves. But, it doesn’t require scale. Best innovations relating to complex systems are done in manageable steps. Q: Judges and legislators are worried about property argument. How do you deal with that? A: Good point. We have forgotten what makes property function in a society of the whole. From economic perspective, propertizing everything is not efficient.
Eldridge case – Freeman, Buchannen and others said property argument is wrong. Need to have balance. Q: If you leave this to market – then the market won’t secure openness or freedom. You are claiming that environment for innovation is damaged? A: Yes, there is a balance, a mix that is most encouraging to innovation.
|
The Technical is PoliticalAccess to an open information environment Dr. Yochai Benkler, Offering a map on what we will be talking about this week. Overview
Models of Communication Trying to describe the structure of the network in terms of who gets to say what to whom and who decides who gets to say what to whom Internet model represents spectrum of ways to organize production and exchange of information Models
He tried mapping components of a model (communicative functions in a communications channel). He wants to show difference wo who gets to control the functions in the different models. (e.g. Communicative functions in a Broadcast model – interesting graphic). Basically diagrams shift of control, or universe of discreet components, that are controlled at the core, or the edges. (Obviously, internet model most of the components at in the edge). The Stakes
These stakes seem to align in favor of more open processes (internet-like qualities) State of One way to organize these problems is to think of the problems in terms of the layers of the communications process Content layer Logical layer Physical layer Physical Ok, want to send page with some scribbles – I need hardware – some lead in my pencil, paper, copper wires, a mac |