 Okay okay okay good afternoon ladies and gentlemen thank y'all so much for joining us attention please my name is Daniel I am a program coordinator here at Berkman Klein Center we're so excited you could be here to join us just a couple of event reminders and announcements so next week our lunch will be technology disruption and the practice of law will the profession survive with Raj Goyal and Ari Shahadi and December 2nd we also have an event coming up from 9 a.m. to 12 p.m. algorithms law and society building rights for a digital era if you'd like more information on either of these events you could go online to our web page cyber.harvard.edu slash events and lastly this event is being live webcasts recorded so be mindful you are being recorded and if you don't want to be recorded well okay and without further ado I'd like to introduce Berkman Center senior researcher Ryan Buddhist to introduce our speaker. Hi good afternoon everyone and thank you for fighting your way through the crowd of law students to join us here today who would have thought that a week and a half after a presidential election a bunch of law students would be interested in discussing executive power apparently so big surprise there but but we know that we have the more interesting conversation here with Scott Bradner so thank you all for for joining us if you would like to to tweet about this you can use the hashtag bkc harvard and so so I'd like to introduce Scott you know internet governance has been something that the Berkman Klein Center has been interested in and worked in for quite some time going back to some of the original ICANN meetings to then in 2010 heading up the accountability and transparency review team and then most recently being a part of an expert panel that helped the NTIA assess the most recent news about the IANA transition and that proposal so this is something that the Berkman Klein thinks is a very important issue and so we're really pleased to have Scott here talk about IANA and why it's important but not for what they do Scott Bradner was involved in the design operation and use of data networks at Harvard University since the early days of the ARPANET this included designing the Longwood Medical Area Network and the New England Academic and Research Network he was founding chair of the technical committees for these networks as well as the Corporation for Research and Enterprise Network Scott has served in a number of different roles in the internet engineering task force which is the network of engineers that develop the protocols that help the internet run and function he was also elected trustee of the internet society where he was vice president for standards and secretary to the board of trustees Scott recently retired from Harvard University in 2016 after 50 years working in the areas of computer programming system management networking IT security and identity management so please give a warm welcome to Scott so good afternoon so we're gonna talk about the IANA well let's start off talking about silliness absurdity mr. Mr. Cruz there he's a bright kid he's a Harvard boy Harvard Law School but he was remarkably stupid in this he created a persona that the internet was going to end and these are some of the headlines that showed up the little the this eagle down in the lower left-hand corner is off of his website was special website he set up to talk about the evils of this transition of the IANA going off and doing something state sued Obama gives away the internet and with it our liberty federal judge just let Obama give away the internet seven days before Obama gives away the internet and national security all of that is complete we were being videotaped so I can't really tell you what I think but so let's go back a little bit I'm gonna go back and forth a couple of times here let's go let's go back a bit to the beginning this whole IANA thing started with a need to coordinate so this whole IANA thing started with a need to coordinate information not to control information but to coordinate it started with a network ad hoc network research group which sort of wanted to figure out what to do with a ARPANET that was about to be deployed and then the series of requests for comments on talking about different technologies or different proposals for technology fellow named John Postel then a graduate student took on the job of being a series editor and then he took a few years later took on the job of coordinating what were then the identifiers of different types of communication over the ARPANET which was the socket numbers the concept the the name IANA showed up many years later and 1988 although the function that was there from 1972 on the the actual name showed up in 88 so coordinating more than just sockets coordinating IP addresses internet protocol addresses when IP addresses started to be handed out the IP addresses turned out to be a little hard to use I don't tell you to send mail to me at 128 103 836 which is the which is the IP address of the computer it used to be in my desk and William James Hall so we come up with this inner indirection called the domain name system and somebody has to coordinate that there's a route the root service the .com.net.org the hierarchy there is makes things easier so all of the parts of what is today's IANA were in place in night by 1984 all of those different pieces of it by 1984 but let's go back again to the beginning once upon a time computers were expensive and big my iPhone is about the power of those large computers that took up buildings back in the late 60s the federal government could not afford to give every institution their own computer so they wanted to share it and the Defense Department was had a lot of researchers out there doing research and they wanted to share these computers