[00:32.510 --> 00:42.210]  So my instructions here tell me that I should read announcement, and I don't know if there are any announcements, but consider them read.
[00:45.040 --> 00:47.060]  So my name is James Vasile.
[00:47.260 --> 00:58.880]  I work at the Software Freedom Law Center, which is a pro bono, non-profit law firm that does work for people who make free software.
[00:59.060 --> 01:03.600]  Basically, we do legal work for hackers for free, because we love you.
[01:04.320 --> 01:13.580]  And because we do legal work for hackers, we're involved in all the issues that hackers think about.
[01:13.600 --> 01:16.460]  And we consider ourselves hackers and tinkerers ourselves.
[01:16.660 --> 01:18.500]  I mean, we hack legal systems.
[01:18.500 --> 01:23.200]  We all, of course, write code, and some of us are electronics geeks as well.
[01:23.360 --> 01:30.500]  But we think about a lot of the same issues that people who come to a conference like HOPE think about.
[01:30.500 --> 01:37.040]  And one of those issues is freedom, privacy, anonymity in the online space.
[01:37.040 --> 01:43.720]  When you're acting online, data that is gathered about you and what happens with that data.
[01:43.720 --> 01:47.500]  And we've been thinking about this issue for a while now.
[01:48.080 --> 02:03.660]  My boss, Eben Moglin, gave a talk at the Internet Society in February, where he talked about something called the Freedom Box, which is an idea that he had of an appliance, a little network appliance, a small computer.
[02:03.740 --> 02:09.360]  In his mind, it's a plug computer, like a Pogo plug or a Shiva plug or something, that you plug into the wall.
[02:09.360 --> 02:13.180]  And it intermediates between you and the Internet.
[02:13.580 --> 02:16.020]  Sometimes it impersonates you on the Internet.
[02:16.280 --> 02:32.360]  It does business for you on the network and affects your data as it passes between you and the outside world, all with the intention of protecting your control over your own data, over your own privacy, and over your own anonymity.
[02:32.360 --> 02:35.860]  So, this talk is about the Freedom Box.
[02:36.000 --> 02:42.340]  I'm going to talk to you a little bit about... very little bit about why this is a good time for a device like this.
[02:42.980 --> 02:46.460]  I don't think that's a big case to make at a place like HOPE.
[02:48.200 --> 02:52.720]  But I'm also going to talk a little bit about the specifics of what we think the box might contain.
[02:52.860 --> 02:57.500]  Now, his talk was sort of the first public throw this idea out there.
[02:57.980 --> 03:12.940]  He talked about the history of the architecture of the Internet, of the replacement of peer computers with a server client architecture, and clients at the edge of the network without a lot of power, and not doing a lot of computing, just consuming services,
[03:13.240 --> 03:18.680]  and how that process of centralizing processing also centralized data.
[03:19.060 --> 03:23.540]  And centralizing the data was what led to a lot of the privacy concerns that people have today.
[03:23.740 --> 03:28.940]  So, he gave a historical background and said, one way we can solve this is with something called the Freedom Box.
[03:28.940 --> 03:31.340]  And he didn't give a lot of detail about what that box might contain.
[03:31.920 --> 03:37.720]  I'm also not going to give a lot of detail about what that box might contain, but I'm going to try to take it one step forward.
[03:37.900 --> 03:42.180]  We are at a point right now where we're just trying to feel out what it is that this is going to do.
[03:42.180 --> 03:57.680]  And hopefully, this talk, especially in this audience, will serve as an invitation to some of you to get involved in a project like this, to help us figure out how to build this thing, what to put on it, and how it's actually going to operate.
[03:57.800 --> 04:04.200]  We think most of the problems are solved problems, and the other problems are solvable problems in a short term.
[04:04.200 --> 04:10.500]  So, that's where we're going with this conversation in the next 50 minutes.
[04:11.540 --> 04:19.000]  So, the need for privacy, we would argue, is greater today.
[04:19.000 --> 04:22.080]  The need for a device like this is greater today than it has been in the past.
[04:22.080 --> 04:30.320]  And one of the reasons why that happens is because we walk around in the world, and we have certain beliefs about our behavior.
[04:30.660 --> 04:39.840]  When we act in the world, and we interact with each other, and we interact with the world around us, we engage in behavior in public that we think is suitable to be in public.
[04:40.140 --> 04:45.400]  And in private, we engage in behavior that we think is suitable to be engaged in, in private.
[04:46.680 --> 04:51.720]  What we're increasingly learning is that on the network, there is less and less private.
[04:51.720 --> 04:53.500]  There is less and less private space.
[04:53.660 --> 05:00.600]  So more and more, people are engaging in activity that they would consider to be private on public computers.
[05:00.940 --> 05:03.220]  Computers that they don't own, computers that they don't control.
[05:03.580 --> 05:07.680]  And in that public space, their activity can be watched.
[05:07.880 --> 05:09.520]  And it happens in public, right?
[05:09.720 --> 05:13.420]  Things you do in public, people watching, that doesn't really raise a lot of eyebrows.
[05:13.420 --> 05:20.200]  But if we're going to increasingly do private things in these places that are observed, right?
[05:20.420 --> 05:26.260]  On Google's servers, on your Facebook page, even over your email.
[05:26.500 --> 05:35.560]  If you're going to increasingly do these things in these public spaces, you also have to realize that these public spaces are increasingly monitored.
[05:35.940 --> 05:42.840]  It used to be that you could do something stupid in a public space and probably no one would ever know because no one was ever watching.
[05:42.840 --> 05:44.660]  Certainly no one was ever recording.
[05:44.940 --> 05:47.780]  But now we know that everything we do is remembered forever.
[05:48.280 --> 05:51.840]  Everything we do now online sits... there's an entry in a log.
[05:52.180 --> 05:54.400]  And that log never gets deleted.
[05:55.100 --> 05:56.100]  It doesn't have to.
[05:56.280 --> 05:59.060]  It only gets deleted when its useful life has disappeared.
[05:59.260 --> 06:01.260]  And who knows what the useful life of data is these days.
[06:01.480 --> 06:02.520]  It might be six months.
[06:02.560 --> 06:03.380]  It might be six years.
[06:03.500 --> 06:05.440]  It might be 20 years, depending on the data.
[06:05.440 --> 06:09.760]  So, more and more of what we do online is recorded.
[06:10.080 --> 06:18.700]  And we live now, to the extent that we live in our digital world, in what might be described as a panopticon.
[06:18.860 --> 06:24.720]  Everything you do, somebody is watching, somebody is recording, because you are doing it on somebody else's server.
[06:27.380 --> 06:38.640]  We are at a point now where the density of information available to people who want to know about you is so high that they can infer all the little bits about you they don't know.
[06:38.640 --> 06:45.880]  And if you are a person who opts out of giving them your data, they can infer that you're the other guy who didn't give them the data.
[06:46.120 --> 06:48.300]  We have a guy at the Sartre Freedom Law Center.
[06:48.500 --> 06:49.660]  His name is Bradley Kuhn.
[06:49.960 --> 06:55.220]  And he has never, ever given anyone his social security number.
