All right after a morning and early afternoon of a lot of stuff about policy and things it's time for a technical talk in party track and we'll hear about a new tool and new service that will be exciting and once again we have a first-time attendee and a first-time speaker.  So and Elijah is totally calm, not worried about it at all!  What's that?  Eijah, I apologize, he's thrilled to talk to you guys about his tool that he's been working on for a long time, demonsaw, let's give him a big party track welcome!  
(Applause)
	>> Thank you. Recently, our right to share files on line has been under assault by governments, corporations, and others who fear openness and personal privacy rights.  People have been persecuted, fined and even imprisoned simply for sharing data electronically.  As private conversations transition from the home to the web, we're losing our fundamental rights to privacy and personal beliefs.  
	Imagine for a moment a secure file sharing application that lets you share whatever you want with anybody you wanted across all of your devices without fear.  Governments, corporations, and others would not be able to know what you are doing, what you are sharing and know who you are.  They wouldn't be able to track your IP address, and not being a centralized base network they could not take you down.  
	Now, imagine a million file sharing networks just like that.  This presentation is a free data manifesto, historical analysis and recipe for creating a new type of file sharing application.  My name is Eijah and I believe in the right to share.  
	Glenn Greenwald said one can easily remain free of even the most intense political oppression simply by placing one's faith and trust in institutions of authority.  But I think the Bible says it even better in Psalms 56:3, when I am afraid I put my trust in you.  The modern Internet --
(Pause)
The modern Internet is a very different place today than it was in the past.  We can define the modern Internet by four different states.  The state of trust, a state of convenience, a state of control, and a state of change.  
	Over the last 15 years, the Internet has shifted from a web of isolated content to a cloud of in the operable networks and trusted third parties.  A primary reason for this state of trust is technological innovation.
	>> Speaking of change... 
(Laughter)
	>> Shot time!  
	>> So companies have begun to support open source technologies, web and data standards, shared protocols and in the operability.  This has been going on for the last 15 + years.  But it's more prevalent now than it ever was before.  
	Interoperability being the mutual coexisting of disparate domains, single sign-on, shared authentication protocols and open ID which is what we use with Facebook accounts.  Now I'm ready.
	>> All right.  So you all know the drill.  By the way, I wasn't gonna interrupt a Bible verse walking in here.  I had to pause for a second, I was like, oh, shit!  It is very hard to get selected at DEF CON.  So congratulations, welcome to the team!
[Cheers and applause]
Thank you, sir.  So the benefits of trust are cost-effective applications, companies can roll out apps fast they are than ever before.  Faster to get apps on the market, easier to share content across domains, different companies are working together that never did 10 or 15 years before.  And there is a level of convenience to us as users of this content.  
	Now, trust is made possible a level of convenience that we have grown accustomed to and can no longer live without.  We have improved application usability like signing in with our Google and Facebook accounts everywhere.  There is everything at our fingertips, all content, our bank accounts, social networks, everything we can get to from laptops or home computers, phones, tablets.  And everything is simple now.  
	There is obviously concerns with this.  Such as privacy.  Our anonymity, potential abuse from these companies and an overall loss of control.  For a model based on trust, the cost of convenience is control.  When our personal data is in the hands of others, we have lost control.  
	At that point, we are faced with some rather difficult questions.  Who owns our data?  Are we simply a source or are we greater than the sum?  
	A fundamental flaw in the model of trust is this loss of control.  And once we have given up control, it's gone forever.  My favorite examples of a loss of control and never being able to get it back are I have three of them.  One is release of the AACS processing and keys and continual release of those.  Number two is and I still find this absolutely fabulous, the Tom Cruise scientology videos, if you haven't seen those, I highly advise them, seven minutes well worth to watch those.  
	And of course WikiLeaks and the great work they are doing with we leasing documents.  Loss of control brings concerns to us as private citizens.  Data retention and oversight.  Do we trust the companies to take care of our data?  
