Skip to main content

tv   Technology Policy Institute- Internet of Things  CSPAN  November 24, 2017 3:53pm-4:48pm EST

3:53 pm
policies that affect their industry. we heard from verizon's vice president and darrell issa. in conversation took place aspen, colorado. started.l get this panel is on the internet of things. they are net is a wide-ranging term. describing commute -- computing devices and machines connected to the internet and to each other. i am slowly connecting my lights to alexa. you have to do new things like worry about your lightbulbs updating their firmware. my wife is not happy with this, but i keep telling her it is getting better and better.
3:54 pm
soon turning off the lights will be as easy as flipping a switch. household items are the least important parts. there's the future to allocate resources more efficiently. are two principal challenges which we're going to talk about. the network. .hatever structured we need it may need ubiquitous coverage but they don't need to send video to each other. how do we secure that network? it is one thing for terrorist and know i need more ice cream but another when you're talking about yourself driving car or the electrical grid. we have a great panel to discuss these issues. will go to discussion. ralph is the vice president at joined from the
3:55 pm
fcc where he guided u.s. wireless broadband policy, and before that he shaped medications policy at the white house office management in budget. relations,overnment she is responsible for promoting the company vision to provide next generation connectivity. she is a former capitol hill staffer who worked with chris dodd and the election of barack obama and tom dashiell. he briefly say served as chief economist to the fcc, his current research includes public broadband policy and markets. filesharing and net neutrality.
3:56 pm
issa represent people of california's 49th congressional district in the u.s. house of representatives. his committee memberships include the house judiciary where he served as chairman of the subcommittee on the internet -- and hedell been a speaks to these issues not just as a representative of his district but a practitioner. he holds 37 patents. during that time he served as chairman of the consumer electronics association. young is vice president for public policy of verizon for ,e identifies emerging issues and assessing key technology and communication trends. he offers practical experience. he worked at verizon's r&d technologies ip, and image
3:57 pm
compression, and he has 10 patents. he's a member of the comedic asian society. 27 more to go. david. i can talk a lot faster than this if you want me to. so, let's get started. i gave a quick definition of the ot. maybe you can expand on that a little bit. what do we mean? it is a pleasure to be here with the congressman and my esteemed colleagues. thank you for scheduling this panel. it makes iot seem like petty minutia. but, i do have a quibble with the agenda. you have put iot after ai. i think the connected devices that are part of iot are input to the applications that we
3:58 pm
think about when we think about ai. that term, the narrow definition, not the super intelligence that will rule us all, but the narrow verticals of applications and transportation, and health care. a number of other segments of the economy. the answer to your question is, iot is everything that is connected. we can break that into smaller categories. enterprise versus consumer. maybe to set the stage it is useful to talk about aspects of why it is interesting. it is growing a lot. you look at the analyst forecast, analysts are never wrong. the forecast is two times growth over the next five years, probably likely more than that. they differ on where we are starting from.
3:59 pm
by some counts we're at there are a lot and it is poised to grow more. there is some skepticism around that, as it is all just type. i think there are organic trends that are driving everything that could be connected, eventually becoming connected. driving down the costs and up the capability of devices. that is generating a lot more to produce data. we have analytics to make use of that data. we have a network bandwidth that is making more room for connected devices. there are signs pointing in the direction of growth. we heard a lot of that on a prior panel, in terms of the application that could make use of all the state of. saving lives -- all this data. saving lives, increasing
4:00 pm
productivity. if you get into what is the next layer below all of these trends, there are challenges. insurmountable, but things are a little more complicated than they might appear. some of the challenges are technology-based. a big one is power consumption. as we think about billions of devices out there, a good number of them are not going to be plugged into the grid all the time. [captioning performed by the national captioning institute, which is responsible for its caption content and accuracy. visit] >> they're going to need battery power of some sort or the ability to generate their own power. there is some trade-off with their ability to communicate and their energy consumption that will drive certain considerations in terms of the architecture going forward. there is potential trade-offs with how you program these devices. they will be out in the wild for five or 10 years. have you can some updated -- how do you keep them updated? there are some challenges in the market. not all devices can talk to each other. you have some ecosystems that
4:01 pm