[06:55.360 --> 06:58.200]  Like, this is a number that he knows and his mother knows.
[06:58.320 --> 07:00.100]  And nobody else on the planet knows this number.
[07:01.220 --> 07:08.040]  When he signed up to work with us, he whispered the social security number to, you know, the social security administration.
[07:08.340 --> 07:10.440]  And nobody else in the office has access to that.
[07:11.220 --> 07:16.100]  Our health insurer, which requires everyone to give social security numbers, he refused to give it to them.
[07:16.380 --> 07:20.000]  I mean, he has kept this number as locked down as you can keep it locked down.
[07:20.340 --> 07:25.320]  And about a year ago, Evan said to him, well, you know, you've really lost the battle.
[07:25.520 --> 07:26.920]  Google knows your social security number.
[07:26.980 --> 07:27.940]  And he said, how is that possible?
[07:27.940 --> 07:30.060]  I couldn't, I've never given them this number.
[07:31.060 --> 07:39.740]  And Evan quite rightly pointed out that Google knows the social security number of everybody else who was born in Baltimore the year he was born.
[07:40.220 --> 07:42.520]  His social is the other one.
[07:43.740 --> 07:44.300]  Right?
[07:45.860 --> 07:54.720]  The gaps between data have grown so small that we can peer at them and see the outline of the missing piece and know what is in there.
[07:54.720 --> 08:09.880]  So those of you who are sitting in this room, who are opting out of giving your data to centralized servers, to companies that are going to data mine everything that passes across the wire, you think you might be getting a certain amount of privacy.
[08:10.280 --> 08:11.640]  And you probably are.
[08:11.980 --> 08:17.860]  But you are getting less than you might otherwise get if all your friends weren't also giving up their data.
[08:17.860 --> 08:19.600]  So that's where we are now.
[08:19.780 --> 08:26.700]  We are in a world where even the people who want to hide, who want to opt out, increasingly have difficulty doing so.
[08:27.000 --> 08:36.060]  And the reason why we live in that world is because the data logging and the data processing and the data mining has gotten all pervasive.
[08:36.060 --> 08:43.520]  And as more information moves online, more activity moves online, the amount of data available is just going to increase.
[08:43.880 --> 08:51.580]  And when we reach the point where we are living more of our lives online, we are going to have less privacy even than we do now.
[08:51.680 --> 08:56.140]  Now, some people don't worry about how much privacy they have.
[08:56.240 --> 08:57.540]  Some people don't think about those things.
[08:57.640 --> 09:02.180]  They say, well, I never really give private information to these companies.
[09:02.180 --> 09:04.860]  I never tell them anything that I think is so bad.
[09:05.140 --> 09:16.320]  But it turns out that from relatively innocent data, we can find things about you that you would consider to be much more private than the innocent information you gave up.
[09:16.760 --> 09:31.820]  For example, there are students at MIT who, by scraping Facebook page data, were able to understand or make extremely accurate guesses as to the sexual orientation of other people on Facebook.
[09:31.820 --> 09:33.620]  Just by looking at your social network.
[09:34.920 --> 09:41.840]  And to some people who might consider that private information, they never told Facebook their sexual orientation.
[09:42.180 --> 09:45.860]  They never announced this to anyone in the Facebook community.
[09:46.100 --> 09:48.840]  Maybe not even to anyone in their offline world.
[09:49.220 --> 09:56.700]  To suddenly have that information discoverable, inferable from all the other data, that might be a scary topic.
[09:56.840 --> 09:57.860]  That might be a scary proposition.
[09:57.860 --> 10:02.460]  So we live in a world where the amount of data about you is increasing.
[10:03.400 --> 10:10.680]  And the things we think of as innocent information are not so innocent when you aggregate them.
[10:10.920 --> 10:16.200]  And when you take all the data you can get about you from Facebook and add it to all the data you can get about you from Google.
[10:16.400 --> 10:23.440]  And add it to all the data you can get about you from online databases and mass marketing databases and government records.
[10:23.440 --> 10:27.180]  We can paint a fairly convincing picture of who you are, right?
[10:27.300 --> 10:34.180]  You sprinkle in a little bit of demographic information, and suddenly we have a picture of who you are as a person.
[10:34.360 --> 10:38.640]  I mean, you know, Google knows more about you than anyone else on the planet.
[10:38.840 --> 10:40.960]  Google knows if you're cheating on your spouse.
[10:42.000 --> 10:43.900]  Google knows your HIV status.
[10:44.340 --> 10:47.580]  I mean, Google knows if you're failing out of school.
[10:47.880 --> 10:50.060]  Google knows any secret you might have.
[10:50.800 --> 10:52.020]  Google has.
[10:52.020 --> 10:52.160]  Google has.
[10:52.660 --> 10:55.840]  So that's the landscape.
[10:56.100 --> 11:01.000]  The landscape is privacy is harder to achieve now than it ever has been.
[11:01.440 --> 11:10.900]  The things that we think of as things we can do in public, and it doesn't matter if anyone's watching, it turns out that those are betraying our privacy as well.
[11:15.780 --> 11:16.660]  So there we are.
[11:16.820 --> 11:17.460]  That's where we live.
[11:17.640 --> 11:21.260]  We live in that world where privacy is increasingly difficult to get.
[11:21.260 --> 11:32.000]  We also live in a world where people claim to value their privacy very highly, but are willing to sell it very cheaply when actually engaging with online services, right?
[11:32.160 --> 11:35.240]  Everyone in here would think of themselves as a privacy nut.
[11:35.900 --> 11:38.460]  But I bet you a lot of you guys are using Gmail.
[11:38.940 --> 11:41.220]  And I bet you a lot of you guys are using Facebook.
[11:41.500 --> 11:46.420]  So we all value our privacy very highly in this notional way.
[11:46.420 --> 11:50.860]  But if you look at behavior, we're not putting a high price on privacy.
[11:51.080 --> 12:00.000]  What we're doing is we're wishing things were better, and we're selling the data sort of semi-reluctantly, but not thinking about it very hard.
[12:00.000 --> 12:21.120]  And if we, the people in this room, who are the most likely people in the world to care about this issue, are willing to take that step, it's very difficult for me standing here to imagine how we're going to convince the kinds of people who wouldn't be in this room to value their privacy in actual behavioral practice.
[12:24.450 --> 12:29.490]  So we have to realize we can't protect all those people.
[12:30.530 --> 12:34.690]  Most privacy discussions start from the point of view of, well, how can we change society?
[12:34.870 --> 12:39.910]  How can we move the privacy environment so that we all get more privacy?
[12:40.190 --> 12:43.930]  And I would argue that that's the wrong approach.
[12:44.190 --> 12:50.330]  I would argue that at this point, that battle is lost, at least for now.
[12:50.330 --> 12:59.410]  What we should be asking is, how can we get privacy for ourselves in an environment that is extremely hostile to our own notions of privacy and anonymity?
[12:59.650 --> 13:08.890]  And so, when I think about the Freedom Box, I think of it as a device that is not necessarily going to sweep the world off of Facebook.
[13:09.150 --> 13:15.970]  I think of it as a device that is going to allow you to decide, this is control of my data.