	Legal compliance matters.  What corporations are giving our data over to the government?  What are governments doing with that data?  Security breaches and what happens with data when companies cannot protect what is ours.  Overall transparency.  Do we have the right to know that what companies are doing with our data behind the scenes?  
	Finally, the bottom line, because everything a company does is for its' own best interest and is driven by financial means.
	The truth about this modern internet is that convenience does not require trust. We can achieve the same level of convenience that we get today, without giving up control and without trusting governments and companies.
	The power to change the world is in our hands.We're experts in our fields. We're passionate, envied and feared. We have the knowledge, tools and experiences and most importantly, we have the free time to code.
	The path forward involves becoming advocates for distrust, even Ronald Reagan, I've quoted the Bible now I'll quote a Republican, even Ronald Reagan said, trust but verify.  We can break the mold of old and stale thinking, create something beautiful and secure and we can open up the Internet for what I call digital self-expression.  
	Stephen Hawking said, "The increase of disorder or entropy is what distinguishes the past from the future, giving a direction to time."
And Aaron Schwartz said, "Information is power, but like all power there are those who want to keep it for themselves."

	Reinventing file sharing means that we need to understand the current file sharing applications available to us now.  We also need to understand how secure or insecure these applications are.  So let's look back in history and look at the existing file-sharing applications to see what type of network model they are based on.  There are two primary models for network sharing applications.  Centralized model and the decentralized model.  
	The centralized model is a tried-and-true model with over 45 years of history.  Did you know that FTP is over 45 years old?  
	I did not know that until I did a little bit of research for this presentation.  That is amazing.  
	The pros of the centralized model are fairly obvious, it's a simplified programming model.  If you take in any computer science class or you get bored at home and code like I do you know how simple it is to open up a socket connection and send data.  Very, very simplified program model.  There are dedicated hosting benefits for centralized.  Web hosting now only costs a fraction of what it did five, 10, 15 years ago.  You can get a shared hosting environment, Linux or .NET for $3 a month.  That is extremely inexpensive.  
	The cons are logging inaudiability.  Lot of credential based security with a centralized model.  Companies are storing our information, there is a question of digital rights management.  Identity or IP and usage transparency, and the centralized model is easily susceptible to down.  A single point of failure and very very easy for governments or corporations or the legal system to go after a single point.  Here are examples of different types of file sharing applications that use the centralized model.  Most will be familiar to us.  We have file system base like samba or RTFS.  We have IRC, FTPs, we have Web-based choices like mega, rapid share, we have new streaming applications like Netflix, prime, go, et cetera.  And we have got Cloud Computing which entered the market a few years back like Google drive, Amazon web services, Dropbox, et cetera. 

	The decentralized model, on the other hand, is a rather newer model for file sharing but imperfect as well.  The pros are no central point of access or failure.  It could also last forever.  A well designed decentralized file sharing application or network, once live, has the potential to never be taken down.  However, there are some concerns with the decentralized model as well such as identity and IP usage transparency.  That's the same con as we saw in the centralized model and we'll talk about that later.  You also have the loss of anonymity, IP address is revealed and that could be linked to your name.  
	There is personal liability which is something new.  The legal system can go after us directly for downloading files via BitTorrent.  And then there's a security issue, too.  Really quick I want to point out some of these, these should be all familiar but BitTorrent to go way back, Napster, the start of it all, stuff like plex, are sync, plex is newer to the market and then streaming applications that I think are very exciting like Chromecast and of course XPNC which has been around for a long time now.  
	File-sharing is an amazing technology.  But this doesn't mean it's perfect.  There are some fundamental problems with file-sharing.  For example, file share is insecure.  Either we trust a third party source like a centralized server or set of applications, or we risk revealing our identity via a peer-to-peer or decentralized network.  There are also legality issues with file sharing.  It's also inflexible.  I want my files available to me at all times.  I don't want to force sync content across boxes and what if I'm offline?  I can't stream Netflix.  Or I can't grab my files via the Dropbox, so I want more flexibility in file sharing.  
