93% of the data invested by the weather models are satellite data, and the bulk of that by far is somethingd 85% of that segment comes from the polar satellites. with the responsibility to do all we can to maintain the accuracy of national public forecast, i would have to place the polar satellites highest on our list. geostationary and the founders and instruments there are not far behind, but there's other elements within the infrastructure that noaa relies on to do its portion of this job. there's other key infrastructure points. we obviously depend critically on a telecommunications backbone and architecture to get the data to the right places, to the prediction, down to the satellites. there are risky points in that system, in keeping that current and robust so there's not multiple embedded single point failures is something we pay close attention to. when you get down to the shorter -- well, let me go the rest of the way down the chain. when the weather data, the central guidance model, images arrive at a weather forecast office, jim and his colleagues use a workstation called the adv

the data centers move to where power gets generated. what is the next hardest thing to move? big data. did it is still really hard to move, even though we have lots of wavelets. that says that the application moves to the data is. if you think about facebook and google and all those folks, they build data center's right were the power is generated. typically near hydro plants and so on. what you find is that more and more applications will move to where the data is. moving those big chunks of data is very difficult. in terms of enterprises in the cloud, there is certainly no reason to suspect that the systems used by cloud vendors like amazon are not capable of geographic replication and redundancy. it is absolutely the case that someone like netflix, for example, could survive an outage of two simultaneous amazon did a centers through geographic redundancy and so on. this stuff exists, and the technology exists within the cloud providers to make sure that once your data gets there, it is not going away. it is just not a cloud provider solution. most enterprises will have thin

93% of the data invested by the weather models are satellite data, and the bulk of that by far is somethingd 85% of that segment comes from the polar satellites. with the responsibility to do all we can to maintain the accuracy of national public forecast, i would have to place the polar satellites highest on our list. geostationary and the founders and instruments there are not far behind, but there's other elements within the infrastructure that noaa relies on to do its portion of this job. there's other key infrastructure points. we obviously depend critically on a telecommunications backbone and architecture to get the data to the right places, to the prediction, down to the satellites. there are risky points in that system, in keeping that current and robust so there's not multiple embedded single point failures is something we pay close attention to. when you get down to the shorter -- well, let me go the rest of the way down the chain. when the weather data, the central guidance model, images arrive at a weather forecast office, jim and his colleagues use a workstation called the adv

93% of the data invested by the weather models are satellite data, and the bulk of that by far is somethingd 85% of that segment comes from the polar satellites. with the responsibility to do all we can to maintain the accuracy of national public forecast, i would have to place the polar satellites highest on ourist. geostationary and the founders and instruments there are not far behind, but there's other elements within the infrastructure that noaa relies on to do its portion of this job. there's other key infrastructure points. we obviously depend critically on a telecommunications backbone and architecture to get the data to the right places, to the prediction, down to the satellites. there are risky points in that syem, in keeping that current and robust so there's not multiple embedded single point failures is something we pay close attention to. when you get down to the shorter -- well, let me go the rest of the way down the chain. when the weather data, the central guidance model, images arrive at a weather forecast office, jim an his colleagues use a workstation called the advanced

and accuracy of the data. news media sources that are trusted tend to be -- that are out there tend to be better trusted sources of data. they build up the trust over time. next slide. even non crowd source data can be wrong. i didn't realize this was going to be on tv. this is all my personal information. i am moving in two days and my new address in my new town is wrong in my car. so google has it okay. street maps has an okay. 412 southwest, south as in north southwest. not southwest is in one street which the bottom is gps has it wrong in the car. even verify -- data that you pay for can be wrong with regard to this data. the promise of the crowd source data, open street maps is the middle piece that i have here is actually correct in terms of my new address. if it was wrong, as a crowd source platform i could go in and correct it. it is like the wikipedia of maps. if something is wrong and i see the information is wrong i can go in and correct it as part of larger crowd. i don't have to wait for a new rele

minimizing, we've seen the tension with private industries feel there needs to be little data kept as possible, whereas on the other hand, law enforcement, and we've seen in the house bill which under the purpose of cutting down -- >> guest: plushed on the bill. >> guest: yes, i was hat the hearing where you testified about that. can you speak to the tension, and the white house proposal appears to be sympathetic to law enforcement's need to be able to access information. why do you see the proposal coming down on that divide? >> guest: i don't know what the outcome will be, but studying the history of privacy law in the united states i think that one of the accomplishments in one of our original privacy legislation was to say to law enforcement quite explicitly, you should really on the collect the information that's necessary and related to the criminal investigation that you're pursuing. it is actually the case that currently and wiretap law that there's min myization procedures and other legal obligations ensuring that information about innocent people is not gathered. that's the