[13:15.970 --> 13:18.110]  And it won't necessarily control anyone else's data.
[13:18.250 --> 13:19.910]  It won't necessarily convince anyone else to do anything.
[13:20.110 --> 13:23.870]  But this box has that sort of limited scope, at least in my mind.
[13:23.970 --> 13:26.630]  And, you know, who knows what it will actually look like in the end.
[13:26.810 --> 13:29.170]  But that's one thing that we have to consider.
[13:31.510 --> 13:34.210]  Okay, so that's my little spiel on privacy.
[13:34.310 --> 13:35.950]  That's sort of where I'm coming from on privacy.
[13:36.190 --> 13:37.750]  We live in an increasingly hostile world.
[13:37.930 --> 13:39.310]  What can we do to protect ourselves?
[13:39.310 --> 13:41.670]  Everyone else is lost.
[13:41.950 --> 13:42.970]  We can't save them.
[13:43.350 --> 13:43.490]  Run.
[13:44.750 --> 13:47.770]  So that's... that's the privacy landscape.
[13:48.050 --> 13:54.470]  We need to now talk about what does the box actually do to exist in that landscape.
[13:54.750 --> 14:00.970]  And we've got a guy at the Law Center, his name is Ian Sullivan, who loves buying gadgets.
[14:00.970 --> 14:08.810]  And every time, you know, Engadget covers some small Linux ARM box, it ends up on his desk three days later.
[14:08.950 --> 14:12.510]  I'm not quite sure how he manages to get to the front of the line on that, but he does.
[14:12.810 --> 14:15.010]  And so we've got a bunch of these little gadgets.
[14:15.350 --> 14:21.310]  And we envision the Freedom Box as a plug computer with an ARM chip.
[14:21.470 --> 14:22.990]  That's the target architecture for now.
[14:23.130 --> 14:24.250]  You plug it into the wall.
[14:24.490 --> 14:25.670]  It is your router.
[14:25.870 --> 14:27.010]  It is your firewall.
[14:27.010 --> 14:29.890]  It decides what packets come in and what packets go out.
[14:30.030 --> 14:32.790]  But more importantly, it implements policy.
[14:32.950 --> 14:35.590]  As in, what packets are we going to encrypt?
[14:35.930 --> 14:39.150]  What packets are we going to send to certain people?
[14:39.330 --> 14:42.610]  If somebody requests information about you, who are we going to give it...
[14:42.610 --> 14:43.810]  Are we going to give it to them?
[14:43.990 --> 14:45.590]  How much data are we going to give them?
[14:45.670 --> 14:46.310]  And in what form?
[14:46.690 --> 14:49.230]  So the box is not just a router.
[14:49.430 --> 14:53.490]  It's a router that is designed to have your privacy in mind.
[14:53.490 --> 14:57.070]  It is also a Tor exit node.
[14:58.490 --> 15:01.250]  It is also a VPN box, right?
[15:01.430 --> 15:10.250]  So people who happen to live in countries that stick them behind giant firewalls might be able to access the free Internet by talking to your computer first.
[15:10.530 --> 15:21.210]  So this is a device not just for your freedom, not just for your privacy, but also to enhance the privacy and freedom environments of other people, to give anonymity to other people.
[15:21.210 --> 15:26.450]  And that's why we think it's important that it includes something like a Tor exit node, like VPN software.
[15:27.910 --> 15:32.950]  And the software, we envision a stack, a freedom stack.
[15:33.250 --> 15:43.630]  At the bottom is this sort of networking stuff, is the sort of thing that will just allow you to communicate with other people without worrying as much about privacy as you do now.
[15:44.130 --> 15:51.810]  Above that level is some sort of system layer, something that is going to send data using these protocols.
[15:52.010 --> 16:10.010]  And that's going to start to look like an encrypted database, an authentication mechanism for deciding who can access stuff, for deciding that you are you, for being able to say, I am you, for being able to authenticate to other computers as your agent on the net.
[16:10.010 --> 16:28.070]  And this is infrastructure, generalizable infrastructure, that we would hope would get used for building applications that allow people to implement some of the services that we are currently using that steal our privacy, instead as services that only log to this box,
[16:28.170 --> 16:29.550]  or better yet, don't log at all.
[16:29.550 --> 16:41.130]  And so, by using these set of services, we are hoping you can build a set of applications to replace things like Twitter, and Facebook, and Flickr.
[16:41.330 --> 16:59.250]  And so, if you want to use the private versions, the privacy enhanced versions of these services, this box will serve the data for you, and most importantly, hold it, and not give it to some other company that is going to then data mine it, and sell it to other people.
[16:59.430 --> 17:04.010]  And the idea is that your data is safe because it never leaves your box.
[17:04.370 --> 17:09.430]  And your data is there, locked up, for only you to decide what to do with.
[17:12.810 --> 17:15.390]  So that's the box from top to bottom.
[17:15.530 --> 17:16.850]  It's a simple stack.
[17:17.290 --> 17:20.090]  It's a simple device that plugs into the wall.
[17:20.690 --> 17:31.450]  And above the application layer, you've got a bunch of protocols for moving data between applications, for moving data between boxes.
[17:31.630 --> 17:33.370]  And this is how the boxes get social, right?
[17:33.470 --> 17:46.870]  This is how, instead of centralizing your social networking in one website, your box is going to talk to other websites, to other boxes, and decide, okay, I've got an RSS feed of photos of my last party.
[17:46.990 --> 17:50.150]  You can have it because I know you, and I trust you, and you're in my network.
[17:50.390 --> 17:53.210]  But I'm only going to give it to you because those photos are private.
[17:53.210 --> 17:56.070]  Or it can decide, this is my blog.
[17:56.390 --> 17:57.890]  Anyone's allowed to read my blog.
[17:58.050 --> 17:59.710]  And it will serve that for you as well.
[18:00.050 --> 18:06.450]  So the simple idea of having user control over their own device is not a new idea.
[18:06.710 --> 18:14.530]  The new idea is putting it in an appliance that you plug into the wall and don't have to do a whole lot of configuration on.
[18:14.690 --> 18:17.810]  The new idea is to make it as turnkey as possible.
[18:17.950 --> 18:24.710]  And we're finally reaching the point where we don't even have to tell people, download this software and install it and configure it.
[18:24.830 --> 18:29.630]  We've reached the point where all we have to do is say to people, buy this piece of plastic.
[18:29.950 --> 18:31.490]  Plug it into the wall.
[18:32.170 --> 18:33.010]  You're done.
[18:33.850 --> 18:35.190]  Maybe tell it your name.
[18:35.530 --> 18:38.830]  And then everything else is just standard application software.
[18:39.170 --> 18:41.590]  All this software already exists.
[18:42.210 --> 18:44.190]  We have Tor software.
[18:44.430 --> 18:47.610]  We have people trying to write the system level.
[18:47.750 --> 18:50.950]  And we have people hacking on the application layer.
[18:50.950 --> 18:57.270]  So that's the basic idea behind the three pieces of the box.
[19:01.450 --> 19:08.890]  So the hard part with doing something like this is that there are certain parts of the software that don't already exist, right?
[19:08.990 --> 19:13.370]  Like, I mean, we want this box to pull down torrents in an encrypted manner.