	It's also inconvenient.  I want to watch the show that I'm paying for.  I don't want to watch "Ads" and by the way the thing that most upset me about HuluPlus, is the requirement that we watch ads.  Some are funny but seeing the same ad four times over in an hour is not funny anymore.  Why do we have to watch ads?  
	File sharing is also unreliable.  If you look at streaming providers, they have exclusive content and they have licensing disputes which changes the content available in their sites.  Netflix, prime does this all the time.  Network outages.  Why when they go down should I suffer?
And there's not enough seeds for BitTorrent.  I want something nobody else wants.  How do I get that?  
	File sharing choices are sometimes expensive.  Has anyone looked into upgrading their Dropbox accounts?  Free two gigs that were given costs substantially more when you want 100 gigs.  That shouldn't be the case.  Hard drives are cheap these days.  The cloud is also expensive.  And not only the price, but what are these companies doing with our data?  Who has access and what if it's stolen?  Netflix and Amazon prime are doing it the best.  HBO Go not so much, I don't want cable or satellite.  My HBO subscription is through U torrent.  Some of my friends get theirs through the HBO subscriptions of their friends and their logins.  I would prefer to watch HBO through Netflix or Amazon prime.  Granted HBO finally caved in recently.  If you want to watch shows five years old.  What about the shows that are out now? 
So my Game of Thrones subscription is through uTorrent.  The modern Internet has also taught us to value something if it is good, convenient, and reasonably priced so I want to throw those 3 things out there.  The modern Internet teaches us that convenience is a good thing and we like convenience.  But it's also taught us to value things that are good, convenient, and reasonably priced.  
	This is why we still torrent, this is while we still download illegally.  I find from my personal use if a show is on Netflix or prime I'm likely to stream it because it's good, convenient, and reasonably priced.  Actually more work to Torrent it.  Finally file sharing problems or file sharing can be unfair.  Not all usage patterns are the same.  BitTorrent network is a very important and powerful network and we all use it but it's designed to adhere to a state of pseudoequality across the board.  If it did not, then we would not be able to download if we didn't have to upload.  So the ratio of seeding to downloading is very, very critical.  But not all usage patterns are the same.  
	What if I only wanted to download?  Is that possible?  Well, it should be allowed.  The differentiation between acquisition, which is the initial acquiring of new content, and the subsequent aggregation of that, which is the disbursing and sharing should be separate models in a file sharing app and configurable separately.  
	So these problems with file sharing have led to the creation of a niche market for solutions.  Inadequacy breeds innovation.  VPNs, proxies, file wall, dark net, tor, peer block, which doesn't work but people like it none the less, and hacking our neighbor's access points are all cures to the symptoms and not the problem.  You laugh, many of us do it.  I probably shouldn't admit that I would do such a thing but if I did I would laugh as well. 
	(Inaudible) 
	Yes.  Exactly.  
	So reinventing file sharing, we need to -- in order to reinvent file sharing we need to leverage the power of our Internet access.  I will quote Aaron Schwartz who said something absolutely amazing 2007.  He said "When you have a distributed network like the Internet everybody can be a server.  You can take your home laptop and run a server off it that can distribute movies and music and Web pages and e-mail in the same way that the biggest computers at Google can.  There is no fundamental difference between the computers they have in the racks in their server rooms and what you have on your desk."
	He was absolutely correct.  Reinventing file sharing means we need a new solution that's secure, anonymous, free and everywhere.  What does it take to reinvent file sharing?  Norm Chomsky said if you want to achieve something you build the basis for it.  And Julian Assange said the only way to keep a secret is to never have one.  
	Let's talk about five key principles that will secure file sharing network and maintain the privacy of data.  
	Number one, saving cyberspace means that we need to reclaim the authoritative source.  Well, what is the authoritative source?  The authoritative source is the primary repository of trusted data.  It's basically our data that we're about to share.  
