PANEL: How Scaling Impacts Privacy (Devcon5)

[Applause] okay hi everyone the panel that we're doing right now is called house scaling and tax privacy I apologize my voice is a little rough today but I think I can manage my name is Liz Steiner I work at the authority but I'm just here to moderate a panel and we have four wonderful panelists to talk about this topic today and I'm going to let them introduce themselves and then we'll also in the more concerning I'm Josh Cincinnati I'm the executive director of the CGI Foundation a us-based biomimicry nonprofit focused on pricy financial transactions online permanently through through some cash I'm very excited talk about all this day this breezy said the camera a little rant about privacy as per this reduction so what yeah we had that so yeah why just scaling how do you think scaling in pets plants you might even get my it's discussion on yeah for me I think somebody's becoming apparently liquor apparent is that oftentimes the discussion around practices is at odds with what we can do the things that can able scalability and I sacrifice one in order to have the other the other the other thing that I'm like really fundamentally concerned about is actually more about scaling but scaling a public discussion around Whiteface is important because if we don't really socially agree that it is we're going to have a very hard time actually convincing people to integrate the things necessary to make this awesome hi I'm Ellie I am co-founder and chief scientist at off chain labs I'm a professor at Princeton and some time US government official so so for me they one of the big issues about scaling and privacy is what happens as we try to meet this vision of web 3 of scaling up this technology to the point where it's something that everybody uses all the time and I think there's a lot we can learn from the history of privacy on the web in web 2 which were not good people are very unhappy with how that worked out and it's not because there was not useful privacy technology there we knew how to do end-to-end description we knew how to encrypt it in place we knew how to do strong authentication we knew how to do a lot of things that somehow did not actually get done in practice with our name we had horrible problems with endpoint security we have all the problems with all of our data is sitting in corporate servers where who knows what happens to it the point being that having the technological tools there is not enough if we just redo everything that happened in web-one and web 2 but now with more stuff on a public ledger we're going to get a result that is no better and so we have to think beyond just building the technological tools we have to understand how to build a market and how to build a structure in the community that's going to behave differently and then others have before when the big money shows up hi so my name is Ian virus the shopping scientists see Josh and that's that's all for the University of Maryland loan first time cryptocurrency because I was worried about kind of aspects of it because if occurrence is basically Twitter for your bank account everything's help everybody that's not accepted for the same reasons that this said that's also an actually link I think to the scaling problem like a lot of people think that privacy it needs bonuses here they make sting a part of the things like the it's not spam or slow when it comes to you verified in on watching it actually makes things faster because Twitter for your bank account also distress stealing problem somewhat actually the Twitter have these hilarious problems when they first started is they were trying to broadcast huge amounts of data to everybody and so you solve the privacy problem by removing that data and they came closer verify you've also apparently solved scaling problem so I think these two things are linked they're actually thing where you can ask energies between them my name is Jacob Evos I'm a research at Technical University of Berlin and so why I'm here is I work on different things that the contacts are off changing and one of these things was to Socrates framework which is kind of programming language in school box that allows you to easily were boring to be specified programs that are provable in CK smarts and verify them conveniently on the blockchain what's really interesting to note here is like me indicated at CK's March than these things that are not often categorized as privacy tech we're like privacy was not when the motivation to do that it was actually scaling because the idea was if there is a way to verify a very complex station with a constant amount of computational effort and we can do that on the blockchain then we can scale very well so I do three that these things are not opposed at all and if the scaling things will actually brought me there and now it's also nice to to address privacy my take on privacy and security and how they relate is both are key challenges that need to be addressed for DES to actually become successful and make mainstream usage because I do think people will not become comfortable tweeting not only their financials transactions or web block chains but actually all their interaction with any IT Service over to watching so let's say we solve scaling in a year and drive asou tech does not keep up I don't think we will see mainstream adoption off maps right so for me it's a key enabler great two things everything else for your introductions until we've covered a few different things here well actually there's a lot just captured my parents for statements so um let's talk about how things are working now and why we should why we should care let's dig into that a little bit more so right now sometimes projects in space that it's more important sauce and technical problems to solve like you know use case problem and well they are privacy on top later or that yeah that let's get the scaling first and then we'll do the privacy part because you know there's not that much we shouldn't be too worried so let's talk a little bit about maybe what is what is the concern right now with the technology that's being used and how is it not good for privacy now and how this could scale up and it's beast think maker in the future so I mean the way it works now for anyone that is I'm going to use this material or many other people differently actually you have a procedure they use it now it is that they use working the system that effectively Peet's the transaction graph for anybody to see and it only takes like one short conversation