[19:13.530 --> 19:14.790]  That's really easy to do.
[19:14.950 --> 19:16.610]  We have software that will already do that.
[19:16.610 --> 19:21.970]  What we're missing is the stuff that allows you to do the authentication.
[19:23.830 --> 19:33.590]  Decentralized authentication turns out to be really hard, especially if you want to do it in such a way that if your box dies, you don't lose any of your data, right?
[19:33.630 --> 19:41.310]  You want your box to back up all the pieces of data to all your friends' boxes in an encrypted manner so that they can't read it.
[19:41.390 --> 19:44.570]  But if you ever need to, you can say to them, hey, give me my data back.
[19:45.150 --> 19:47.110]  This is not the easiest thing to do.
[19:47.230 --> 19:55.250]  It's hard to figure out what is the thing the user will hold that will allow them to say, I am me, and you should give me my data.
[19:55.430 --> 19:56.170]  Unlock my data.
[19:56.170 --> 20:11.870]  So the unsolved problem of how we decide when you get data and when you don't get data in this decentralized way across the network is one of the things that I think you guys in this audience are particularly good at thinking about.
[20:12.150 --> 20:21.330]  And that's the reason why we came here today, was to ask you guys to think about this sort of idea and to help us architect that layer of the system.
[20:22.890 --> 20:27.050]  So authentication turns out to be one of the really hard problems.
[20:27.190 --> 20:28.510]  There are a lot of people that have looked at this.
[20:28.630 --> 20:39.090]  I mean, every decentralized Facebook-like application, every decentralized social networking application, like Crabgrass or something like that, has wrestled with this problem.
[20:39.470 --> 20:42.430]  None of them claim to have solved it in an adequate manner.
[20:42.630 --> 20:44.710]  None of them claim to have nailed it.
[20:44.890 --> 20:53.670]  So we need somebody to start thinking about this in a serious way and we hope to provide a forum where people can do that.
[20:54.530 --> 20:56.170]  This is a solvable problem.
[20:56.350 --> 20:57.450]  It is an architecture problem.
[20:57.470 --> 21:01.950]  But it is not an impassable problem, even if it might be difficult.
[21:03.730 --> 21:08.730]  So the second major challenge of a device like this is security.
[21:10.370 --> 21:16.490]  Obviously, we're building a box that is going to hold all your data and everyone's going to have the same box.
[21:16.490 --> 21:20.190]  It's going to be maybe an enticing target.
[21:20.390 --> 21:20.770]  I don't know.
[21:21.410 --> 21:23.690]  Somebody's going to want to crack on this box.
[21:23.830 --> 21:25.110]  Somebody's going to want to knock on its door.
[21:25.710 --> 21:31.230]  And not only do we need to make sure that when we ship it, it's got the latest and best security we can find.
[21:31.250 --> 21:33.370]  We need to make sure that the thing updates properly.
[21:35.350 --> 21:42.750]  And on the one hand, security updates are a solved problem in the Linux world.
[21:42.750 --> 21:49.810]  On the other hand, the first time this thing breaks open, we're going to lose everybody's audience.
[21:50.310 --> 21:56.890]  So security on this box is not just keeping the box safe.
[21:56.890 --> 22:06.610]  It's knowing that the data on the box is safe and knowing that when you move the data to other boxes in an encrypted way, that it's going to get there and not get intercepted in the middle.
[22:07.690 --> 22:13.710]  And finally, the last real challenge of building something like the Freedom Box is the interface.
[22:13.990 --> 22:17.170]  And I said plug it in and walk away, and that was a little simplistic.
[22:17.450 --> 22:30.650]  But ultimately, it needs to be as easy as plug it in, answer a couple of questions, be up and running, and every time you want to do something additional, all you need to do is click a couple of buttons.
[22:31.150 --> 22:36.050]  And yes, you can get under the hood if you want to, but we don't want users to have to do that.
[22:36.310 --> 22:39.290]  It should be as simple as, I tell you how you know I'm me.
[22:40.150 --> 22:43.630]  I tell you how you authenticate me to you as a box, right?
[22:43.710 --> 22:46.510]  Plug a cable into the thing, hook up my laptop, whatever.
[22:47.250 --> 22:51.630]  Once you do that, it should have automatic defaults for as much as possible, right?
[22:51.770 --> 22:54.050]  Don't ask me, do I run the Tor network?
[22:54.190 --> 22:55.490]  Yes, just run it, just do it.
[22:55.490 --> 23:01.930]  So, assembling that interface from a bunch of pieces is a hard problem.
[23:02.070 --> 23:06.610]  And we in the Linux community are getting better at interfaces, but we haven't actually gotten there yet.
[23:06.790 --> 23:12.510]  So those, I think, are the three major areas where we need help actually making this box happen.
[23:12.610 --> 23:18.610]  We need a high-level view of what it's going to do, and then we need to start drilling down on these three hard problems.
[23:18.790 --> 23:22.310]  And ultimately, this box, it's just a Linux distribution.
[23:23.110 --> 23:28.570]  All this box is, is one variant of Debian, probably.
[23:29.610 --> 23:31.210]  It's got a target architecture.
[23:31.410 --> 23:32.610]  It's got a target format.
[23:32.950 --> 23:34.690]  But theoretically, it could run on anything.
[23:34.830 --> 23:35.890]  It could run on your old laptop.
[23:36.150 --> 23:39.230]  It could run on any box that you have that you want to use for this purpose.
[23:39.490 --> 23:41.790]  So that's the Freedom Box in a nutshell.
[23:42.070 --> 23:44.810]  It's a box that is designed just for this one purpose.
[23:45.030 --> 23:48.190]  Managing a little bit of your personal data and not giving it away.
[23:48.190 --> 23:53.970]  Letting you keep the data in your own apartment, without anyone else seeing it, unless you want them to see it.
[23:54.390 --> 23:55.670]  I will take questions.
[23:58.290 --> 23:58.730]  Yes.
[24:13.080 --> 24:18.940]  So the question was, how does this address the problem of, once data is out there, it's out there.
[24:19.060 --> 24:19.840]  You can't get it back.
[24:20.000 --> 24:21.660]  And it doesn't.
[24:21.780 --> 24:27.240]  It doesn't address the problem of, if you give your data to other people, you lose a little bit of control over your data.
[24:27.240 --> 24:31.220]  What it does do is it gives you control over who you give the data to.
[24:31.440 --> 24:37.000]  So if you only give the data to people you trust, to your friends, that's a private conversation, right?
[24:37.000 --> 24:39.360]  We have private conversations with people all the time.
[24:39.780 --> 24:42.400]  And they can repeat the things we say.
[24:42.560 --> 24:43.740]  They can forward our emails.
[24:43.800 --> 24:45.300]  They can gossip about us.
[24:45.480 --> 24:51.080]  There is a certain level of risk and trust involved in any interaction we have with any other people.
[24:51.080 --> 24:58.360]  The problem right now is we have interactions with individuals that happen at a certain expectation of trust.
[24:58.480 --> 25:04.300]  But the whole conversation is listened to by outside parties who have no respect for the trust.