	When we give away the authoritative source we give away the control of our sensitive data.  
	The price of online data storage is our anonymity and potential for abuse of our data.  The foundation of file sharing is data.  Right?  Everything is based around data.  Data should be the first thing that is secured and the last thing to be compromised.  It should never leave our control.  Companies I used to work in the financial sector, companies I learned are very, very good at securing their devices.  Their corporate devices like laptops and phones, and USB drives, they excel at securing those, but they are very bad at securing network transmissions between those devices.  The weakest link are the users and the methods by which they exchange data.  
	When we lose control of our data, we lose control of everything.  This includes the power to protect what's ours, the certainty that our data is protected, and the choice to respond to attacks and the ability to remain anonymous.  The solution?  We need to take responsibility and reclaim the authoritative source.  
	Our data should not leave our personal devices.  It should never go into the cloud, it should never go into Dropbox, it should never go beyond the devices we own like this laptop or my phone or my tablet.  
	We need to do what companies continue to fail us on and that is secure our data.
 	Number two, saving cyberspace means that we need to redefine authentication.  Well, what does it mean to authenticate?
We do it all the time, multiple times per day, simply authentication is the process by which I prove who I claim to be.  That is really it.  Right now, there are many, many forms of stateful authentication.  Stateful authentication is used for credential-based or certificate-based  authentication where there is some sort of state being maintained.  Usually in a database.  It's a necessary model but in some cases it can be antiquated and unnecessary.  Relying on trusted third parties is not always the best solution for authentication.  Stateless authentication is a form of shared secret authentication that leverages shared assets or other known data to mutually authenticated clients.  
	The benefits are obvious in that the modern Internet gives us a wealth of information or shared secrets.  Such as pictures that we have uploaded to Facebook or posts we have made or tweets, et cetera.  There is an immense amount of information available that can be used to share secrets across multiple people.  Mutual coexistence by shared knowledge is the basis for stateless authentication.  
	This means that the shared secret is something that is obvious to the very specific group of people.  There is no registration process and no data storage.  No need for databases, SQL or otherwise.  And the beauty of this is we can use a series of dynamic encryption algorithms based on this information because no state needs to be maintained.  So there's no certificate authorities, there is no databases, there is no, nothing in the background that can link what we are doing to our identity.  
	So the solution for stateless authentication is redefining the need to authenticate.  Let's create a temporary trust.  We have learned trust is bad and we should be advocates for distrust.  But temporary trust that can be changed and modified over the course of seconds or minutes or hours is valuable.  
	Our social networks have an immense amount of shared secrets we can tap into.  No data is stored and the loss of security because insignificant with stateless authentication.  Why?  It's not necessarily a secured model that needs to stand the test of time.  It's a transitory nature, stateless authentication that has the ability to adapt via dynamic encryption algorithms.  
	Number three is saving cyberspace means that we transmit a module approach to security.  Modular security is something that is based on layered security and layered security being the practice of combining multiple mitigating security controls to protect data, make it more difficult to compromise the whole.  While modular security takes layered security and builds on top of it by creating more of an obfuscated and disjointed separation of duties.  I use the double-lined approach so nobody knows the entire secret and there's an inability to store complete secrets.  It's not feasible in a network design based on modular securities to have any secrets.  It's a bit like a free-for-all, a hail Mary in a way, but a guaranteed touchdown so perhaps a bad analogy.  
	There are existing applications today which are two tempting targets for take down.  Too many single points of failure that reveal too much information.  Modular security provides plausible deniability.  We should always assume that someone is listening.  Always be paranoid.  It's better to be paranoid than to be caught. And what they don't know won't hurt us.  They cannot audit what we do not have.  The solution implement multiple modules of defense to resist penetration.  As messages propagate through the transmission stream, they will increase in strength.  But nobody will be able to know the entire secret and even a compromise of 99% of a network will not reveal the entire secret.  Each module therefore is isolated, autonomous, self-sufficient and resistant to attacks.  