because I couldn't give each one each of the researchers their own computer so they decided to build a network to share those share those computers the network was a packet based network was a packet based network in that that you broke the data up into chunks and the chunks fiddle fiddle through the network to get to the destination the concept of this came was done by Paul Barron back in early 60s and at Rand he was off trying to figure out a way to build a network that would survive a strike nuclear attack because he figured that if we could if the communications would exist after nuclear first strike nuclear attack then the enemy would feel think twice about hitting a first strike because we could retaliate but he also wanted to give that same technology to the Russians so that we would think first think about doing first strike but it was it was that technology but that wasn't the reason the reason was to share the big computers so ARPA went off and they built a network they built a network out of this packet switched networking switched networking each of those chunks of data trundling through the network has enough information to make it's a destination that's a destination address and it contains some data that it wants to transport it makes no assumptions as to the underlying networks capabilities that may packet may be lost duplicated put out of order or anything it just it doesn't know it doesn't care that that is a problem that's an issue for the end systems to deal with not the network itself no service guarantees there's nothing here it's just I send you some packet a bit packet of bits and it gets to you maybe and that's all that's that's what the this network was so the ARPA built a network 69 were the first nodes by 82 there was a couple of hundred nodes international then shortly thereafter commercial networks showed up started to show up and by the mid 90s we had a significant number of commercial networks late 90s we had commercial networks all over the world the interconnected mesh of those networks made the internet that is the internet it's that interconnected measure wasn't the ARPA net ARPA net was a separate network but when the ARPA net moved over to TCP IP and interconnected with these other networks that became the internet but no one cared no one in the power structure that is the traditional networking people said this doesn't work there's no guarantees there's no service guarantees no reliability guarantees no quality guarantees why wouldn't anybody use this it's a toy it's irrelevant it is completely and utterly irrelevant and that was great for us who were playing around on the network but IBM refused to bid on the the switches that made up the original ARPA net they also said in 1992 quote you cannot build a corporate data network out of TCP IP end of quote that's a direct quote from an IBM researcher that's because their definition of a network was entirely different it was a heavily controlled heavily regulated heavily managed network where everything was guaranteed and the internet had none of that digital equipment went off and built their own protocol AT&T just refused to be involved in this at all they just thought this was silly and irrelevant just like the aerodynamic theoreticians that said the bumblebees couldn't fly the powers that be said the internet couldn't fly and that was great because that meant that the regulators ignored us because this was never going to be of any value why do we need a regulator why do we need to think about how to control it how to manage it from a governmental point of view we don't have to because it's never going to mount anything well it kind of did amount to something we'll get back to that the IANA function went into a steady state around that time and for a decade was just basically just plunking along nothing really special sending out doing coordination of protocol parameters for the IETF doing some address allocations to regional registries they didn't even do the allocations to the end systems like when I got addresses for Harvard I sent a note to John Postel said I need addresses and he sent me back a note said here they are but very quickly after that it was got delegated to regional registries around the world so the IANA only gave out big chunks of addresses to the regional registries and but the IANA ran the DNS top level it was the database which pointed to the computers that ran each of the top level domains that's what the IANA did and March 1994 John Postel wrote a RC 1591 in which he said there's a small set of top level domains and it's extremely unlikely than any other top level domains we created so this is all the internet needs for coordination there are only these few centralized functions everything in the internet is cooperation it's you and I agreeing to use in the same standard for email for email transport you and I agreeing to use the same standard for web display there's no protocol police out there saying you have to do that it's just if you and I disagree on which one to use it ain't going to interoperate and it's not going to get as profitable as it could be but it is there's no there's no requirements no coordination no nothing other than that very simple level so the only centralized functions are we have to agree on what the protocol parameters are when I send you a packet it has a field in there says protocol when I put a 25 in there you have to know that means email when I put an 80 in there you have to know that means web that's there's no there's no requirement that you have to use it it's just that in order for you to interpret the packet when you get it you have to we have to do that so it's one central place where that's recorded that's the IANA we have to have some distributed consistent way to distribute IP addresses