you need to pick as a consumer before you really take advantage of some of these technological advances. there are issues of consumer trust in devices. there are other challenges out there. twottwo out better -- i put out that are most relevant for this room because there are horizontal challenges. those are spectrum and security. as you have billions of more devices come online, we need resources to make sure they can perform their function. we need to make sure as the number of devices gross, we do not see the -- grows, we do not see the number of attacks grow. >> yorty have something to say, so go ahead. -- you already have something to say so go ahead. i am done. >> i am going to jump in here. rob knows this stuff more than i
4:02 pm
do. in fact, i'm at the opposite end of the spectrum. some months ago, he called me up and said how would you like to give a paper on cyber security, and i said i do not know a dam thingabout the -- a damn about that. i said i do not even have these things in my house. i do not have a smart refrigerator. i said, wait a minute, i have a smart tv. i thought, i had to connect this to my wi-fi so it is on the internet and it is a thing. ok, i do not have to worry about it too much. and then i was reading something -- it is a samsung. the samsung tv's have these things where you can gesture at it. or you can speak to it. yes, you can speak to it and it will follow your commands. ok, i then read it also sends that to a third-party. hears, incenset to a third party for analysis -- it sends to a third party for
4:03 pm
analysis. it also have a camera so you can gesture to it. you don't not have to use a remote, you can make signs and stuff. they also sent that to a third that to at also sends third party. here is the kicker. the tv is in our bedroom. [laughter] >> all of a sudden the internet -- it kind of hits home. occasionally, my wife and i have to move the tv out of the bedroom. that is what the internet of things is. that is kind of the other side of what rob was discussing. that is what it means in practice and reality. [laughter] >> that is on youtube, right? >> ok. that is all i wanted to say. >> privacy and security.
4:04 pm
obviously a big deal. the also raises questions of the best part of what this infrastructure is all about. in this case, apparently we need lots of and with 4 -- bandwidth for video streaming. ashley, has your company's think about what the demand for this will be -- how do your companies think about the demand for this in the future? what is the investment for the internet of things? >> thank you for the opportunity to join this panel. when you think about the internet of things, you usually think about the device itself. but it is really a system that is made up of different parts. the device is an important part, but another important part is the network that connects that device to the internet or the thirdnd systems and the
4:05 pm
parts that are important are the back end systems where the data is collected, with analytics take place, where the control of the system occurs. those three pieces are all very important. on the network side, there is no one-size-fits-all. streaming andideo 5g. connections are going to be incredibly powerful for internet of things and the low latency of five g are going to be extremely enabling. but that is not the only requirement, because not all internet of things applications require very high bandwidth, upstream or downstream, or very low latency. of'll see a diversity solutions for connecting these devices. for some, your broadband connection with wi-fi will be perfectly adequate. for others, you will want it to work anywhere you go so you will
4:06 pm
want a very widely deployed lte that isike m1 specifically designed for internet of things. one of the things that makes it well suited for connecting devices is that it allows for very low-power operation. rob mentioned the importance of that. it also does not deliver the very high bandwidth. applications, that is not the most important thing. for older ubiquitous connectivity, you may look to satellite. >> i agree with everything david just said. just to bring up to a higher level, when legato and we think about what do you need from -- what? pun? tjat ahat a [laughter] >> the key is spectrum.
4:07 pm
they all have different characteristics. of the the networks future are going to require each of those in a mix of those in order to provide and serve the diversity of applications that the iot will result in. we are focused on the enterprise side of the equation, was role require ultrapetrol -- which will require ultra reliability and coverage. which satellite does provide. we are also strong believers in being tech neutral. a allould either -- ll different applications could benefit from all those technologies. some of the mission-critical things will require the satellite for coverage for basic connectivity and then the ground
4:08 pm
systems come in for the lower bandwidth applications. >> modest talk about the policy aspects of it. congressman, this is probably a good place for you to weigh in. what do you see as the biggest policy impediments to develo ping an internet of things, and what his congress's role? >> the interesting thing for policy individuals is we have to unwind the bad decisions of past policy decision. -- thatnot a small- is not intended to be a joke line. it is very sincere. when you look at how we have sold it and make decisions. i will not name names but earlier there was a discussion about how did we come up with the wi-fi, which today is the clear backbone of so many internet of things, including the switches that your wife continues to be confronted with.