	Number four, saving cyberspace means we need to leverage existing protocols.  Our choice of protocol message or otherwise does not define us as an application.  When we create a new file-sharing app it's really about art and science.  This is my second file-sharing app in the last 15 years and it truly is always a journey.  This one I'm about to share with you shortly has taken a year part-time, as a labor of love.  But creating a message protocol is true wizardry and have more difficult than creating a file sharing app.  Use standard protocols the benefits are obfuscation, which there's a Japanese proverb I love that says "The nail that sticks out gets hammered down.  This means calling undue attention to ourselves will only result in problems.  So let's blend in with the rest of the http network traffic, pretend like we're no different. 
It's interoperable, will work with existing web services and sites, and hopefully undetectable transfers.  The solution for a profile, use these preexisting protocols, stuff like HTP, XML, JSON, etc.  And finally number 5 is saving cyberspace means that we need to support a more flexible and adaptive model for file sharing.  What does this mean?  Flexible and adaptive.  Well, first of all, we need to be simple and effective.  It needs to be easy to use our file-sharing app.  BitTorrent is an amazing, amazing protocol and uTorrent is an amazing application.  I use uTorrent most every day.  It's very complex and a lot of my friends aren't necessarily like us, they might not have the same technical background so if we can make something that everybody can use, then file sharing can be everywhere.  
	Individuals, families, and organizations, small, medium and large, should be able to use this application equally.  It should be available on all devices, phones, tablets, work stations, laptops, servers, and even low-power consumption devices like the Raspberry Pi and Ouya and even these badges should be able to run file sharing networks.  Very, very possible.  Also operating systems should be agnostic, Windows, Linux, Mac, Android and iOS should support all of them.  
	Finally, we need a hybrid model.  We need a file sharing app that takes the best of the centralized model and the best of the decentralized model and removes everything else that is crap.  How do we create this hybrid model?  We need to abstract the content from the transfer.  No fixed servers and no direct peer-to-peer. So nobody knows about your IP address and no fixed servers, needs to be pervasive and ubiquitous, everywhere and throughout everything and it needs to be secure, anonymous, free, and everywhere.  
	Edward Snowden recently said, "I don't want to live in a world where there's no privacy and therefore no room for intellectual exploration and creativity" and Bjarne Stroustrup, his name is very difficult for me to pronounce, one of my personal heroes, founded C++, said people who think they know everything really annoy those of us who know we don't.  I think that's true.  Right.  
	He's a very -- far smarter man than I am and he seemed to get annoyed quite frequently!  Do we really need, let's be honest, do we need another file-sharing application?  I struggled with this a year ago before I started writing Demonsaw.  There's a lot out there and every year there's a new file sharing that comes out.  Recently I found myself halfway around the world.  I won't tell you what area of the world or what country but it was far, far from home and I found myself there for many, many months.  And I realized that I had no good way of getting access to all the files that I had downloaded previously.  I had no good way of sharing or accessing my content.  And I felt a little vulnerable and I thought, why is this the case?  I could have FTP'd or uploaded it to a cloud or done Dropbox or Amazon a shit load of money to host my content but I realized I wasn't happy with the current state of file sharing. 
I wanted to share with friends and family or 10,000 complete strangers.  I wanted to do it whenever I wanted on any device.  I wanted access to all my content from anywhere in the world.  I wanted convenience and control.  Why doesn't a solution for me already exist?  
	I realize this is a problem that we can solve.  I would like to introduce you to Demonsaw.  Demonsaw is a tool that allows me to share my files my way.  In fact, just this morning, we made Demonsaw available to the public.  Demonsaw.com, if you go there, you can download it, have fun, do whatever you want.  It's free.  