93% of the data invested by the weather models are satellite data, and the bulk of that by far is somethingd 85% of that segment comes from the polar satellites. with the responsibility to do all we can to maintain the accuracy of national public forecast, i would have to place the polar satellites highest on our list. geostationary and the founders and instruments there are not far behind, but there's other elements within the infrastructure that noaa relies on to do its portion of th job. there's other key infrastructure points. we obviously depend critically on a telecommunications backbone and architecture to get the data to the right places, to the prediction, down to the satellites. there are risky points in that system, in keeping that current and robust so there's not multiple embedded single point failures is something we pay close attention to. when you get down to the shorter -- well, let me go the rest the way down the chain. when the weather data, the central guidance model, images arrive at a weather forecast office, jim and his colleagues use a workstation caed the advanced w

so i abrogated all of the data. anybody want to guess? out of 40 of the largest companies in the world, how many of the were using one or more of these -- there was about seven different applications, all delivered out of clout, -- cloud. that's over. best guess, not over. 38 out of 40. of the largest companies in the world. using one or more of these. when we debate the question of will corporations use the cloud, i would say they already have. we clearly are still in the early adoption days. i rattle off some very interesting stories right now of companies you know right now who are using lots of interesting computing and storage cloud services as well. i just wanted to get people to start to think. companies are moving this way. there is a bay area company -- i will not name names, but the cfo walk into the cio's office and said, "no more capital expenditure." the reason why i talk about this as a business model is -- and, by the way, i would be happy to send any of you guys a copy of this book. or if you want epub versions, we can do t

since the advent of data centers. many companies enter into the market. it is not a trivial market to enter. the cost of building these things, by the way, is $1,000 per square foot, which is expensive even by california standards. cost of power dominates this. this has been an interesting space for a variety of reasons, and i will leave it there, but datacenter clouds services is the other component being delivered, right? finally, i will introduce you to last things. first is something that have been very innovative about five years ago when jeff bezos said he would

first, it requires the cfpb to collect data. this is to ensure against discriminatory lending in the small business space. the data requirement imposes limited cost on lenders, but provides an important protection for small businesses, particularly those owned by women and people of color. more regimely though, there's authority over almost all service providers, large and small. the regulations could effect credit, but i want to emphasize it's premature to judge the impact on financial providers or the impact of the cfpb on small business credit costs and availability. instead, individual rules will need to be evaluated on their own marts when, and if they are proposed. the dodd-frank act imposes safeguards on rule making to ensure against unregulatory burdens. this rule makes and adjudication is subject to the act and supght to the regulatory flexibility act. the cfpb is one of three agencies required to have regulatory flexibility review panels under the enforcement and fairness act. they are required to consult with regulator

but when you open the gates in this way enable this kind of data flow from the private sector to the government, it is really the interests of the individual user that i think needs to be safeguarded. and the way you can do that is through legislation. guest: so what kind of legislation are we talking about? would you like to see warrants in terms of using data? guest: we think that you do need additional approval before you intercept private communications. that is the core principle on the federal wire tap law. you have exceptions, but you want us to occur around the edges. you do not what the core principle of judicial review before there is an intercept taking place to be replaced by a new core principle that says that the government routinely gets access to user data from isp's. that could easily happen over time. i think if some of the language in the proposal is not tied up a bit. host: you have endorsed senator ofhy's the security act 2011. why are you in favor? guest: i think what he is trying to do is strengthen data breeze notification. this is an interesting development.

. >> candidate bios and the latest polling data, plus links to c-span media partners in the early primary and caucus states all at c-span.org/campaign 2012. >> this week on "the communicators," two critics of president obama's proposals reducing cyber threats against the u.s. marc rotenberg of the electronic privacy information center and larry clinton of the internet security alliance. this is part three of a monthlong series of cybersecurity. >> host: and this is week three of "the communicators"' series on cybersecurity. this week we're going to talk with some interest groups who have a stake in cybersecurity issues. first up, marc rotenberg of the electronic privacy information center, he's the executive director, and we want to talk to him about some only the privacy concerns from the white house cybersecurity proposals. when you look at what the white house has released over the past few months, earlier this summer, where do your concerns lie? >> guest: let me begin by saying i think a lot of what the white house has done is actually very good. there are a lot of different agencies