[25:04.560 --> 25:11.360]  So once you eliminate those other third parties, your data is going to go as far as that person will take it.
[25:11.400 --> 25:15.320]  And you're left with just about as much privacy as you have in the offline world.
[25:15.440 --> 25:17.920]  If you share your data, you've shared your data.
[25:17.920 --> 25:19.760]  The question is, who are you sharing it with?
[25:19.900 --> 25:22.140]  And that's what this box is designed to protect.
[25:22.820 --> 25:23.200]  Make sense?
[25:24.460 --> 25:25.280]  Anybody else?
[25:25.640 --> 25:26.900]  Yes, the gentleman in the back.
[25:39.260 --> 25:43.700]  So the question was, what about physical access to the box?
[25:43.800 --> 25:46.940]  How do you keep the box secure if somebody has physical access to it?
[25:46.980 --> 25:52.120]  And the answer is, look, if someone has physical access to your box, you lose.
[25:53.140 --> 25:55.020]  That's pretty much universal.
[25:55.020 --> 25:57.100]  It will lock everything down and encrypt it.
[25:57.400 --> 26:03.500]  But if they come with a warrant, you're going to either give them the information or sit in jail.
[26:03.860 --> 26:05.580]  You're either going to give them the password or sit in jail.
[26:05.960 --> 26:08.500]  So, yeah, the box will have everything encrypted.
[26:09.720 --> 26:16.060]  Beyond that, if they want to, you know, brute force your key, there's not a whole lot you can do about that.
[26:16.060 --> 26:18.820]  I mean, it is a computer, and it's not much more than a computer.
[26:25.970 --> 26:29.050]  The question is, can you be compelled to give a password?
[26:29.350 --> 26:38.410]  And the answer is, people have been compelled to give passwords in certain instances.
[26:39.410 --> 26:46.950]  You know, we all read about the case of the system administrator in San Francisco who sat in jail because he wouldn't cough up the passwords to the local city network.
[26:46.950 --> 26:51.670]  So, yeah, that is a thing that could happen if somebody knocks on your door.
[26:51.810 --> 26:55.390]  But that's no different than any other risk you run by keeping data on a computer.
[26:57.110 --> 26:57.550]  Sure.
[27:16.640 --> 27:19.740]  Well, I mean, no one can use the data against you unless they have it.
[27:19.740 --> 27:24.040]  The idea is that if you keep control of the data, and you don't have to keep much, right?
[27:24.160 --> 27:31.460]  I mean, the thing is not... we're not going to intend for end users to do a whole lot of maintenance on it.
[27:31.860 --> 27:34.500]  So, it's not going to keep a whole lot of logs.
[27:35.100 --> 27:39.460]  So, it's going to keep just enough data to do the job it's designed to do.
[27:39.620 --> 27:42.620]  And if you delete something, it will really be gone.
[27:43.400 --> 27:47.900]  Beyond that, you know, if you're going to keep records, you're going to keep records.
[27:48.080 --> 27:51.600]  And yes, putting them all in one place, like I said, makes it an attractive target.
[27:51.600 --> 27:53.340]  But that's why we need security on the box.
[27:54.520 --> 28:05.160]  It is much better to have the data on your box and vulnerable to hypothetical threats than already in the hands of people who don't like you, which is where we are now.
[28:05.340 --> 28:05.660]  Yes, sir?
[28:29.460 --> 28:30.440]  Sure, sure.
[28:30.720 --> 28:32.180]  That is absolutely true.
[28:32.340 --> 28:38.880]  That the box itself as a physical object would be valuable to anyone who has the ability to break the encryption on it.
[28:38.880 --> 28:45.100]  And frankly, if people have the ability to break our strongest encryption, we've already lost.
[28:45.860 --> 28:47.780]  You know, people can take the box.
[28:48.080 --> 28:59.640]  Your average home burglar, if he has some giant, massive array of computers to brute force your key or some magic quantum technology, I don't know why he's robbing your apartment.
[28:59.900 --> 29:02.600]  But, I mean, you know, maybe it's his hobby.
[29:03.040 --> 29:06.400]  I mean, you could see Richard Feynman doing something like this, right?
[29:07.120 --> 29:07.880]  Yes, sir?
[29:08.460 --> 29:10.400]  My apologies if I...
[29:13.430 --> 29:15.130]  Like, so does this, like,
[29:18.550 --> 29:23.310]  browser, and they start giving away all the information on Facebook, is it going to stop them from doing
[29:26.440 --> 29:26.820]  that?
[29:27.060 --> 29:36.240]  I mean, you know, I don't think that we're going to pop up messages that say, warning, you were about to tell Facebook something you shouldn't tell them.
[29:37.240 --> 29:40.740]  Rather, people who buy this device are already interested in privacy.
[29:40.740 --> 29:48.500]  And so, it will be perhaps a tool for configuring Facebook to give you the best privacy within that environment.
[29:49.660 --> 29:59.900]  It'll perhaps have a competitor to Facebook, maybe one that's interactive with Facebook, so that you can participate in the Facebook community to a certain degree without giving them all of your data.
[30:00.800 --> 30:03.200]  People are going to want to act in the real world.
[30:03.520 --> 30:06.760]  And they're going to want to interact with companies that don't respect their data.
[30:06.900 --> 30:08.180]  There's nothing we can do about that.
[30:08.380 --> 30:11.960]  If you want to go do that, there's only so much protection we can give you.
[30:11.960 --> 30:16.960]  And we want to serve people along wherever on the spectrum they might want to be.
[30:17.240 --> 30:24.560]  But yeah, if you want to go and tell Facebook everything about yourself, we're not going to jump in your way and physically stop you.
[30:25.720 --> 30:26.240]  Sure.
[30:26.240 --> 30:27.200]  About
[30:30.540 --> 30:39.380]  ten years ago I went into some data mining and AMP was doing about a terabyte of data a day from their cash registers.
[30:39.720 --> 30:47.480]  And the worst thing you could ever do was to shop there one month after another and use your credit card.
[30:47.480 --> 31:09.840]  Because basically they knew if you drank Pepsi or Coke, had children who are dentures, shoe polish, I mean the amount of statistical data that you basically gave over was just astronomical and because your credit card data was all there and pretty much all day in the supermarket.
[31:09.920 --> 31:10.740]  Sure, I mean...
[31:10.740 --> 31:11.520]  But my question
[31:21.260 --> 31:23.420]  is for your box, which I think is a great idea.
[31:23.900 --> 31:32.260]  I think it's long overdue to basically say to Google, you have my social security number, give it back.
[31:33.320 --> 31:34.340]  Well, I mean...
[31:34.340 --> 31:35.500]  Or to anybody for that matter.
[31:35.540 --> 31:37.240]  The box isn't going to do that.
[31:37.380 --> 31:38.480]  I mean the box is not a lawyer.
[31:38.700 --> 31:39.620]  There are criteria for
[31:42.700 --> 31:43.500]  other methods.
[31:44.080 --> 31:49.440]  I mean, certain well-intentioned organizations may look at a point and say, no, but you
[31:54.630 --> 32:01.830]  feel that it's part of your effort to basically get back what is quite valuable in form.