with a watching analytics firm to realize that this system as it exists today it within all of these projects is measurably worse than the patchwork of offline cash transactions and various financial intermediaries that actually do have some degree of privacy guarantees and the way they the way works today so I think we can all hang universally agree that like these systems need to need to change in order for us to actually feel comfortable using them and I don't know day to day and a broad basis because otherwise it's gonna learn that happening is we're going to recreate this brand-new financial system but then people actually feel comfortable using gaps because they'll know that all the data can be traced and open for their competitors to see people won't actually engaging in any crypto currency transactions because they know that companies and governments will be surveilling earlier and that's like a very frightening thought so I like how you fix that Tomita to smart people to figure that out but I mean one of the ways to do is to figure out how many I need sanitize privacy on this chain right how do we get people to actually say that there's it's cheaper to do this various chains and socially how do we connect the really saw the disconnects that people have of like great right now we talked about this before Elizabeth McGlynn when people interact and buy something online at their credit card or use music rip to currency without any kind of protection the only they don't really think about how many people can actually surveil and understand the data as being sold on their behalf or observed and their behalf they can you imagine going into a stories and casually buy a newspaper the online equivalent their glyph 30 people in the room watching you do that and then taking notes and then recommending other newspapers really buy while your process of doing that and most people don't view it that way so if you have a big social challenge in terms of convincing people that that is I want to just sort of take issue with the approach of which you talked about that obviously weren't suggesting that we'll just build a thing first and then we will figure out privacy later because if you say that what you're essentially saying is we're going to design a system in a particular way and once we've invested a lot of engineering effort and we have a big installed base then we're going to fundamentally redesign it that's not gonna work if you don't start out with privacy of mine if you don't start out these thinking about what your privacy strategy is going to be and at the very least not engineering in a way that's inconsistent with that strategy you're gonna find yourself locked in to a design later that is very difficult from a privacy standpoint just apathetic continue to think that you don't need privacy if you're not doing pass so actually Josh you want me to give the first answer I think you expected me to give the standards feel like it though well we can't do it even than Republic it's Twitter for your bank account you you could stock whether ex girlfriend or boyfriend your business competitors will see what you're doing foreign governments use it for counterintelligence this is a rant I think for 20 minutes but it's tempting I think particularly for this audience to go well if I'm not doing the payments if I'm doing crypto kitty if I'm doing something a little bit different then there's no progress the problem because you just can't think in terms of what will be actually I have that problem too what about small contracts but it's not the case right there will be some privacy problem that stuff succeeds if your stuff is worth money someone will want to know what the data is I think you have a very good point where we're planning this event like that so actual contracts right you don't weep the fact there's a business relationship between two different parties every time you sign up is right but in a watch when you establish that kind of relationship it's public everybody and that's just not a thing companies or enterprises you want to do and so this goes back to your point then if we don't solve these privacy problems even for things that are quite money it's going to be a major barrier to adoption I think you made a very good point privacy needs to be early in the process of designing applications and also platforms but in light my perspective is that the people are also ill-equipped to attract these privacy concerns from a technological perspective right we build two theorem as platform that allows you to do something you could some people say also do a Bitcoin to large-screen degree but it was super complex so the theorem came along act as nice programming wall and smart contracts that added general opportunity to the field the problem is now seat has Walter forbid convert resolves to same issue for a theorem the theorem is much more generic and there will be no there will not be a privacy solution technical solution that fits all the new spaces that can be built on the platform so from my perspective it's key that we establish good schooling and equip the engineers that built II centralized applications with the tools they need to make their stuff privacy where and I'm not only talking about the traffic mechanisms or but I think they could bring you a long wave as community we need to think about how we enable people to build privacy where applications because now even if people think about it I think they can often just not pull it off another everybody has to see cache engineering - well one more sort of engineering you don't hear and that is obviously is a safety property and a safety property can be lost at any layer in the system so you can have the strongest possible platform underneath you this privacy preserving but you can still build an application on top of that the disgrace in the same way that the fact that you have HTTP to transmit data across the internet is an excellent way to protect the privacy of that data but there are plenty of services that use HTTPS and yet that terrible price because of the application level they give away everything to be gained at the lower levels so um so some one thing that you were saying is that looking to the future we're just we try to avoid the mistakes that we make with red one and one two and do one three different and so the answer is privacy by design that privacy tool many things like that giving developer to see opportunity and the knowledge to build in privacy by design and make it easy for them so do we just that's how we