because IP addresses have to be globally unique because they're telling you how to get to and what is the node that you're trying to get to and what is your address the saying who you are any globally unique way those have to be globally unique there are alternatives that people have thought about but at the moment that's the way it's set up and that's done by the IANA the coordinated the top level is done by and we have to have a single DNS at the moment and they're different with their theories ways you can get around that but at the moment there needs to be one definition of saying here are the top level domains so when you go to Ford.com you get to the same Ford.com as your mother gets to when she drives it from her house but that's it that's the only thing we need to coordinate and that the actually centralized management and the internet everything else is distributed it's Harvard runs a piece of its DNS MIT does Ford does everybody runs their own little piece and they cooperate all of the internet service providers cooperate they interconnect with each other but it's a business decision on how to interconnect when to interconnect it's entirely their own there's nothing requiring it no governance no management and then the proverbial turn it the rotating device the NSF said that NSI could start charging for domain names up until up until then there were free registrations I free as in the government paid for them but in 1995 September 95 the NSF said okay you can start selling these domain names a fee of $100 for two years 50 bucks a year this was big this was money was up about a hundred million domain.com names at that time so that's 50 bucks a year for a hundred million names that's real money and there's money to be minted here because but Verizon the but the verisigns network solutions is the only game in town so what do we do there are some individual country codes that sort of saw their hand the money on the wall and went after it like Tuvalu with dot TV and they sold they they leased out their country identity to a company in I think Canada to sell off to sell off domain names and dot TV so but other it was this was big it really got everybody pretty excited the IANA has three functions it's got the domain names the IP address is the protocol parameters the only only makes any difference to anybody has domain names because that's where the money in the lawyers and the trademarks and all that kind of stuff are you can't get away from the fact that this is awful stuff it's an awful swamp but it is the swamp that we've got it's also the only thing that the media can vaguely understand is their main names they talk about internet addressing they mean domain names it's not actual internet addressing but they mean domain names their natural reaction to anybody in the capitalist society is oh there's a shortage let's make more so it's a lot of call for new top-level domains including from John Faustel proposed that number of us looked at that and said it doesn't solve any real problem if you're IBM comm you're not going to move to IBM dot foobar because you have to unwind the web and repoint everything and you can't do that so it's not really competition for existing registrations might be competition for new registrations but the more domain there's the more domains there are the more likely the only one you're going to remember is calm so it really doesn't gain you much but it we need to do something because we have monopoly and monopolies are ugly and evil and all that kind of stuff and so there needs to be something John proposed to move the control of the domain name system and by the way the other stuff that comes along with the IANA just because it was there not because it made any difference to the internet society which he was a board member of at the time as I was and they were going to set this up an ad hoc group to set up a management for the domain name system under the internet society doing it on their own not blessed by any governments the government the US government have been paying for the IANA function through ISI internet the University of Southern California is for a number of years up from from the beginning but he thought let's just spin it off and set up our set up our own own separate organization within the internet society and he set up a ad hoc international ad hoc group and it was ad hoc group the representatives from a number of big big groups like the NSF and the World Intellectual Property Rights Organization international telecommunications standards and groups like that but it was really an ad hoc group it was doing something on their own they had not been blessed by anybody to do it they just went off and did it great idea the idea was to come up with seven new TLDs what we do seven new TLDs but one of those TLDs was .web and a company or actually an individual going under a company name of the image online design sued John over .web because he claimed that John had a verbal agreement to sell .web to him now John was not in the power position to actually selling the domain names at that time the TLDs at that time but the image online design said he did they said he reneged on this John later told me that this lawsuit and the lack of support he got out of ISI in response to the lawsuit was a reason that he reenergized his activities to try and get the IANA off out of under out from under U.S. government control into a private organization because he felt very burnt and also there was a redirection of the the route by Eugene Kasparov who went to jail over it to redirect to alternate so there was there was a lot of stuff at play Department of Commerce who had taken over the general management of the contract with ISI for IANA functions saw that we needed to do something so they put out a request for comments