4:09 pm
junk spectrumly nobody wanted that they could throw for useless things. i should not the useless,, because baby monitors were considered, yes, the baby would cry and you would. in the next room on this well powered device. it was sort of your placement for the other junk stuff. gh 32d these other-- althou megabits is suddenly in again. you look at this history and say, ok, if you are going to start again and if i could put everyone in this room with their agendas, i would say what works for people were not in the room, which is your wife, the consumer. and you would say, ok, what do we need to do. the first thing is to recognize the technology of today is no longer and with -- bandwidth can find. meaning i make a product and it
4:10 pm
can do x amount of bandwidth with an antenna. these products that can listen before talking, that can operate in very diverse ways with very consumptionser means you can have devices that are better than they once were. where do you see it? do you see a wi-fi where the newest wi-fi is finally a mesh network that is somewhat intelligent and covers all the available band and uses a third band so it is not trying to actually make the decision about what you are doing on a band that was in fact junk band but rather went to one that troubles further than the cells? so when the cells cannot talk to each other, the devices in the mesh network can. those are succeeding in spite of, if you will, history. let me quickly go through two
4:11 pm
things. i promised everybody that i was going to comment on this being an artificial intelligence device. the switch that you are talking about, that you are putting in is inur lighting system fact in many ways a shovel because you can push on the button and it turns on and off. it is ae horse becuasause little smarter when you hook it to your iphone, and it is even a little smarter, well-trained, may be a well-trained dog when you tell alexa you want to turn the lights on at 15% on red, if you have that feature. it is not yet -- we are not yet talking about artificial intelligence. artificial intelligence is clearly that next step where it continues to think and learn and get smarter than anyone director of -- than any one director of it.
4:12 pm
none of it particularly needs government involvement, which is a good thing. you will innovate out without us. let me bring you back to two closing items. when i was a young lad, there were three things that came to my house. i am one of six children, so we always debated which when i was related to. a post man, a milkman, and a bread man in the each delivered to the house. the milkman and the bread man are gone today. that does not exist, although amazon has sort of replaced it. the fact is that at one time, the government made a decision that there is a universal delivery obligation. in other words, there was a service that was there -- essentially the service was free even if you paid for usage, and we made sure it hits every point on the planet. is trying tornment decide about this essential new service, have we fallen short in
4:13 pm
three areas? 100% coverage of our entire co untry, even when it is inconvenient, as the post office does. real standard setting by government or private enterprise for true universal low-power -- mid-power. and obviously the question of dietary high-power. -- proprietary high-power. that is receive a question of are we there. have we really done that? i will close even though i said i will already close with, then we get into cyber. that is bigger than the today.ion for obviously, we will all have to work on it and government does not have answers and the private sector does not yet have a perfect answer, but my view is that if you go back, not to the shovel but to the milkman and
4:14 pm
the bread guy and asked the question of, is there one universal guarantee to every point in america and if not, is that the government's primary role? ensuring universal access, which is not done just by licensing. and has to be done by initiatives be on the. at. >> you have been thinking about the institutions thatdthat do d . securityt for ito security. how does that relate to what you just said? >> it is me. ok. [laughter] >> no one is going to count on the congressman to do it. [laughter] >> to you. >> let me discuss the institutions issue in the concept of cyber security. the cybertion
4:15 pm
security issue associated with the internet of things. that has generally been perceived that the internet of things is a whole lot easier for hackers to put viruses on. they tend to be unprotected, nobody who has them -- like me - - is paying any attention to them, and, in fact, we have had at least two very serious attacks attributed to the services, one of which was last october. dime, whichlled runs dns servers. therefore, it took down a lot of people who use that dns server. inther one at a company calle december. and thee new viruses, protection against them is very little. when rob asked me to do this, i sort of said i need to read about this.
4:16 pm
broadbandabout internet technical. also the department of homeland security saying what are the problems and how do we fix them. it was basically the technical aspects of how to protect against the internet of things, cyber security. most recent think i have seen came from cablelabs, which goes a step beyond that and it is actually quite good. i would like for you to take a look at it. it was published this summer. they are simple stuff, ok? hard to do. it is also focused on consumer goods, too. there are a lot of other internet of things. use the best current software,, at the security on the design phase. simple. we designed the internet like that and we would have -- if we designed to the internet like that, we would have a different story. automated security updates. vulnerability management, unit?