	When I was stuck across the world, I realized that most file sharing networks, they really break down to three fundamental components.  When you look at any file sharing network.  BitTorrent, FTP, doesn't matter.  They all come down to three components.  That is you have a client or multiple client nodes.  What these do is these either upload or download.  They either share or acquire data.  Then you have this type of node that serves as really a message propagator.  What it does is controls program flow.  It makes sure messages get sent where they need to.  This can be torrent trackers, or this can be built into the client in the case of more direct applications like FTP.  Then finally you have a server component that is really responsible, the dumbest of the three components.  All it does, it's like a tank, all it does is gets data sent through it, 99.99 % of the bytes get sent through the server, all it does is propagate data.  Most file sharing applications have combined these three components into two components.  Either the client and the server are combined, or the router and server are combined and the client is separate.  
	Demonsaw is a completely new type of file sharing application.  Architecture is based on a combination of network routing, Web-based and security patterns, but it's succeptively simple at its core.  The magic behind Demonsaw is it keeps these three functions, three different client, router and server nodes separate.  It doesn't try to combine them.  Because of that it's able to achieve an enhanced level of security and anonymity we'll see shortly.  
	What I want to do now is give you a quick, very very brief demo of Demonsaw and just show you what it can do, just a local demo but it will be a live demo.  I'll just talk you through a very, very basic scenario.  What will happen here is a router, here just a Windows-based router -- all right.  Thank you.  
	Hmm.  
(Pause)
All right.  We'll use our backup demo, then. Which is a static demo. Hopefully that will work.  Okay.  Sorry about that, there are always technical difficulties in live demos. 
	I will be available after this, in the cafe.  And I will be glad to walk you through the live demo or answer any of your questions that you may have.  
	So I always come prepared.  Let's talk about the demo.  Julian Assange also said non-conformity is the only real passion being ruled by, Sean Parker, very very famous person, anybody know Sean Parker said you can now be the master of your own destiny.  
	What you will see here is you'll see the client.  Basically there is three downloads that I have made available.  There is a Windows-based router which serves only as a message Facilitator.  All it does is groups clients and controls message flow and that's it.  
	There is the client which you see here and that is just going to share, search, browse, and transfer.  And then you have a Web-based server I have available which is a .NET web app that you can drop in any .NET web hosting provider and all your data will be sent through that.  For $3 or $4 a month you can go to a hosting company and get something that supports .NET 4.5, in fact, fuck it, get 25 of them for 100, get 25 hosting environments for 100 bucks a month, and you could basically be routing as much traffic as you want through it.  Because web hosting is so cheap these days.  
	But the web server will allow the router not to handle large streams of data.  In this example we have searched for I think Daft Punk.  For some reason I want to listen to the new Daft Punk this day.  So you just type search, bang, gives you the results instantly.  In the next window you have a browse function.  This allows you to navigate through any of the content in your secret group.  Now, a secret group will be defined by an AES key that will be dynamically created by some sort of shared asset.  In this version of Demonsaw that will be simply an image.  
	Let's say the front row and I all have that picture of the Christmas party where things got a little crazy.  Right.  Got a little crazy, we all had a good laugh and then we wanted to forget about that party.  Let's use that similar imagine and the 12, 13, 14 of you are in a group that nobody else can get to. 
Anything we share will only be available to us in the group but even through using the client we are not gonna know that I'm downloading from you or you are downloading from me or you have X, Y, and Z files available.  But we are gonna isolate ourselves to a specific group simply by this shared image from which our dynamic AES keys will be created.  
	We'll be able to browse each other's content and, by the way, the names of the clients here are really irrelevant.  You can make them whatever you want.  I just, server and work station worked for me.  
	You can transfer, download stuff into your status.  Very Napster-like.  And here you can share content.  So I can choose what folders I want to share and this is a little browse window where I can get access to my content.  And then right here you see the little image of the little demon guy, that is the image from which the AES keys will be created.  So if you want to deny someone access to your group, this is why security is almost unnecessary and insignificant for Demonsaw, I just change my image and nobody has access to my shit.  Say one of our friends that we don't like to hang out with found that image and he suddenly wants to transfer all our Daft Punk.  For some reason I don't like him.  We just change our image.  There's no need to revoke security access, or we change an image every day, something different.  