get web three we will really help run a scaling issue and not have to well I think if all of the engineers in the world could make a pact that we would only build privacy friendly technologies for the rest of our lives and be careful about who else we train to be an engineer than we'd be in good shape but given the incentive structure that we're being given how everything works we have to be concerned about what kind of market develops and avoiding the situation where there's a race to the bottom and that means we need to think about how about governance because there is a competitive dynamic that will drive everyone to try to collect and exploit more data than the other person in order to eke out a little bit more profit gain a little bit of a business advantage I don't think we can solve this by just all agreeing to not be evil and I don't think we can solve it by bill later one boring Aleya - which is going to make make insecure or London private applications on top of them impossible we need to have some way of governing our behavior that is stronger than just than just slogans and that means that we need to think about how we're going to operate we need to think about rules of the road for the sector and we can try to set those ourselves governments will try to set them for us sometimes governments will actually be helping us when they do that but they're gonna have to be rules because the market by itself leads to an outcome - surprise sir no I just wanted to add to that that I see the decentralized space also as a kind of opportunity because it just show buddy where which data is collected and that it needs protection whereas in the current system it's kind of hidden behind API or company walls and we do not really know because we cannot personally track what's happening behind that so why if by making this issue a very prominent issue by showing everybody hey here later stare into the attacks it will probably encourage this process of speeded up and one sort of follow-up question I have for you and on the idea of how governments to potentially provide you know gas or were additional incentive for us to consider privacy as a peasant sort of central technology systems do you think like things like GPR that kind of transform user data into a much bigger liability for companies and business entities do you think actions like that in other governments or raid agents event of their variants might serve to compel people to effectively change the behavior in most businesses approaching these systems I mean I think ultimately when businesses think of large holdings of data as a source of risk then they're then they're they're calculations will change GPR is one of the things that does that it's legal risk its reputational risk and other kinds of risks that can comfortable in day the government's will be tightening the screws because there's a perception probably correct that privacy has gotten worse over time and companies have become more careless with these super days data over time and if this sector is going to be better something's going to have to happen to me that happened the anybody in this sector or any other sector who gets big enough to matter it's big enough that your investors are happy you will be visible to to policymakers and regulators and if you are not behaving well you're going to find out in the same way that I seal markets FCU crackdowns we're going to see I think crackdowns upon privacy if people are really careless sorry because my question through I think how we make sure you get prized is actually good and not just the illusion of compliance right we've seen this in for payments and Victorian securities privacy needs that like lukla think work now but they really don't and so the anyone who's just has a public dataset you can think you're safe but then when you start working at sophisticated analysis when people have access to data it's also public then this stuff all breaks down and it seems like the reader reaction or yeah an anticipation of GDP are anything else is someone does something completely broken piece of cryptography that's like good enough to get past compliance until five years later where they find their cousin actually warning yeah so I wanna speak to this one because I was at one point the CTO at the Federal Trade Commission which is the US government agency that is most responsible for enforcing privacy laws and one of the things in that wall that drove me absolutely crazy was when companies made blatantly technically false statements about a privacy protection for example claiming that we only store the hash of the 32 bit IP address therefore we have no way to know what the IP address is or even people who claim we have the hash of the 10-digit phone number and therefore we but it is and saying that with a straight face I've seen your policymakers and so I think over time governments can get smarter they're hiring people who know how to call BS on those things I mean that's part of it right part of it is we have to create an environment where people just can't get away with bogus arguments about privacy right and we need to call them out if somebody's making a bogus privacy argument they should be called out by their peers publicly and it should affect their reputation governments will listen to that stuff but most people but more importantly prospective employers perspective investors and so on will listen as well it's sort of up to us to copy us on each other when somebody is saying something that it's really misleading to their customers so besides some the good nature of developers in the community saying that they're not going to personal data and they care about privacy also government regulations and these are and also just catching each other on each other what else can we do to incentivize privacy be effective well I guess this will go back to my previous might a little bit but you really have to make to get major the users want it like I honestly think that if you so there they're all these two external incentives and and but one thing that people should be intrinsically motivated about at least you know as it as an American coming from this sort of constitutional rights perspective like you should care about a Fourth Amendment rights care about us or right two fundamental privacy that is enshrined by these fundamental legal rights but I do agree that oftentimes like in practice I think the user has created bad trade-offs there is one service that is very nice and easy to use it doesn't care about privacy and then you can