4:17 pm
you know? configuration testing. all of this stuff. technical stuff and i said, thank god some of the allstate it because i do not know anything about this. i do know the big problem was how do we actually get this stuff in place. there is the institutional issue, ok? the industry we have here that makes this stuff -- you have to think about what this industry looks like. those are the people who are doing the security. first of all, with hundreds of manufacturers making little computers but going refrigerators. most of them are not in the united states. so they are going to be part of this story. , dozens of sellers who were selling refrigerators home alarm systems -- dozens of sellers who were selling refrigerators, home alarm systems, televisions. they are not even in the same industry. most consumers do not know they have a computer in these things.
4:18 pm
at least with people who have pcs, you know you should download antivirus. but with a smart refrigerator, you do not know in you do not care. the last thing that makes this even worse is the real danger is somethingou will loop into your refrigerator and somebody will figure out how often you open the refrigerator door and, therefore, whether you are home or not. the real thing is watching attacks on other people. if you were one of the people who launched it, you do not even know it. the damages not for you. -- damage. is not for you how do you think -- damage is not for you. how do you fix that? you need customers to be involved. that is what you need. couple of conclusions about this.
4:19 pm
have i talked enough on this or should we do a little more? >> to add another question to at, many years ago i was on a panel where we were looking at software reliability. and always kept coming back to the question of connectivity versus reliability. is very similar trade-off. that the device can be -- is there a similar trade-off. here? jerry mentioned to the report that came out of cablelabs. debut can start off by answering that. and then i would like to hear from the network people, how they view that. >> you can go to cablelabs informed insights to check out our recommendations. in terms of your question, sco tt, if you think about iot narrowly, perhaps the answer is yes.
4:20 pm
a more open iot at the device level could lead to reductions in security. to david's points, that iot is much broader than the device systems in its applications. i do not really think at the level that it is important to consumers there is that trade-off. provide an to example, the lightbulb in your house does not need to be talking to the entire internet. is it is open to the entire internet, it may be compromised in the way jerry just talked about. it can't talk to your home hub, a gateway that has appropriate firewall and has no reduction in -- toy for that label as you as a consumer. when we think about devices and systems and applications and conductivity, how is has connectivity grown in recent history? it is through network affects. to the extent, iot is going to
4:21 pm
be growing because there is value in devices and systems being connected in a smart way. >> at the very highest level. in thismake sense industry to design the devices to be able to go through a hub or something with more controls? or is it cheaper to build something that can connect to anything? problem incan be a the iot space. there are different architectures. you can have device-device communications. device to hub, cloud to cloud. in terms of the trade-off ,etween security and cheapness i think there is a way to bridge that gap. it gets a little to the institutions question that jerry was outlining.