The lack of an image will default to the global or default group, basically what BitTorrent does now.  If I choose not to have an image it defaults to nothing.  
	Now, sorry about that.  Technical difficulties.  
	Now, how does Demonsaw address all these problems we talked about with file sharing?  Right.  First of all, it's secure.  There is no peer-to-peer.  There is no IP addresses that are revealed.  Nobody can know what you are downloading, nobody can even know what you are uploading and there is no centralized servers, routers and servers can be dynamic and shift as much as they want.  In fact, I don't have my phone on me, but -- now I do.  Right now we're working on an Android port of Demonsaw that will bring client and router to your phone.  What that means is I'll be able to host a private file sharing network on my phone and all data that gets transferred can be sent through hundreds or thousands of different servers all around the world.  They can be in countries that don't discriminate against data sharing rules or data sharing in general but I can control my own private file sharing network from my phone. 
That's what we're working on right now on Android and I can download stuff from the phone.  We'll also be working on Chromecast and possibly plex plug-ins so you can download and the 2% or 3% that's done you can stream it to TV.  Anywhere in the world have access, not just local like plex or DLNA or XBMC but you can do this from the phone.  We all have old phones, Androids or otherwise sitting around, doing nothing, plug them in, make them routers.  Take them with you, have access wireless networks and suddenly you have a mobile router that is your file-sharing network that nobody knows what you are doing.  Demonsaw is encrypted.  Everything is encrypted.  Multiple layers of encryption.  We leverage encryption standards.  You could theoretically rotate he description keys and algorithms at will.  That is something we can easily add if there is a need. 
There are varying encryption types and algorithms, AES, mutating, automatic, isolating, data-driven and stateless algorithms.  There are varying encryption techniques, messages and data are encrypted separately.  So if someone sniffs your network and they happen to get access to a message packet and they happen to know the shared Diffie Hellman key by some act of God and they happen to decrypt that message packet they won't even know what you are trying to do.  They won't know what packet, what packet that belongs to, whether it's upload are or downloading, no idea if you are sharing a file or a turbo tax document from 2003.  They'll have no idea.  
	In order to know that, they have to completely compromise the entire end to end chain which is what we saw with modular security.  
	Of course authentication and authorization is all done via keys.  Demonsaw is anonymous, no log-in, no registration, Mo data retention, no loss of control.  I'll go one step further.  It's free and there's no ads.  You can use it however you will.  No callback to demonsaw.com, I don't even know you are using it, I don't even care.  A year ago I started on this journey to write this, to meet my needs and I want to share it with everybody because I thought maybe other people share my need.  Yours to use however you want.  Set up a sniffer and verify I'm not sending any data back to demonsaw.com.  It's modular, separation of messages and content.  We are unable to deduce the type of content exchange.  This of course is the need-to-know basis.  It's simple.  You saw the interface.  It's share, search, browse, download and upload. 
That's it!  I'm sure the Android and iOS versions will be even simpler once we create them.  
	Use it at home, work, or while traveling.  It will go right through most firewalls because it's standard http, nobody will even know what you are doing which is great because I didn't really think about the security ramifications of this but if you set up a Demonsaw client and share a C drive, yes, you can transfer your entire work computer and work from home.  I've not tried this.  I do enjoy my day job and the paychecks that come as a result of it, but that is the theoretically possible.  Firewalls and proxies, as long as http can get through, this shit can get through.  It's everywhere, Windows, Linux, Mac, Android, iOS and the Web.  Right now we just released the Windows version of the router and the client.  And a Windows web application based version of the server.  The router will do server functionality.  Router is a server as well. 
But if you specify additional servers like the web server component it will override that and allow the router only to do message propagation.  