do see cash legal transaction which is very complex in comparison to paintball right so I think it's maybe that's not the best example but it's also like just clicking except on the website for something so we need to provide better trade also maybe regulatory force people to let better chance I think we also need to provide better attacks it'd be really good to demonstrate to people watching off so a good example this is didn't know which us is an app used it's for dinner that's we don't have a good banking system to transfer money that way and for a long time for much of people knew that bill had this default public each other literally was Twitter for your bank account was a web pages where he understood everything that one bit but we were all showing me win and then someone made an art project where they have this website with all this stuff on it and an era and all of a sudden people started caring about this and so similarly for watching I think we actually never go writing tools but advise your eyes look your something in there you can find out about you don't you music wish you have made this a sec right and until we do that it's just a hypothetical oh I can tell you that my arrow is work well I can tell you that this vessel back up it it's not really I mean another thing we can do is give users some hope I think most users most of the time when they use technology just said this fatalistic attitude whatever I do my date is going to go everywhere right and so it doesn't really matter what I choose people don't really believe that if they choose product Abril and product B they'll be in them they'll be in a significantly better position with respect to privacy and in that kind of environment if there is a product a stack if there is some environment that's available to people that actually does offer better privacy and we talked about the idea that it is possible to do this because I think in the in the legacy tech space their narrative that you hear is that you know we really don't want to have all your data but we just need it to run our business it's just a thing that's necessary to make the sector go and I don't think that's true but to the extent that this is the way that our community talks about these things we will make it true so we need to work hard I think to try to give users hope and to try to talk about holding ourselves to a higher standard that we can solve these problems and we don't have to just collect everything in order to make our businesses governor we're not talking about do you trust Google with your data or Facebook with your data but it's actually do you trust some part of your data or is it public everybody it's not just this trade off maybe give us some better ability you can at least - thank you only trust like or maybe have a follow-up question do you think there's really that much of a choice you haven't back three or is it rather there is an implementation that gives you privacy property is technically where you have zero for me it's muscle binary in that space because the longer company that has data and can act that's just me or be yeah be good with the data well in the in the legacy tech space the a lot of privacy action or trouble happens at higher levels of the staff right so anything that is a real risk here in this space as well that at higher levels of the stack you might see more data collection you know from a fundamental technology standpoint you an answer to your question I think there are pros and compared to what came before also this is a do-over on some of the structures and approaches that happened before and maybe we're a little bit wiser about how to do it so speaking of being Liza that's a little bit about the technology and people actually speaking of them let's talk a little bit more about the technology and incentives that can be combined perhaps like we were talking earlier about scalability and privacy can be done at the same time with technology I think it's good to explore that kind of thing too it's an extra opportunity for incentive that perhaps is not better and what beautiful and stuff so yeah do you want to talk about some of the abilities that we can do with the things that we can capability that we have when were doing an assessment not so my usual thing at work with this newer knowledge of proofs and we've seen a massive amount of new excitement of what we can do and it's quite interesting but I think we're sort of people getting distracted by they'll get just the new exciting little pieces of it or not the like general pattern of use the tools here's the techniques so a common thing people think of is without new piece of photography to get there then a bunch of these things to do this this is the existing techniques for doing like it helps me catch some air pollution well or as well and then also that like people miss I need to get hung up I'm like oh we're in the account about bitcoins at UT Excel bottom different none of the techniques over there or over here right and it's actually these are all pretty you use most of these developments on anything so it seems that in fact the incentives of these face right now I'm getting very excited about what's my priority subscribe there were actually many shows like what health inside your car it's the biggest thing there are also things that you can do in terms of design that don't rely on the absolutely newest Chinese most complicated cryptography so you know looking at what we're doing at opportune laughs we've done a lot of work to try to move state of states and contracts off chain we did that originally way back in the beginning motivated by scalability but when you move stuff off chain you when you reduce the footprint on chain you also have a significant privacy benefits so you don't need to use the highest tech stuff there stuff you can do with in this sort of traditional engineering frameworks and there is a synergy between efficiency and privacy in a space where the public more public on chain say is more expensive than what you can do privately that's not to say that you should never use the the high tech stuff or that we would never use it but but you can get a lot without needing to you didn't use that stuff I was only in three and I would like to ask a question to the other panelists because it was my initial question was like where unit is like how do we have privacy while you're scaling for me oftentimes the answer to privacy and scaling poisons the same direction as least because it's like reducing block chains to what they're actually that