4:22 pm
as with the congressman in terms of developing standards and making it easy for new suppliers of iot devices to build to those standards and allowing consumers that eas o -- tthe ease of conductivity that comes from those platforms. the problemget past of the small manufacturer with five people and their company trying to stand up for their product not thinking about security, they can easily adopted the industry standard as the base code for their device, get it out quickly. >> from the network perspective how do you -- >> we actually do not think connectivity and security are mutually exclusive in any way. there is such a diverse range of applications and products. in the iot system itself, you have the applications on the other end. the way we like to think about it is each layer of this stack
4:23 pm
atl require a different look what security requirements are necessary. we do not think they are all created equal. should the electric grid have the same type of security requirements as my young daughter's electric toothbrush that connects to my phone? probably not. and so i think as an industry, what we are looking at is whatever security requirements and best practices almost do we need to implement at the device level or network architecture design level to make sure each application has the security requirements it needs to meet a consumer's expectations. we also believe in consumer education. data is what is becoming extremely important in terms of enterprise and industrial entities taking connectivity to the next level. and what can we do with that
4:24 pm
data and improving our operations and maintenance> making? making sure consumers understand the type of data being collected and how it is being used as externally important. >> you asked about the hub versus directly connect ed devices. having a hub can be advantageous for having consumer devices operating in the home. that just moves the security point to the hub, which is connected to the internet. you still have the same security now it isu hopefullyt now i hopefully being taken care of to make sure it is being done properly. most of the tools are already available to produce the secure internet of things, whether you are talking about digital certificates that will authenticate devices and allow them to be managed securely, updated securely, encryption
4:25 pm
over the network that will ensure the integrity of the data flows. all of these tools are available. the problem is not everybody either knows they are available or uses or chooses to use them. even worse, there are some really basic practices that are often ignored, like creating and internet of things device that has a default password that is published, so it is easily accessed by anybody who wants to go in there and change the settings. the tools are available, it is really just a matter of producing the education, both at the consumer level and across the echo system. -- ecosystem. >> just because the tools are available does not mean they will be used. >> let us look at the incentives as it is never going to be good enough. the education is never going to be good enough. decades ago, i was a young armor officer training for how we were
4:26 pm
going to beat the russians. in the next war and then the soviets -- beat the russians in the next war and then the soviets. there is a difference between how the u.s. went to war and how the soviets went to war. we went to war with every single tank. the tank had a radio and a full command.of the every private have the same level of training of what would happen if we went to war. the soviet union had commander control that use hand signals for the most part and literally did not put radios in anything except the command vehicle. they put up fake antennas on the rest. part of the concept was they did not trust their soldiers. skip forward. you will have losses. in the case of the soviets, our goal was to knock out their control and then the rest of the guys did not know what the hell to do. surrender was hopefully the choice.
4:27 pm
when we are looking at these trillions of devices and we know there will be failures, the first question is how do we 1000, onelure of 100, million is not failure of the system? rather than saying we will have better education, change the passwords, we will put a hub in, the first thing is the system has rebuilt with an assumption that there will be failure. an assumption that the systems -- whether it is the electric grid, the internet itself through the dial-up service protection -- has the ability to protect itself and deal with inevitable losses. some of those losses are electronic devices that simply go haywire. ther ee is the device that creates its own noise. like this one they had to substitute just a moment ago. that.k you start with
4:28 pm
the second part, which has a bigger public interest, is a debate that is not settled. i am very personally involved in it. the former fbi director came be fore congress since were under of that he had no ability to get the information he needed from the san bernardino bomber except by forcing apple to create an active remote backdoor into the product, and that is what the magistrate had ordered, feeling that there were sufficient constitutional protections in this order. a matter of weeks later, an israeli company for $1 million gave him the data he wanted. a few weeks after that, a cambridge professor showed how we can do exactly what a group uy like me said.e ha we have to have a real debate over whether encryptions and protections are real and unbreakable, because if they are not real and unbreakable, then they will be exploited.
4:29 pm
likeey are, organizations the fbi, cia, and others will be constantly disappointed that they cannot get what they desperately wants. this is not a new arguments. back in the 90's, quite friendly, we had 128 bit encryption as a cap. the only way we got past that was that microsoft and others began having their software hacked and given away for free all over the world. it created a pressure because of its failure. we are repeating that right now just as we did with jsu in craftingusit -- with just in crafting software so it cannot be easily copied. do we have an absolute right and an obligation to make these things secure? if so, that inevitably will empower those who use it for nefarious purposes to be protected. back as all the way to the highest levels of governments around the world -- that goes
4:30 pm
all the way to highest levels of governments around the world. >> let me finish appeared.. up here i certainly do not. -- what a finish up here. i certainly do not want to get int the way of the eclipse. [laughter] >> a wide variety of players in this game. howl will we get everybody on board? there are two extremes here. proper first setting up o voluntary organizations and we will seldom. l them. all the way to the ftc saying we will have regulation. i think i identified a number of them. one is a voluntary approach. the problem is this works well have a small number of firms all of them can internalize this. , when you have hundreds of thousands of firms, it becomes
4:31 pm
difficult to police this. >> ok. >> this would work if you could say, let us do the equivalent of a good housekeeping seal of approval. if everybody signed up for that and consumers gave it to them, it would be a good solution. i'm not sure this is the best solution for iot. second would be voluntary standards but with legal enforcement. lawyers would love this because, in fact, product liability becomes the way to enforce it. that works. well, the problem with this is enforcement is costly. it may completely miss the externality, which is to say i have this on my refrigerator. they create all this damage to somebody else. i do not care about suing the guy for this. this is perhaps not a good way to do this. third,ory, this is
4:32 pm
joint efforts with regulatory commissions and the industry. joint efforts. and example of this might be epa and energy star. the appliance people gets put on energy star. that is currently done with the epa. that is a form of joint regulation in an industry. it works well when you have a lot of firms. customers are not involved in the problem and there is some externalities. uis a potential for regulatory caption, which wee about. be concerned this could work well for iot i think.