	We're working on the Android version.  It's written in C Sharp right now but we're putting it to c++ so we can roll it across all these platforms.  Just a matter of time before it's on Linux, Mac, iOS and you should see hopefully something in the next three months we'll launch the Android version, fingers crossed.  It's based on design patterns.  I'm a game programmer, that's my professional job and in game programming we use what's called the entity component system and what this does is favors composition over inheritance and it makes adding new code and expanding the system extremely easy.  In terms of faster, more flexible features and maintenance we will be able to roll these features out fast because it's based on -- entity component system. Demonsaw is silent and unseen.  We leverage standards, we avoid suspicion, and we remain undetectable. 
Flexible and adaptive.  As I said before, share files just with yourself, share them with family and friends or share it with 10,000 people.  A million file sharing networks.  
	Excuse me.  Finally, if you are curious what the architecture looks like here is a picture of a very simple Demonsaw network.  Router is performing message propagation and data propagation services.  There are four clients in this example.  They don't have to be all different devices but here is a very, very simple way you can share your content just with yourself or with three other people.  You just, there is a UPNP functionality in the router and it will just open up and the ports appropriately in your routers.  If you just have stuff and want to get to it from work this is a great way to do it and not trust Dropbox or the cloud.  Since it's your shit on your devices it's free!  This is a slightly more complicated design which I think will be probably the majority of you.  This is a friends and family model. 
Here you have a router that just does message propagation in the center.  Very, very lightweight.  This can be on a phone or on a old Windows box you have sitting around doing nothing.  You have a bunch of clients that can be anywhere and you have some servers here.  One a Web server and one Windows server that does data propagation services so when files are uploaded or downloaded they are sent through the servers and router is very very lightweight.  Finally a much more complicated network topology of multiple routers that are completely isolated with multiple servers and multiple different groups that all have different AES keys because they have a different shared secret.  
	Green groups have the same shared secret, red have different, and purple have their own.  We are currently working on session propagation technology which would allow any routers that opt in to be able to propagate the sessions of people that are logged in which means you could theoretically log into any router and be able to access anybody else.  This is something that we're working on right now, haven't thoroughly tested it, haven't vetted security concerns with it but it's something in the works.  I think it's kind of a cool idea.  
	And in summary, Albert Einstein said "Only a life lived for others is a life worth while."
I created Demonsaw as a labor of love to meet a personal need.  I want to share it with you because I think it met my need and if it meets your need as well I encourage you to download it, use it, let me know whether you like it or not.  If you have feature ideas please e-mail me.  As Steve Jobs said and one more thing, "changing the world, our potential is unbound.  We possess a tremendous amount of talent, we're good at what we do and we enjoy what we're good at.  We can create something new and beautiful that can change the world and we can do it together.  Demonsaw is a tool that deviates from the insecure models of file sharing.  It is a new way to share content without fear.  It's secure, it's anonymous, it's free and it's everywhere.  My name is Eijah and thank you very much.  You can contact me here and Twitter and the website. 
Thank you.
(Applause)
	>> Eijah, I have one question for you real quick.  Can you explain how it is that you are paneling encrypted transportation over http?  
	>> We're almost out of time but I'll take a quick stab and if you want more information I'll be at the cafe.  Basically what we're doing is using standard http so adhering to those protocols and data appears to look just like http.  So it's basically if your firewall is set to allow http content, port 80 or 443, this gets sent but because there is no direct connection coming into the network there's no issues with firewalls blocking incoming transmissions because everything is designed from an inside-out perspective.  If that doesn't answer your question I'm glad to go into more detail.
	>> Do you have the protocol speck published someplace?  
	>> I don't but e-mail and follow me on Twitter and I'll be glad to talk about it in more detail, as well as opensource, I have gotten questions about that.  I haven't had time with development to even think about open-sourcing it but it's something that I'll definitely think about.  If you have some thoughts about that, I'd love to hear them.  Thank you again.
(Applause)