saying we need to do something here's some ideas we've got what do you think and they want to set up mechanism for domain name registration and management thereof now they meant all of the IANA functions but that's the only one they mentioned meanwhile John had been proceeding to try and come up with a way to get the IANA function out into a private organization he'd been working on that on his own and with a bunch of us during that period much in the IETF and elsewhere he consulted with Ironman magazine or who is responsible for the RFC out of the Department of Commerce and by the end of 1997 he had a plan was a plan for what he called institutionalizing the IANA function and that was sort of accumulated in January the end of January 1998 and a conference in London called reengineering the internet Ironman magazine was a main speaker there he said what the US government was going to do sort of forecast what they were going to do which is to support the idea of spinning out the function to not have it under ISI but have it under a separate cooperation John didn't make it so I presented in instead I presented a paper and you can see that on the on my website shortly after that magazine or the Department of Commerce put out what they call the green paper so like a preliminary rulemaking saying here's what here's the direction we think we want to go we want to go to an independent private company to private nonprofit to manage these functions to coordinate and manage the functions and the functions were allocations and number blocks not the actual allocations to the individual but the blocks to the regional registries run the root name server or not run the servers but often make make the edits in the file which indicates what top-level domains exist decide on when new top-level domains should exist and coordinate to technical parameters same set of functions that we talked about before so this is the IANA functions and the green paper said here are the ones that we want to move out for this all of the IANA but nothing more there's no internet governance here it is simply the IANA a lot of comments later on after get reading those comments the Department of Commerce came up with the white paper which is the rulemaking which is actually saying okay this is what we're going to do this is our policy we're going to set up a private company we're not going to set it up we're going to invite invite people to come up with a proposal for a private company to allocate IP addresses run the root server or manage the root server system decide when new TLDS and technical parameters same functions that's all nothing more nothing less so when that white paper showed up a bunch of groups decided to go off and try and figure out what that meant this is a picture from Geneva self-appointed meeting from the International Forum on the white paper and down in the lower left-hand corner there I think that's Jonathan Zittrain because as was mentioned earlier the Berkman Center was involved in that they coordinated that meeting I went to that meeting and it was very clear that the people at that meeting there was thousands of them maybe not thousands at that meeting but certainly many hundreds at that meeting they weren't there for coordinating IP addresses and domain names and technical parameters they were there for internet governance but that's not what was on the table and one person got up at the mic and said we shouldn't be spending all this time worrying about country code top level domains because the internet is going to destroy countries and we won't have countries anymore that's the mindset of the people that went to this they wanted something different the internet at that time was another universe and another universe where we could start anew civilization could start anew it didn't happen in case you had noticed they wanted internet governance by the people or something like that most of us thought that that was a well-needed vacuum we didn't need internet governance the lack of internet governance was a positive not a negative but these people were out there trying to come up with on a new organization that would control the government the internet John produced a proposal for ICANN this internet corporation for assigned names and numbers he submitted it on October 2nd of 1998 14 days later he died complications of heart heart surgery I believe four days later department of commerce said that they would accept his proposal to create or to bless this this organization six days later the ICANN had his first board meetings hadn't been incorporated yet but that the first board meeting it was a closed meeting and unfortunately that forecast a lot of what the ICANN did was a very closed organization not particularly welcoming of help input from elsewhere but it was incorporated and by the end of November department of commerce and ICANN signed a 10-year memorandum of understanding and the memorandum of understanding said here under the IANA functions you're gonna do they call the DNS management function they're really the IANA functions IP number blocks root server system new decide when new top-level domains and technical parameters same set of things no internet governance just those technical functions if you translate the I the MOU into what actually happened the I the department of commerce is over ICANN or was over ICANN but the only thing they did was say yes or no to changes to the file which defines the top-level domains and they'd actually never said no they just was a bookkeeping function where ICANN would say this is the change the changes we want to make send that off department of commerce department of commerce would look at it and send back saying okay but all it was is a list of domain top-level domains and IP addresses of those servers that