[32:02.870 --> 32:09.290]  So the question was, the box is designed to protect privacy, but a lot of your data is already out there.
[32:09.390 --> 32:10.210]  How do you get it back?
[32:10.430 --> 32:13.150]  And the answer is, you can't.
[32:13.670 --> 32:17.790]  I mean, to the extent that data is out there, it is very difficult to retract.
[32:18.330 --> 32:19.850]  The idea is not to give new data.
[32:20.490 --> 32:24.210]  You know, and maybe some of us who have already given up our social security numbers...
[32:24.210 --> 32:26.590]  I mean, social security numbers are never going to be private again.
[32:26.750 --> 32:29.570]  And this box isn't going to change that specific problem, I don't think.
[32:31.450 --> 32:31.890]  Yes?
[32:32.730 --> 32:33.090]  So,
[32:36.700 --> 32:43.240]  despite the fact that you've already linked to the fair amount of data that we all have, why not just focus on pushing out as much more than it is?
[32:46.090 --> 32:55.070]  Yeah, I mean, you know, whether the box will or will not contain, you know, software to make spurious Google searches, I think, is an open question.
[32:55.230 --> 32:59.210]  It's something that will get decided at a much later date than we are now.
[32:59.370 --> 33:02.030]  I mean, right now, these are all good ideas.
[33:02.310 --> 33:04.030]  I don't know how many of them will actually stick to the wall.
[33:04.250 --> 33:05.710]  But yeah, it's definitely a good idea.
[33:05.710 --> 33:06.730]  Yes, sir?
[33:25.710 --> 33:33.610]  Yeah, I wouldn't say that I am a Linux partisan.
[33:33.610 --> 33:38.690]  And I wouldn't say that when I said it's a Linux box, that is a pronouncement of any kind.
[33:38.850 --> 33:42.550]  I mean, it will be a box that will run some variant of some NICs.
[33:42.770 --> 33:48.430]  That will be decided by whoever actually builds the box, whoever is actually doing the architecting.
[33:48.430 --> 33:49.450]  That won't actually be me.
[33:49.790 --> 33:50.530]  So, yeah.
[33:50.810 --> 33:56.270]  If OpenBSD turns out to be the thing that they decide to use, I don't see any reason why it couldn't be used.
[33:56.410 --> 33:58.730]  There's a gentleman in the back who's been waiting for a while.
[33:59.390 --> 34:02.690]  Can you continue with the social security number, for example?
[34:02.870 --> 34:03.330]  Sure.
[34:12.480 --> 34:14.380]  With the social security number.
[34:14.840 --> 34:17.040]  But you don't want to give them your actual data.
[34:36.190 --> 34:43.610]  In this situation, though, both you and the company would have to agree on a particular scheme.
[34:43.610 --> 34:58.370]  I was wondering if you know of any incentives for companies, particularly versions that we use every day, to hold on to just kind of random blogs a day.
[34:58.910 --> 35:04.090]  There's value in them having a lot of personal data, and there's an interest in it for them, not them.
[35:04.610 --> 35:09.750]  But what's their interest to respect our privacy by just holding on to the graphic blogs?
[35:11.110 --> 35:12.770]  Yeah, I don't know that...
[35:12.770 --> 35:23.530]  So the question was, companies right now take a lot of your data, how can we convince them to, instead of taking our data to hold cryptographic blobs for the purpose of identifying us?
[35:23.650 --> 35:25.230]  And I don't know that we can convince them.
[35:25.310 --> 35:30.390]  I mean, the box is not designed to convince companies to act against their interests.
[35:31.210 --> 35:39.590]  If you get enough people together who decide that using the box is great, and don't get me wrong, I hope that everyone will plug one into their wall, but a lot of people won't.
[35:39.590 --> 35:42.490]  Because they can barely plug their computers into their wall.
[35:44.790 --> 35:50.330]  But, maybe everybody knows somebody that they trust enough to run their plug for them.
[35:50.610 --> 35:54.330]  And maybe some of these plugs will be multi-user.
[35:54.550 --> 35:56.270]  And maybe some of them will serve a lot of people.
[35:56.390 --> 36:11.190]  And if you get aggregation of enough people together who are using this technology as a default, then the ability to adjust the technology on the box, adjusts the way people interact with the wider net.
[36:11.410 --> 36:13.650]  And the way they interact with other services on the net.
[36:13.830 --> 36:22.490]  And once you can aggregate that sort of power, you have an ability to have real conversations with companies about what their incentives really might be.
[36:22.790 --> 36:25.150]  Do they actually want to serve this population?
[36:25.150 --> 36:30.070]  Do they actually want to allow these people to use their services and take money from them?
[36:30.250 --> 36:35.930]  If they do, they might be willing to talk to us about what the limits of that data mining are.
[37:01.040 --> 37:08.900]  So, the question is, are we actually making software or should we just design a set of protocols?
[37:09.070 --> 37:15.360]  And the answer to that question is, at some point, you need to hand somebody a box with software on it.
[37:15.360 --> 37:19.900]  That box has to contain more than just descriptions of protocols.
[37:20.480 --> 37:22.280]  It needs to contain running code.
[37:22.480 --> 37:24.630]  That code might run on a BSD system.
[37:24.780 --> 37:26.020]  It might run on a Linux system.
[37:26.400 --> 37:30.550]  Maybe, you know, a project like this could support multiple systems.
[37:30.720 --> 37:32.170]  But we don't see any need to.
[37:32.820 --> 37:39.280]  Frankly, the idea is to get one working version of it, not to get something that compiles on every chip and board in the planet.
[37:40.220 --> 37:42.240]  Get something in people's hands that they can use.
[37:42.240 --> 37:45.670]  Because we are not partisan between various platforms.
[37:45.960 --> 37:48.000]  But on the other hand, at some point, you do have to pick one.
[37:49.320 --> 37:49.760]  Yes, sir?
[38:01.960 --> 38:08.020]  I agree that the boxes are attractive targets for people looking to break in and take data.
[38:08.020 --> 38:09.740]  And that is why security is important.
[38:10.000 --> 38:27.460]  And the other thing, I'm just thinking with the public called Magic Rocks, this is a couple of years ago, Jim Ward, where he was looking at authorization codes going to and from various cable boxes, where he would change some into good codes, some into fair codes.
[38:28.400 --> 38:33.220]  In the same way, but this has got to do more than just look at simple statements.
[38:33.220 --> 38:40.660]  The state has just got to look at a whole lot of stuff going to and from to describe what's good at hand.
[38:41.000 --> 38:46.560]  Yeah, I mean, the question was, how deep do you need to get into the data to actually make real policy decisions?
[38:46.800 --> 38:48.440]  And that's a valid question.
[38:48.600 --> 38:54.580]  But the answer is, you don't need to get very deep at all to make some of the big policy decisions, right?
[38:54.580 --> 39:01.260]  You can look at traffic and say, oh, you know, that is a torrent.
[39:01.600 --> 39:05.600]  We need to route that in a way that doesn't move it towards us, right?
[39:05.760 --> 39:07.280]  This is a VPN connection.
[39:07.520 --> 39:08.860]  We can route that through the VPN.