4:33 pm
if you do full regulation, that to not respond toto regulatory change will. it will be slow to change. in the technical field, that is not a good solution. to go with the joints regulatory voluntary operation. i think that would be a good solution to come up with. it is enforceable, it can change the technology, and it is something that will work well with iot. it will not be perfect. we have had --what happens to the tale of things? -- tail of things? but you do need to have institutional structure around which laws and standards can be built. it is important to think that through. totally private versus totally public and i think a mix of the
4:34 pm
the two is the way to go. >> you said there is a lack of incentives. i would push back on that. one internet is no of things. if you are a business and you are connecting your fleets of trust, you have strong incentive to make sure those systems are secure and you're going to require the supplier you are ensure that willensur security and you will take all the steps you need because this will cost your business in terms of reputation or actual loss or damages. if you are a city in your deploying a smart traffic management solution, you have a strong incentive to make sure that is not packable -- packable. that somebody cannot get in there. i think there are incentives for those sorts of things.
4:35 pm
it is when you start to get down to the security camera for your fr --that you ar ebuying you are buying from a small company. you may not care if it is used in a denial service attack. that it causes that externality. i think segmenting the use cases and the problems and identifying where the incentives are weaker might be hopeful. there are also mitigation solutions that can be put into place. ai came before this, and one of the applications of ai could be identifying abnormal communications and traffic the videoo that if camera starts engaging in abnormal behavior, the ai in your home router or service provider's network could identify that, flag it for you, and suggest you may want to unplug that thing and check the settings and whatnot. i do think there are a lot of incentives already in place.
4:36 pm
if you do fail to properly secure your network and consumer data is lost, you ar eliable to be -- you are liable to be called in front of the fcc. >> that is probably the one i is the least resolved right now in cyber security. the federal commission -- which case. michael doherty they could not name best practices. they did not have a standard. and yet they wanted to hold me for whiche for things they had not published in after the fact went into. you look and said. ther eis w here -- you look and say there is where the rub is.
4:37 pm
government cannot keep up with best practices. but that means when government enforces -- or i guess when lawyers and forced, they are alluding to the idea that you should have known. i think that is where a lot of what the internet of things is going to be all about is industry screening best practices, governments quickly making it known that they are from a regulatory standpoint, a need to know if you are going to manufacture hardware or software for the industry, and then pushing that down. of course, the update question comes along. where this all the part is where we get back to the consumer. i have a bias because i came out of the consumer electronics industry. if they haveto is, to read the instructions, you failed in the product. i know that is not just for the men in the room. the reality is that if we do our job right, the internet of things will be products which
4:38 pm
may have to be shut down when you do something wrong but will not require the consumer to be trained and educated and updated into best practices and it will not require the product, on the day it is launched, be anything other than updatable as threats change and evolved. obsolete double win the processors can no longer meet the demand of a future product. that sequence of events is where -- and i think you have done a great job of playing it. it is a hybrid of things we have done in the past. is the federal trade commission liable when somebody knowingly is deceiving the public? yes. are they the basket of after something happens you must not have lived up to best practices? i think not and i think that is where we are going to have to change government view, keep the plaintiff's lawyers out of it whenever possible, particularly
4:39 pm
when we are talking about technology that updates hourly. >> will said. well said.ions -- >> when you're talking about standards and companies and regulation. it is kind of a place where these efforts might come together. without naming names, who is not paying attention are not willing to go along with what you speak of? >> can i name names? [laughter] >> use a code word. >> exactly. is serviceity here providers and we work for the cable industry that provides
4:40 pm