deal with each other top-level domains that's the entirety of the control the US government had over ICANN now they also could come in and slap them around if they got those two capricious which they did frequently and department of commerce would occasionally come in and say you really shouldn't be doing that you're not open enough and what and what like and ICANN would say oh yeah we'll do much better and then proceed to keep going the way they were so but that's the entirety of the control was to check off on whether the changes to the root zone file were okay in theory the department of commerce could have maybe said take Cuba out of that file now they never did that in theory they could have done that but that's about that's the most possibly ridiculously they could have done if they did that most of the world would have ignored them because they know what the cuba's address is they just plug it back in again it's not a big deal so ICANN was formed by John Postel to institutionalize the IANA deal with these technical boot keep bookkeeping functions no government's just technical bookkeeping functions John expected that the IANA would be ICANN would just decide on new top-level domains but not many of them there's a few may he thought maybe 130 a year for the first year and then 30 or so a year after that has to deal with I domain trademark issues because domain names are trademark issues whether they should be or not is irrelevant and ICANN's worries about the security and stability of the domain name system so ICANN was formed as a little organization to replace John Postel and Joyce Reynolds and maybe one or two other people so a few competent geeks here's ICANN's budget it went from a few million bucks to a hundred and thirty million bucks that's an awful lot of money to replace a few competent geeks top-level domains they just suddenly decided that more is better and we auctioned off a thousand top-level domains didn't gain anything other than confusion most of them didn't succeed in the marketplace most of them have failed or close to failing some of them are vanity ones dot forward they're vanity ones and they're they're not going to go anywhere because they're only for the individual company and here's another statistic I came bylaws I went from about 7,000 words to about 30 37,000 words if you need 37,000 words to say what you're doing there's something intrinsically wrong because that ain't the vision John had this is not technical parameters it's something else I'm not talking about a lot of things not talking about the IGF for the ITU trying to muscle in and the WISIS and other things like that and also really not going to talk about much the ICANN public goodwill because they started out with a lot of goodwill and blew it away almost entirely which was a really unfortunate but after ICANN got set up and running the world started to change the world started to change a lot and that was because there still wasn't any internet governance and the internet was affecting too many people the Arab spring was all effect a lot of that came out of the the communication from the internet the many governments see the internet as a intrinsic threat because it's providing as a fellow from China once told me information that confuses the citizens by information that the government didn't provide the internet it's a big deal covers a billion people and governments want to control it so for many many years the US was able to go in and say don't you worry about that we may be a dictator but we're benevolent dictator we're out for everybody's good and then Snowden showed up and that was proven not exactly what the real world was real world was something different than we're out there for everybody's good and that blew away the US is a primary authority to try and forestall changes to the internet governance structure the people wanted to move it outside of the US move it into the ITU or someplace like that the the I the internet leadership technical leadership that's the internet ITF the IAB the regional registries IP registries the internet society W3C they all got together they get together regularly they call themselves the I-STAR and they got together in Montevideo in 2013 and issued a statement which basically said it's time for the US to let go and then later on the Brazilian government and others came up with their own concept that Mundial was a concept of we'll take it over as a perpetually appointed 25 member team to take over internet governance well there's nothing to take over but to create internet governance I guess now things began to get really interesting because the US said well maybe maybe we will let go there's only this little thumb we've got on the wheel just checking the top level domain file but maybe we'll let go if there's a way that the community could come up with a plan for us to let go and that I can or some other organization can do can deal with this in a controlled way and and continue so they wanted one proposal to transition the IANA to some new environment had to support multi stakeholder which means it could not be just governments it couldn't be just industry couldn't be just ISPs it had to be multi stakeholder it had to maintain the security stability and resilience of the internet and meet the demands of the IANA's customers which is the regional registry is the IETF and the domain name world and unfortunately said and maintain the openness of the internet the ICANN doesn't have any role in the openness of the internet the openness of the internet is a local function of an ISP in a particular country at a particular time but the US government put in their statement of what this new organization had to be to maintain the openness of the internet it doesn't make any sense because the function is there isn't