[39:09.120 --> 39:14.240]  So to the extent that people are moving data around, we can look at the packets just like anyone else.
[39:14.340 --> 39:15.980]  We can shape our own traffic.
[39:16.920 --> 39:22.500]  We can tell it more about what that traffic might look like if we want to give it more information.
[39:22.500 --> 39:25.280]  And that will enable us to enact real policy.
[39:25.540 --> 39:28.660]  So you don't actually need to read every packet that goes by.
[39:28.820 --> 39:31.540]  You can gather data from the context of each packet.
[39:32.160 --> 39:32.360]  Yes?
[39:39.110 --> 39:45.680]  The question was, what about personal legal liability for running Tor exit nodes and what else?
[39:46.430 --> 39:47.790]  And open proxies.
[39:48.060 --> 39:56.790]  So, you know, I'm not going to claim to be an expert in the legalities of running Tor exit nodes.
[39:56.790 --> 40:00.080]  If you really want that answer, you should talk to Wendy Seltzer.
[40:01.520 --> 40:07.810]  But her opinion is that, at least when I've talked to her, is that there isn't a whole lot of legal risk.
[40:08.180 --> 40:16.230]  And if you are worried about the legal risk, you can still run a Tor node that is not an exit node.
[40:16.230 --> 40:17.790]  If you think that's a problem.
[40:18.290 --> 40:19.290]  That's the short answer.
[40:19.290 --> 40:22.600]  I could not give you specifics, not having looked into it specifically myself.
[40:22.850 --> 40:26.250]  This is something I've used to do.
[40:26.590 --> 40:26.890]  Yes?
[40:28.250 --> 40:28.650]  Yes.
[40:29.030 --> 40:29.290]  Absolutely.
[40:29.910 --> 40:30.710]  The gentleman in the back.
[40:31.070 --> 40:31.670]  It sounds
[40:38.380 --> 40:39.500]  like we were in the box.
[40:39.700 --> 40:44.670]  Do you have any thoughts on dealing with the problem that you're really forgetting?
[40:45.030 --> 40:45.430]  Yeah.
[40:45.690 --> 40:48.630]  The question is, what happens if the user gives up the password?
[40:48.850 --> 40:51.670]  Or even more importantly, what happens if the user forgets the password?
[40:52.150 --> 40:54.310]  I think those are both hard problems.
[40:54.750 --> 40:59.990]  When I talked earlier about authentication being difficult, that was the kind of thing I'm talking about.
[41:00.130 --> 41:02.910]  No one has really solved this problem well.
[41:03.150 --> 41:08.330]  And one of the reasons why is that no one has solved the problem of the untrustworthy user.
[41:09.110 --> 41:11.990]  We don't know what that solution is going to look like.
[41:12.110 --> 41:16.370]  I have some ideas, but I don't think any of them are very good.
[41:17.670 --> 41:20.530]  If you have better ideas, I would really like to hear them.
[41:26.100 --> 41:27.520]  I am talking about Shiva plugs.
[41:27.720 --> 41:27.860]  Yes.
[41:28.960 --> 41:29.360]  Oh.
[41:30.400 --> 41:31.080]  Here we go.
[41:31.220 --> 41:31.780]  Shiva plugs.
[41:31.980 --> 41:36.900]  So, if you are looking for an idea of what the box will actually look like, this is it.
[41:37.040 --> 41:41.260]  It's a rectangular piece of plastic with an outlet on it.
[41:41.360 --> 41:41.760]  Plug.
[41:42.040 --> 41:42.840]  So, thank you.
[41:43.380 --> 41:44.440]  The new ones are smaller.
[41:45.160 --> 41:46.260]  The new ones are smaller.
[41:46.460 --> 41:50.280]  And, I mean, the brilliance of this is that when they first came out, they were around $100.
[41:50.500 --> 41:52.960]  I just bought three of them for $24.99.
[41:52.960 --> 41:56.420]  I mean, these boxes are really, really cheap.
[41:56.820 --> 42:02.980]  So, the idea of being able to put this into people's hands at a price level that they can afford is realizable.
[42:03.280 --> 42:06.060]  This isn't the one laptop per child project.
[42:06.060 --> 42:09.040]  This isn't a giant hacking marathon.
[42:09.940 --> 42:10.020]  Right?
[42:10.220 --> 42:12.000]  The hardware is done.
[42:12.400 --> 42:15.340]  The software is most of the way there.
[42:15.340 --> 42:20.460]  It's the crucial interconnecting bits, a little bit of interface, and a little bit of authentication.
[42:21.440 --> 42:22.480]  And that's it.
[42:22.820 --> 42:22.900]  Right?
[42:23.160 --> 42:24.980]  This is a Linux distribution.
[42:25.020 --> 42:26.720]  Or a BSD distribution.
[42:27.100 --> 42:29.840]  And that sounds like a big project.
[42:29.860 --> 42:33.780]  But really, you're only cobbling together a very limited suite of software.
[42:33.780 --> 42:37.920]  It's not going to take 20,000 man hours to put this together.
[42:38.460 --> 42:40.020]  So, thank you for the demonstration.
[42:40.360 --> 42:40.720]  Yes, sir.
[43:17.230 --> 43:18.250]  Well, sure.
[43:18.410 --> 43:27.110]  I mean, you know, he's talking about separating data that is personal to you versus data that has just passed through your VPN or passed through your Tor.
[43:27.430 --> 43:31.990]  And I don't understand why anyone would keep any data that they don't need.
[43:32.010 --> 43:34.530]  And you certainly don't need data about stuff that has passed through.
[43:34.530 --> 43:37.730]  So, if you don't have data, you can't give it away.
[43:38.310 --> 43:43.270]  And the idea would be that if you are running a Tor node, you're not logging that stuff.
[43:44.090 --> 43:52.990]  Well, one of the reasons is that one of the ways to reduce your security is to take your information and put it in other machines.
[43:52.990 --> 43:53.250]  Sure.
[43:53.690 --> 43:53.950]  Sure.
[43:54.050 --> 43:55.130]  And if that is...
[43:55.130 --> 44:04.190]  If spreading your data to other machines is a thing that people want to do, yeah, we would have to do legal architecture as well as hack on it from a tech point of view.
[44:04.530 --> 44:05.170]  Yes, sir.
[44:27.010 --> 44:27.410]  Sure.
[44:27.410 --> 44:33.410]  Your data is probably not going to be out of your hands for weeks or months before they finally say that, oh, you are innocent.
[44:33.730 --> 44:34.850]  Here's your box back.
[44:35.110 --> 44:38.270]  By the way, we kept a log of everything that you have on there just in case.
[44:38.870 --> 44:39.130]  Sure.
[44:39.330 --> 44:39.550]  Yeah.
[44:39.690 --> 44:41.250]  I'm not saying there are no legal risks.
[44:41.410 --> 44:41.630]  Absolutely.
[44:49.890 --> 44:53.750]  If you want us to give the Tor node a different IP address, sure.
[44:53.750 --> 44:55.410]  Or, I mean, I would say a physical...
[44:55.410 --> 45:00.310]  I mean, personally, I would say the Tor exit node being built until it would be a killer for me.