conductivity, we have an ongoing relationship with the customer and we feel a hefty responsibility to do what we can to protect our customers. is notrt of relationship uniform across the iot system and there is a potential of those who want to just shift units and have an ongoing relationship with their customer. it is a challenge and i think some of what has been discussed is exactly -- is not exactly the right type of approach and i would quibble with jerry's energy star analogy. just to be clear with that writes thee epa rules and it takes years to play that out. i think it moves too slowly for iot and i think it is a more complex problem. i think it is up to industry to step up and sort of provide the tools that the ecosystem needs to address these issues. plug, and as of a
4:41 pm
an example for what the industry is doing in this space, cable thrizon is a member of is. part of companies writing code. this is not just talking about best practices. we actually write code that enables security for iot devices. eade that could availabl through open source. these solutions are starting to take root. the success is important to keep in mind. it is tricky. if we go back to the example that jerry raised, the dns provider being attacked by the maria botnet, for the magnitude of the attack in perspective, researchers have concluded the population of devices that comprise the botnet that led to that attack and took down
4:42 pm
twitter and netflix and other had a steadyces devices.2000 and 3000 the public a lot until you think about the billions of devices out there. that is less than 1% of devices on the market for iot today that were available for a major event like that. i do not know that we will have perfect security, but certainly i think there are tools in place now to make incremental improvements. >> that is a different kind of question. setting aside the issues you have had with your spectrum -- >> what are you talking about? [laughter] >> entering into this industry, are certain aspects of setting up a network for this sort of future internet of things easier? arer more difficult -- there more difficult things like getting into some of these groups? so we actually view it
4:43 pm
as an unbelievable opportunity in a lot of ways. i think with the congressman is talking about, how do we undo a lot of the things, we're talking aout something with significant amount of spectrum unencumbered right now. that leaves an opportunity to look from the start of the tools from talked about. the best practices that are out there. focusing on core industrial industries, like the utility sector, oil, and gas that are desperate and looking for an advanced level of connectivity that is not quite meeting the needs they have right now as they look to further automating their systems. the story is not new to anyone in this room, but we are looking at it at the other side and we really see a unique opportunity to deploy greenfield spectrum in the bestt uses
4:44 pm
technology we have end walls we currently have today. >> a little time for questions. >> the answer is 11:43. ,> as one of the authors on the report on security i was intrigued about what rob was saying. we said that the iot industry needs something that is the equivalent of wi-fi alliance that can put pressure on the device standards and defined test procedures this software -- test procedures. this soccer takes a step for -- this takes it a step further. congratulations. >> we just set it up last year, but the intent of the form is to make it a consumer brand
4:45 pm
that people can recognize it. make a transparency issue and make it easy. >> not that iot has a unique security problem, the internet itself is not very secure. developed,as first it was not all secure. this is kind of a process that new technologies go through where you have the big challenge to make it work at all, an d thn you worry about how to make it secure. there is sort of a change in that mindset where we really want to start designing securi ty at the beginning. and the code is going to help. >> question? ok. that seems like a good place to wrap up. thank you all. think the panel. -- thank the
4:46 pm
panel. [applause] we can all go upstairs and upside and see nothing, if it all works out. we will have viewing glasses for you all. [captioning performed by the national captioning institute, which is responsible for its caption content and accuracy. visit] >> congress is on break for the thanksgiving holiday. monday, the house will work on 2018 federal spending current funding expires december 8 legislation on an additional $44 billion for disaster relief requested by president trump to respond to hurricanes and wildfires. the senate continues to work on judicial nominations, including a vote monday to confirm a u.s. district judge and the district of columbia. later in the week, senators are expected to vote on the senate tax reform plan, which includes
4:47 pm
oil drilling in the arctic and repealing the health care law's insurance mandate. watch live coverage of the house on c-span and the senate on c-span2. c-span, were history unfolds daily. c-span was created as a public service by america's cable-television companies and is brought to you today by your cable or satellite provider. >> the kennedy clinton dinner, hosted by the new hampshire democratic party speeches from representative john delaney of maryland and tim ryan of ohio. this is two hours and 15 minutes.


info Stream Only

Uploaded by TV Archive on