anything that does it today and there wasn't anything then so what do you do well there are a bunch of proposals he's a bunch of people they specifically said this is a list of folks that have to be involved the folks who are the customers of the IANA of the IANA plus Verisign because they're run the TLD they run they actually edit the TLD file other global stakeholders things like that and they said they would not accept a proposal which was intergovernmental so you couldn't move it to the ITU ITU being a government-based organization the decisions are made by government representatives and the NTI said we're not going to let you do that it's not a permitted thing you've got to have it so there's multi-stakeholder but who's going to tell who's going to slap I can't around when they get mess when they get messy the department of commerce theoretically did that but didn't really so but somebody really should do that be able to do that if commerce moves out of the picture so they came up with a they come on a community process on the community process which has been signed on by ICANN is incredibly strong that community process can can veto changes to the bylaws can prove changes to the bylaws can remove individual trustees can move the entire board it can really do can really nuke the place I along slog to get there but you can actually do that so a very powerful accountability process then I keep changing it but so I am anti I looked at the proposal that came in said okay it meets the requirements now I've four times I've made this so that the NTI anti-IA logo is grayed out I mean and Microsoft keeps helping me by bringing it back because I probably didn't really mean that I guess so anyway meanwhile folks in the kind of the Congress got all been out of shape because the US was giving away the Internet and we were going to give away our freedom all that kind of stuff the Congress blocked the anti-IA from acting said here's a block of funding block to the end of our fiscal year so they're at the end of the end of the fiscal year you can't do anything you can't spend any money to cause a transaction they're a transition the IANA contract was extended to September 30th 2016 big block from Congress was September 30th 2016 but Congress just went nuts there are pile of people in Congress just went nuts we can't possibly do this all those headlines at the beginning we can't do this because we're giving away the Internet and our freedoms that just going to get expunged the first amendment rights on the Internet are going to go away all malarkey but it's all it was very strong it's a done deal September 30th came and went the funding block expired so it's anti-IA could spend money again the contract the IANA contract expired so on October 1st 2016 we have an independent IANA which is now it's part of ICANN but it's an independent IANA it's under the community it's not under the US government anymore can't be undone this was a contract that expired even the president the incoming president can't unexpire a contract which has already expired now could throw it throw the FCC at it say or we're in charge of IP addresses well that's only in the US it's not clear that anything else could happen of course some of the folks like Ted Cruz may try and do something but it's not clear that they could do anything this is a done deal and note that the very thing that they were most worried about is that the UN or countries would take over Russia and China would take over governance of the Internet even though there wasn't any governance of the Internet they would take it over would have much been much more likely if the anti-IA had not let go because the UN had been already been asked to vote to take it over and if the UN votes to take it over the countries in the the countries of the world vote either in the UN or in the ITU to take it over then how's the US going to say no the US is now a minority player in the in the Internet there are many many many more people on the Internet outside the US than inside so it's an almost irrelevant player so much more likely that if they if they had stopped the anti-IA that they very result that they feared would come true so remember this is technical functions the IANA at protocol parameters address blocks root root zone file new top level domains coordinating functions not governance ain't no governance here these are critical functions they have to work but they aren't governance it's what the IANA doesn't do which is important it doesn't do governance there is no Internet governance the Internet is so damn important it's trillions of dollars a year it's billions of people how can it not be governed it's the only international anything that isn't telephone system the the shipping on the on the on the oceans radios the satellites all those are governed the Internet is not governed how can that possibly be well it was because nobody cared for long enough that it got away from them if they cared to the beginning if they had been a little bit more concerned with what it was going to bring to them then they might have imposed governance early on but they didn't so there is no Internet governance but is that a problem we have to fix use the microphone for the people out there in audio land thank you sure David Belson break it off my I'm the editor of the state of the internet report to be published who under whose authority do the RIRs operate so I know they're independent I know at least Aaron is an independent corporation all of the RIRs are independent but do they is there any sort of oversight or they're just now they see when the John when he first started out was assigning addresses directly as I said I got my addresses for Harvard directly from John but some folks in Europe came to him and said you know let's make this thing