if we test them right, if we use our knowledge and experience the data here, we are going to have the best performing automobiles developed by the companies that have a great interest in saving us from oil and making sure that our country moves in the right direction. this is our new economy, one that i am proud to say we are doing across major cities across the country. but we in san francisco want to lead the way. we want to be the experimenters, the ones that push out this idea that we can do it in the bay area. if we do it here, it sends a trend across the country that they can do it elsewhere. get ourselves off of oil, into a more pollution-free society. i just thank all the people working together on this. it takes everybody working together to line up not only our educational institution with us, car manufacturers with the stimulus package fund from the government, and with our own local leaders to make sure we are doing the right thing. again, thank you very much for using your many years of leadership in the auto industry to bring forth new technology, new ways of doing thing

the report as 1800 data elements. as of today, we have 94,000 deaths in the system, of which 8.3% were due to child abuse and neglect. the gao report is right. we know that more children die of abuse and neglect than is reported. the cdc had funded a surveillance project in seven states. in an average year, michigan reported 16 child abuse death certificates. law enforcement's had 26 good child protective services had 40. when these sources were synthesized, it was 100 deaths a year. we compared them to the encamps data for the same year. i agree with the did gao findings on some of the reasons for the under reporting could death due to neglect are especially under reported. most definitely happen when caregivers egregiously failed to protect children from hazards. children dying in bathtubs, in house fires, left in cars on hot days. they're different definitions of abuse. what mississippi might call conn might call an accident. states have different system. when the deaths are not well investigated, we do not really k

now when the data collection -- nothing it free. there is some cost involved. congress has set the parameters for this data collection, but within that where we have discretion about how to do it, we are going to seek to, you know, make the most of the benefits of these data for the public, and only impose the burdens that are necessary to achieve those benefits, and we intend to make sure that the data have, you know, have integrity and will be useful in the end, and the issue of privacy that you mentioned of course is an important consideration, a consideration that has arisen in other data collections, and those issues are important and we intend to pay attention to them. >> thank you. no further questions. >> okay. >> thank you, mr. chairman, mr. ranking member, and thank you for being here today. one of the things i continually hear from my district down in south florida is small business owners and access to credit. everything worth reacting to can be worth overreacting to with dodd-frank. do you think that we need to have regular review of some of these r

you have to have the data. even then the data is questionable. because the people that develop these methods like bill sanders, william sanders is the father. says you can use it to identify the very top and the very bottom, but they are not useful to the middle. you cannot -- he has said you cannot identify individual teachers. dan, who you may know, is one the leading researchers who advocates for using test scores. >> guest: you are supposed to be the interviewers. is this a question? >> host: no, i'm responding to what you said. which was dan, who is the leading proponent of value-added assessment said you should not name names. in your books, you praise the "los angeles times" for naming names of teachers. >> guest: i think that public servants who are doing the public's work with taxpayers dollars ought to be named based on their performance with the caveat that everybody knows there's test measure. there's no denying if a teacher scores terribly on the test, we need to emphasize the viewers these are not simple tests that measure how the s

redesign the way the data center's work. they own fire in the grounds, there probably the biggest owner of fiber-optic cable in the world. and so, it costs less for google to do these things than other companies and a few years ago when some were saying they must spend $800 million a year just on the end with, that was wrong. they didn't tell me and exactly how much they paid it was clear to me that those estimates were way over the top. >> i want to get to a few things about where google is now. and i want to talk about china, which you bring it in the book, and china is an interesting -- we haven't yet said don't be evil witches of course google's unofficial corporate motto they came up with earlier i guess in sort of a little session they needed to come up with a kind of macho. >> that is an early googled engineer sitting in this meeting. getting kind that the meaning of the value and reading these things and people were saying the kind, help customers and things like that. he said we don't we just say don't be evil and ge

we have also moved the new data center without incident. the hub and mainframe is in production with no missed court calendars and data quality has improved. we will show you an example of that, when the sheriff's department talks about their programs. this is the look going forward. what we would like to do with the state department's corp. in the next six months is winning to increase the capacity for development and production. this is critical. we need to connect to the court system for testing to the hot by year's end. mike's team has been meeting with us weekly to make this happen. we also need to upgrade the sheriff's the jail management system, assist the district attorney's implementation of their subpoena module, connect the public defender's case to the hub, assess the police department of support requests for their crime data warehouse. in the next six months, this is our commitment to getting the job done. over the next 12 months, we will focus on cordray with the police department to provide data on domestic violence crimes to