and not doing as much as possible on the blockchain and if you remove computation steps and then like to to verify the montane you can use different techniques that can be simple integrity checks with hashes it can be complex mechanisms like stock starts whatsoever it doesn't matter but the key aspect is we move stuff away from the clock chain keep as much of the information off that public lecture in the first place and then there's much less we need to bother about and at the same time we push this loan on the blockchain so it's naturally gonna not scared better but for the same amount of utility require less computation to be done mantra in the first place so do you think there is a synergy between privacy and scalability or a trade-off so I I think I'm the lone panelist who who said that there was a trade-off and I was at least technical in years maybe I wrong but I actually like I think there is a trade-off present that because it creates it creates a degree of user complexity for one thing when you start offloading stuff into other layers and doing things if you roll up something like but you have to figure out then how you actually engage that interaction layer on the user side then the other thing that I I think it creates is you know it does require like shiny or newer photography that has potentially like unknown unknowns about what way are built really play something hello where there is Vista to be like a weakling trustless every cursive you know the trilogy privacy but there's so much that's not known about it and you said to understand those trade-offs when when approaching this doesn't say that there isn't still some alignment on like yeah there being a mutual advantage to having less state I'm getting blockchain and more of it sort of private and an ulcer that just creates new complexities that have to deal with like Josh is saying at the technical level that seems to be true which is an interesting switch because usually in traditional centralized systems when you added cryptography maybe it's slower and so you make a justified vs. here it's reflective of your new feed data an assistant that's all the burden that makes a pass code but the social complexity of the spike shedding of rich photography do we use or hopefully if it is not common sense I'll write this whole debate up which weights assumptions which elliptic curves right which you know do we believe in falsify about subject you've seen these things on Twitter can slow things down it's very impractical from a socialist perspective developers well in the interests of scaling the debate and scaling the system we're gonna put those off for five years and what it's going to ignore and that is tempting but if we do that we will find out five years that we can't have privacy and we'll be doing the pixel there's a role for Standardization here as well and doing with the issue that ian is talking about that you know the sort of prank mutation of the complexity you see this in other areas as well where those things that are similar across products can be standardized while allowing people still compete in the areas where there's more innovation we saw that you seen that in cryptography for example right there are standard ways to do many of the older cryptographic primitives it's understood how to do them like key size to me how to generate this or that even formats and so on are standardized and that's all useful for a bunch of reasons one first because it allows more vetting by the community of particular approaches but also because it helps with the emergence of open source implementations so you don't have people reemployment in the same stuff so how do you think that interacts with what seems to be emerging there where we have companies that are basically marketing not product but literally go away we don't see where knowledge encryption proof that like trying to push these things so you're gonna that could disappear in like five years that just effective practice is a very nascent new system is this be a problem of like you're not gonna standardize this is one guy says they want to standardize on they and someone saying something completely different but if we have a strong cryptography stood on a you'll remember I mean this standardization works where there is enough agreement on where how some part of the system would work you know you can come up with standards for state verification of zero knowledge proof maybe without needing to have a standard for how generate the proof for what sir how you build the circuit that you're proving and so on so you know and similarly in other systems there are parts that are standardized and the standards can be pluggable where people can plug in their own secret sauce in the areas where there is a lot of innovation and where it's a little bit less so I'm not an expert on the history of cryptography but I do think that there was kind of similar things taking place with early encryption right where it was in private companies and they selected some parameters and was like old doc magic and over time these things where battle-tested became established were republished and more open source wait so maybe we're just on track but I do see what they're getting at that this could be an issue the reason I mentioned it seems to be middle that's exactly right or something like this and such but a lot of it was all done by RSA for the basic encryption and then Netscape for plsss all the time but it seems like in those cases there was one comment expertise and now in this cases democratized which is good but there's now more head fighting and that sounds different but that may be because I'm just sitting I mean if you look at cryptographic standards some of the most popular standards like the hash function standards for example or a yes we're done through government organized processes although they were open competitions and that was a way of organizing the community and you had a lot of it was community work and so lots of lots of diversity different opinions about how to do that but yeah a strong standard emerged and one thing that got trust out with that process obviously governments can't run these things badly and the result of that can be can be worse than nothing as we don't know but there are good examples good the good existence proofs that this can be done well and you can get to standards even for things like Cyprus where there's a lot of a lot of different ideas about how to do it and I guess standardization will help with making easier tools for privacy medicine so yeah