easier for you we'll take care of Europe you give us a big block of addresses and we will assign them low so John blessed right being set up and he blessed AP Nick being set up in the Asia Pacific area in the US he had blessed network solutions as the source of that network solutions at some point decided that their bit their future was the domain names and the IP addresses were getting in the way so they want to spin that out they spun that out as Aaron that went all the way through the US government then at the Network Resource Council the federal federal councils in the US which blessed the idea of spinning it out without ever questioning whether there was an authority to assign addresses and so that's an open question there's there has been threats to sue all our many of the RIRs I don't know if all of them there's now five over whether they have authority to do what they're doing but the way they've all been set up when ICANN came in they were also they were all remodeled so that every one of them has a separate policy development process that is completely open to everybody you can go participate in Aaron's policy development process even though you're not getting any addresses from Aaron you you can do it if you're if you're in Europe you can participate they there's a bottoms up policy development process which is the policy for assigning addresses out of Aaron same thing for ripe same thing for AP Nick and Latin Nick and and after Nick so they're they're operating under originally John's delegation and never been challenged thank you I can understand people's feelings that somebody must be running things but I still find the internet to centralize with the central source of names and addresses where's research being done or an effort to have a much more distributed or decentralized internet there's there's been a lot of thinking think research about this particularly in the DNS area for a long time there have been proposals to have ever since ICANN was first formed actually their proposals to have some kind of distributed maintenance of the TLDs but it always comes down to you getting the same for dot-com as I get and you if even if you even if it was distributed in the some kind of semaphore based creation of two TLDs you still have to coordinate the you still have to centralize or at least distribute that mapping so there's there's a lot of thinking about it a lot of people would prefer to have geographic addressing to get away from address space issues so from centralized address space but a lot of things have just not been proven out some people are very serious about the geographic addressing but it turns out the connectivity and telephone world you if you're offering telephone service you have to interconnect at all of the peripheries you have to interconnect that the local offices there's no requirement in the in the internet so when when I was first started had my connectivity at home in Cambridge connecting to Harvard through my ISP connecting to Harvard went through Washington DC because that was the closest connection if we'd had geographic addressing everything up and down the East Coast to have no where my house was and that doesn't scale so there's a lot of thinking about it but so far they haven't come up with an alternative though they're not they're not stopping thinking hi I'm Paula Villarreal I'm a fellow here at the Bergman Center you mentioned the incoming president and do you think this poses an possible scenario of the US becoming more like China in terms of internet like with all these cyber security talk and all that it's certainly possible the way Trump signed on to the cruise issues with IANA seen me sort of capricious cruise mentioned it and Trump just picked it up but it didn't seem like it was any deep thinking involved in that the security issues a big one the recent attacks the denial of service attacks from the Internet of Things stuff those those are very real and then certainly the the gut reaction is to fix that and you fix that by protecting us somehow rather than looking at the root cause which is crappy crappy equipment so the the quick reactions would would tend toward trying to do that China is this very special case there aren't there are a few other countries as well but how China imposed how China deals does this thing is all it connections in and out of China have to go through a China owned connection that's not the case in most countries most countries have multiple service providers they have separate interconnections they don't have the control point that China does but certainly England is now imposing a a monitoring rule and a net nanny rule which is very very very strict that all ISPs operating in India and England have to obey so we're we're on a cusp of perhaps doing that we're moving back towards what the telephone world was which is a a Westphalian model of every country is its own to thief them that's the internet avoided that but with the regulation may move us more to that though can more likely be Europe North America kind of thing rather than individual countries hi this is Jay Carlson not comm there are a couple other things that the internet seems like it needs global at least near global agreement on and we've seen those in the issues of buffer bloat and the related issue of congestion control and I'm not sure how these are handled except through sort of the use net model where pure is the peer pressure is the only real control that people have that you recursively throw people off unless they obey and that's also another model that I am a could have taken will throw people off is kind of a different difficult issue Iana doesn't have any contracts or agreements with individual internet service providers or individual like for Harvard Harvard's an internet service writer for