and what you can see in the tennessee star data, what you can see in the tennessee star data is, if youry to predict the effect of being in the high court a kindergarten on adult earnings, using the discourse for the kids when they're leaving kindergarten, their task or benefits translate into increased earnings of $600 more per year, over their entire adult time period, despite the fact that the test score fadeout immediate like we do -- this doesn't guarantee that we will see the same sort of benefit in head start experiment, but it does make you realize that very rapid fadeout of test score impact is not necessarily inconsistent with long-term improvements and life changes. okay. the final thing that i want, is supposed that you were concerned, we don't know what this rapid fadeout means in head start experiment. so suppose you're concerned about this and you want to do something about it. what should we be focusing on to try and lock in the game's head start has made? what a lot of people focus on is the possibility that low-quality elementary schools might essentially squandered th

it depends on the type of data being captured. in a situation like this where you would want to know how many of the complaints or problems with the boy rather than something else, it does not track that. i would have to go into each of these cases and look at a notation that tells me that. we are asking for that type of data to be tracked in the new system, but it is not like that right now. i have to look of the notice of violation, and then be able to tell you that. having looked at 295 cases for the last calendar year, i will tell you a very small percentage of he completat complaints had o wideal with the boiler. the majority of complaints had to deal with this time clock not being sent properly or the thermostat being defective. if it was a single-source heating device from a gas or electrical compliance in a dwelling unit, it was usually something having to do with the heat-sensing device or maintenance. that was after going through each one of the cases individually, which we have been doing since last week. >> are they tra

just again, not right. >> host: we don't have the data in front of us. i think what i'm trying to do is find some common ground with you and say i think the charters can be intensely helpful if they would collaborate with public education, if they would stop with the kind of hedge fund driven business here to put you out of business we are better than you are. that might work in the business world and the educational world there should be collaboration supposing acre of public-school teachers would step back and say let us create the school and college a charter and take the toughest kids take the dropouts and the hardest to educate come back and help you. that isn't happening. you think that could ever happen? >> guest: if the union stopped fighting them at two and nailed in every way they can with ridiculous lawsuits, with little things such as the volunteers that the harlem success schools for the children zone, the teachers union said they can't use the lunchroom. little things, big things, anonymous phone calls about principles who actually try to su

the data isn't always helpful.s a very, very steep hill for an incumbent president, consumer confidence diving. people are looking around for signs of hope, and they don't see it. >> unless, john, unless he doesn't have an opponent who has a better plan. >> sometimes, it comes down to a choice, but often, it comes down to a referendum and people say you promised us something you didn't deliver, we're going to try something new, even if something new looks a little risky. people say, why can't these politicians get along? the people bear some reasonability to this. in 2010, they sent in a new republican majority who said the opposite. they sent in a majority that is the exact opposite. the government is the problem. this isn't just ugly politics. this is a competition with some big ideas. >> that's a good point, john. >> they're fighting about big things and the american people sent them here. >> which is why later on, fareed zakaria is going to make the case that a parliamentary system will work better. >> no, no,

i described the data quality program. he said there are 27 new governors coming in and we have to be done in 100 days. the insight and speed and urgency that he brought to this task to try to lay out a policy framework to avoid a really dumb policies you are describing -- this is not thoughtful public policy. it is standing in the way of anything we have described here as possible. we see a lot of it in k-12. if you look at the digital find agnow.com, you'll bunch of very specific almost redundant language. it is because each of the bullets under each of those elements go after stupid barriers that exist that try to stop learning at a county or district boundary. digital learn and now is an effort to create a policy framework that is thoughtful about this shift going forward. we very well may need another form for community colleges or extend that to k-14. it is the leadership of the two governors that are trying to create a framework for this revolution. that's why i think the work is so terribly important. we appreciate

redesign the way the data center's work. they own fire in the grounds, there probably the biggest owner of fiber-optic cable in the world. and so, it costs less for google to do these things than other companies and a few years ago when some were saying they must spend $800 million a year just on the end with, that was wrong. they didn't tell me and exactly how much they paid it was clear to me that those estimates were way over the top. >> i want to get to a few things about where google is now. and i want to talk about china, which you bring it in the book, and china is an interesting -- we haven't yet said don't be evil witches of course google's unofficial corporate motto they came up with earlier i guess in sort of a little session they needed to come up with a kind of macho. >> that is an early googled engineer sitting in this meeting. getting kind that the meaning of the value and reading these things and people were saying the kind, help customers and things like that. he said we don't we just say don't be evil and ge