[45:00.510 --> 45:01.130]  I would not...
[45:01.650 --> 45:02.410]  You could turn it off.
[45:02.630 --> 45:02.770]  Yeah.
[45:02.890 --> 45:03.890]  It will be configurable.
[45:04.510 --> 45:05.050]  Well, okay.
[45:05.050 --> 45:05.110]  Yeah.
[45:05.670 --> 45:09.970]  It wasn't clear that you were saying before to give them an option just to...
[45:09.970 --> 45:10.410]  Oh.
[45:10.730 --> 45:15.110]  You know, my idea for configuration is sane defaults, right?
[45:15.330 --> 45:16.910]  And I believe that's a sane default.
[45:16.910 --> 45:17.590]  Y2.
[45:20.180 --> 45:20.700]  So...
[45:20.700 --> 45:21.160]  Okay.
[45:28.090 --> 45:31.430]  With GNU for both PTP, the original social network.
[45:31.710 --> 45:32.010]  Mm-hmm.
[45:32.230 --> 45:34.410]  MonkeySphere was a presentation in this room earlier.
[45:35.410 --> 45:39.150]  And GNU social P2P.
[45:39.450 --> 45:47.070]  So the suggestion was that we look at both MonkeySphere and GNU social P2P as pieces of software for the box.
[45:47.110 --> 45:48.110]  And I think that's a great idea.
[45:48.430 --> 45:50.770]  I think those are both good pieces of technology.
[45:51.610 --> 45:52.270]  Yes, sir.
[46:09.490 --> 46:09.930]  Yeah.
[46:10.050 --> 46:13.290]  The question is, what about getting data into the box?
[46:13.650 --> 46:18.070]  And, you know, that again is an open question, right?
[46:18.150 --> 46:20.370]  I mean, how are you going to tell it what you've done?
[46:20.590 --> 46:25.310]  I guess you could give it your credentials, let it scrape, let it download.
[46:26.270 --> 46:31.390]  You know, the question of how we do that in a legal manner is an open question.
[46:32.390 --> 46:32.830]  Yeah.
[46:33.610 --> 46:35.310]  What about any concern
[46:46.560 --> 46:47.020]  in the box?
[46:47.230 --> 46:53.880]  Well, when we talked about the system layer, that sort of standard privacy model is the kind of thing that would need to be developed.
[46:54.060 --> 47:03.980]  Now, I mean, we live in a world where user permissions and group permissions are the starting point for a lot of how we think about access controls.
[47:03.980 --> 47:12.460]  And I expect that that would be the starting point for this conversation as well, identifying groups of people and saying, treat them in this manner, right?
[47:12.520 --> 47:13.380]  This is my friend.
[47:13.660 --> 47:14.660]  This is my mother.
[47:15.000 --> 47:20.390]  Don't show my mother pictures that I only want my friends to see, because she'll freak out.
[47:20.940 --> 47:21.500]  All right.
[47:23.260 --> 47:23.890]  Yes, sir.
[47:37.910 --> 47:40.610]  Well, the idea is that this box is your firewall.
[47:41.750 --> 47:45.450]  Obviously, if you plug it in at work, you might have a different problem.
[47:46.770 --> 47:51.950]  There are ways to bounce information off of computers that aren't firewalled.
[47:51.950 --> 47:56.990]  And anyone who has done any work in the torrent world is aware of that kind of stuff.
[47:57.290 --> 48:05.030]  But yeah, I think punching holes in firewalls is a thing we as a community have gotten really good at quite recently.
[48:05.970 --> 48:08.990]  For the most part, though, we really hope the box is your firewall.
[48:09.390 --> 48:16.530]  And in any conversation between two people, we think there's a high likelihood that one of those people is not behind a firewall.
[48:19.680 --> 48:20.360]  Yes, sir.
[48:37.060 --> 48:37.540]  Sure.
[48:37.740 --> 48:44.920]  I mean, the question was, how can you receive email or contact if you never tell anyone where you are, if you're always hiding?
[48:45.060 --> 48:49.200]  And the answer starts with dynamic DNS, so people can find you.
[48:49.580 --> 48:57.440]  And the back end of that answer is some form of encryption, so that people can write to you without other people listening.
[48:57.640 --> 48:58.520]  Now, that's privacy.
[48:58.740 --> 49:04.400]  If you want anonymity, you're going to have to start routing things through something like an onion network.
[49:04.600 --> 49:09.600]  And there are ways to route things through anonymizing paths to get to you.
[49:09.740 --> 49:13.260]  You can even do something that looks a lot like email in that context.
[49:14.040 --> 49:21.080]  I don't know that this box is going to do that right off the bat, but it is something that we would certainly hope to see happen at some point.
[49:23.880 --> 49:24.700]  Any other questions?
[49:24.940 --> 49:25.300]  Yes, sir.
[49:29.020 --> 49:29.460]  Yes.
[49:32.200 --> 49:36.560]  Well, I mean, we're not going to sit down and write the Facebook clone.
[49:36.720 --> 49:44.740]  We're really hoping the Diaspora guys succeed in a way that makes a lot of that technology reusable, in a sense.
[49:44.780 --> 49:48.960]  In the same way that we're not going to write an onion router, we're going to use Tor.
[49:49.600 --> 49:53.380]  If there is off-the-shelf solutions that work, we would really like to use them.
[49:53.780 --> 50:11.180]  We really believe that by creating a network of decentralized servers, where everyone can have their server back instead of being a client in the system, that people will start to write applications that use that new slash old network architecture.
[50:11.180 --> 50:15.100]  And, yeah, Diaspora is definitely a project that we have our eye on.
[50:19.840 --> 50:21.180]  I don't see any other questions.
[50:21.520 --> 50:29.200]  So, I'm just going to say, this project, as you can see from this conversation, is still at a very early stage.
[50:29.720 --> 50:31.580]  There's a lot of things that need to be worked out.
[50:31.780 --> 50:35.920]  And this conversation is not about telling you the box is going to do X, Y, and Z.
[50:36.060 --> 50:38.180]  The box is going to have ABC software on it.
[50:38.400 --> 50:42.520]  This conversation is about telling you that the box is being planned.
[50:42.520 --> 50:52.700]  This conversation is about saying, if you are interested in helping to develop a box that protects your privacy, the time to get involved is now.
[50:52.860 --> 50:56.420]  The time to give us your ideas is now.
[50:56.540 --> 51:04.880]  And the time to help us identify other people who would be interested in putting some real energy behind an effort like this is now.
[51:05.020 --> 51:19.380]  So, if you specifically have ideas, have access to resources, have names of people who you think would be well served by working and collaborating on this idea right now, talk to me afterwards.
[51:19.600 --> 51:31.720]  Because I would love to talk to you, get your information, and start actually connecting people who might want to start doing the actual work of hacking instead of standing here talking to you about stuff that doesn't exist.
[51:31.720 --> 51:35.620]  So, thank you very much, and I hope to talk to all of you afterwards.
[51:35.620 --> 51:35.640]  Thank you very much, thank you very much.
[51:35.640 --> 51:35.760]  thank you very much for the experience.
[51:36.060 --> 51:36.400]  Thank