its community I am doesn't have any contract with Harvard the particular issue you mentioned buffer bloat the technical issue it's a real one but it mostly affects the people that have it rather than just the general overall the internet of things these are denial of service attacks from crappy equipment that's a much more serious thing because it does it does affect a lot of folks and has potential affecting us all whether that's done by a central coordination it's certainly not within the purview of Iana today it could be and certainly there's a kind of thing that ITU would worry about but what would you do about it would you say that giving the power to a local regulator to tell ISPs within a country that they have to kick people off whose security camera isn't properly configured that's certainly not something is going to happen from the center it have to be very distributed and it's very unlikely the peer pressure thing is the only thing is there now and it doesn't work I guess I have the microphone so I get to speak I want to go back to your previous discussion about countries effectively partitioning the ethernet the way China and Iran is doing but Russia seems to be taking just the opposite approach that they have lots of connections into them and one of the theories behind that is they don't want anyone else to have the ability to shut them off because they're afraid that people will blame them for all these hacking incidents and then cut them off why would they blame Russia for but on the other hand there have been some private groups I would call them terrorist groups that just cut off cables in the Persian Gulf to Egypt for example to try to shut that shut it down when they didn't want it but other groups that are like ISIS that are launching attacks from their areas so it looks like there are these things outside of what governance organizations vigilanteism that tend to come into play when things get really out of hand with distributed denial of service attacks botnets and other things doing harm and how do we ever control that well it's a great question to ask and it's not clear there's an answer yes there's the vigilanteism is a real issue the question of whether the Egypt cables were it wasn't Egypt's cables with cables off Egypt were cut on purpose or just a trawler do me stupid not determined but it could be either could have been either we know that countries spy on other countries by using the undersea cables but there's whether you have a court it's not clear how you can have a coordinated thing other than intergovernmental and I think that cure would be far worse of the disease rich borough of Harvard Extension School with the growth of IPv6 usage and the explosion of numbers that have to be assigned in and distributed in management what changes do you see in this this organizational structure coming out of that I know DNS can handle IPv6 but there's a lot more address spaces to be given out the DNS is one of the huge successes because it is completely distributed operationally distributed yes there's a central there's a root which says point points to various different levels down but it's completely decentralized so it scales extremely well so I've got I've run my own piece of DNS at home for my own for my own sobco.com Harvard runs a piece Harvard delegates a piece over to Dana Farber for their sub-domain so as far as the scaling goes it's not a real issue that's the number of addresses in v6 yeah it's big but we're not seeing any issues at all with that kind of scale we have some scaling with the routing system unless we can get more hierarchy in the routing system these routers get bigger and bigger and more and more expensive the backbone routers there are million ball million dollar boxes now when we thought that would never happen when we were first doing this stuff but that's the scale of the routing table these days and that's not that that is a real issue it's not clear how to how to push that push back on that that was one of the things that I would have liked to address when we did v6 we wanted to look at the routing infrastructure see whether there's a better way to do routing and we punted on it and that I think was a mistake but I think we're out of time now according to the clock on the wall so you you mentioned that I can sort of has a spotty history of transparency and then you talked about the various accountability reforms I was wondering you know looking forward if you could talk about whether you're optimistic or pessimistic about the role that I can is going to play you know with respect to the broader internet you know community of stakeholders and you know how you see them playing nicely or not and the opportunity for sort of improving on their past track record I'm not sure that it's relevant as I described that the functions we need out of my IANA I can functions are not governance functions the only thing that's remotely like a governance function is coming up with new TLDs and they've already blown that a thousand new TLDs just means commas valuable and nothing else is so I'm not sure that it's relevant they've the IANA has never interfered with the assignment of protocols or with this address assignment with v6 as rich pointer v6 each of the registries got a big block of v6 addresses some of them we never go back to IANA for another block because that's the size of the block they got the first time was so big so there's no opportunity to enforce on anything and protocol parameters they don't the main name system is about the only the only thing we got left so even if they are completely wacko and transparent and spend and move their budget up to 250 million dollars all that just means is more people are tied to pay for it not that it gets in the way of anything great well thank you so much for