is there anything else that you won't want to say before we open up through our questions okay just are there any questions from the audience for it's like no I think some rest could you talk about the role of crypto currency exchanges in this debate and pass in your answers could you debate amongst yourselves or talk about what you'd like to see press of what you think will happen so the question was how can exchanges in the future scale in the taxi market is going to be very creepy but with exchanges my major concern is the last exchanges maybe for a bit but maybe users and users behavior and it goes back to privacy and Trust and like the new responsibilities of users as well because if it looks statistics how many people actually keep private keys themselves and how many people trust exchanges with it it's kind of worrying to think what I don't think we can think that these people will take responsibility for privacy yeah if if we all view data as a liability then exchanges are effectively like centralized channels are legally mandated nuclear-waste ups the like they they're worth everyone's data that goes and they're legally required to write like and I don't see that changing to be honest so I my preferred outcome that I don't think it's actually likely to be honest is that at least Senate not assume is that people figure out other ways to exchange against these systems that do actually preserve privacy without having to go through the nuclear waste stuff any other questions like yeah questions I guess in terms of the pattern of saying I see repeat in America but our neck the most users are or will be in the most in Asia China independent of what does your experience which in that sense is there anything like we can see if there's developments there or or or certain thing about me so I have last month I was in Tokyo on a panel with some people from campus a which is the Japanese financial services administration I believe and they regulate cryptocurrency and that - in Japan and that was actually interesting because there was an appreciation of the privacy wasn't English should be aware of they wanted to do all of its first but that actually wasn't think they were concerned about they acknowledged that needed genuinely to have privacy of these things they're really interested in male is that the regulatory environment janma say is responsible both for consumer protection and privacy and for enforcing like act money water and stuff versus in the u.s. the guys who you money laundering is Treasury Department Vinson and when you talk to them about privacy they basically go yeah you sound right but that's not our problem so why are you getting our way so I think there may be some better hope in parts of Asia in other places well I'm not gonna comment further on that anybody else I mean there's also going to be the difference between - ether companies versus fantasy the individuals - right there any other thing that but even in the most permissive hostile government you can think of those governments have enemies leader other governments they don't let those other governments seeing everything but just imagine what would happen if you go looking everybody who worked in the security service of a company go oh yeah that guy there is having trouble paying his rent Hugh Trevor I would be there going all over that they give us all the classified documents it has right so there is an incentive even in regimes at the hospital privacy what these have security from other governments they don't apply so it's kind of any funny to see how it's been at a place out in the questions well then I'm gonna ask a fun question all right so so if you could so we like before the internet do or have some idea of what life was like before the Internet here we go again load three what is looks like maybe think about what would be your coolest training for three outcomes and what's life gonna be like if everything works out as we want for some aspect of that I just think in the future you're gonna do or the opposite like what was it like back in the end before we like you know you're the person who I date with to find out that we did you everywhere you've been and everything you Bob just googled address right yeah it's some poll but I just hope that this could be like like computing infrastructure as powerful with better privacy with less dependency because as it turns out information technologies like a key Trier nowadays and that access and solid infrastructure would be nice one more daily life level I think there's still huge opportunities for integrating services reducing vanderhoff in all these kind of things that just make your daily life better but mail see on Spencer over the next 10 years my biggest hope is that we learn from what has happened before us we have a chance planet to before I do over to rebuild the technology stack that people rely on and we have a real opportunity to do a better job it is probably not as easy as we think to do that people who came before us faced a lot of trade-offs they made a lot of decisions most of them still seem like the decisions in hindsight and yet they ended up in a place that has some real disadvantages but I hope that we do better this time and I think one of the keys to doing that is to get well two keys to doing that at first to really pay attention to what happened before and learn from it and to not to give ourselves too much credit for being able to see ahead for being more altruistic than we actually will turn out to be yeah if I couldn't describe like my ideal view for what the end result of this movement might be like I actually I think it'd be more this in my in my day-to-day I would like to have the device on me that is not tracking me that when I go to buy a book it doesn't it the rme on it I paid for it without having to enter in my email or user account and without anyone knowing who I am buying that information I just want to live in a world where it kind of emulates like what it was like when you went into a bookstore and they actually feel like that's a return to form and that we could build that under formal gaining all the convenience that we have in today's society so you're saying you want to look like you're saying you wanted to look like what it was before they lost the other half so we want the mobile device in our pocket that's not acting like the mobile device trapping us in a pocket that we separate out and we once have been with a cache but that's digital all right and here's our last comment that 617 seconds over time [Applause]