 OK, so who's used Skype or Viber, WhatsApp, Facebook? Are any of these genuinely free technologies? They're not free. Are they run by charities or non-profit organizations? They're run by companies. I mean, what is the purpose of a company? It's to make money. I mean, if you were starting a company, you would want to make money. And to make money, you need a product. So if you're making a service like Skype or Facebook, what is your product? If you're giving your app to people for free, where is the product? Advertising? The customer's other product, yes. So they're selling the customers not just to the advertisers, but to anybody who wants to study or manipulate the users of those networks. It's not just advertising. It is control. You can look online and you can find plenty of examples of Facebook tweaking what you see to manipulate you or control your emotions, to understand how much money you have to work out if you're earning $1,000 a month or $1,000 a day, because advertisers would like to know that so they can actually give you different offers. They might charge you more money for something if they think you can afford it. So it's not just straight advertising. It's to manipulate you. It's to understand you. And it's to profit from you. Now, Skype is a bit like a sausage, that if you knew what was inside it, you really wouldn't want to eat it. But billions of people do, unfortunately. Many people eat sausages, too. But can we change that? Do you believe that we have the skills and the experience here at FOSAsia to replace Skype with a genuinely free alternative within a year if we really put our minds to it? Does anyone believe that's possible? Yeah, I do. I think we have the skills. Here's one of the people who will help. This is Alex from Savoie fair Linux. They've recently launched a new product called Ring. It is a peer-to-peer distributed system for voice and video communication. So it's no central point of failure. So that's just one example of a new initiative to replace Skype. And they're making really good progress. Come and see his talk at 1 o'clock tomorrow. Here, we're going to have a look at Metcalfe's law. Can everyone read that OK at the back there? So Metcalfe's law tells us that the value of a communications network, v, equals the number of users squared. It may not be exactly that, but it's certainly proportional to the square of the number of users. It's not linear. But if you start off here with 20 users and you add another 20 users to get to 40, you don't just double the value of your network. You go from here all the way up to here. So the network becomes four times more valuable just by doubling the number of users. And this is really important in the way we solve this problem with free software. So if you imagine getting all the users in the first, say, two rows of this room onto a product. So if I came around to each of you and asked you to install my app, Lumicall, and we could all call each other, the value of that network using Metcalfe's law would be 100. Now if we reached out further to more people in the room, so if each person got one more person to install that same app, then that doesn't just double the value of the network. It quadruples the value from 100 to 400 just by each user adding one more user. Now if you fill the room with, say, 200 people and you get everyone to install the same technology or technologies that are compatible with each other, the value of that network is 4,000. Let's say we got everybody in Debian, about 1,000 developers, to use the same technology for voice and video communication. You've now got a network with a value of a million. You can see this is shooting up very quickly, isn't it? Just by adding more and more users. Let's say you get the whole free software community, all the people that come here to FOSSAsia, the people that are over in Boston at Libra Planet, the people that came to FOSSSTEM, all the people that you collaborate with through projects like GitHub. Let's say that was 50,000 people and we get them all communicating with SIP, for example, or Ring, or XMPP, or another solution. The value of that network is 2.5 billion. It's not so many people, but the value has just shot up dramatically again. If they all get some of their friends, I've put five million friends, so that's each developer getting 100 friends, which they may be able to do over a week or it might take them six months. I think most people could get 100 friends onto a platform within six to 12 months. Then the value shoots up dramatically again. And finally, we've got the whole world. So 36 billion billion billion is the value of that network and network that spans the whole world. What's really significant about these figures is that they also reveal the value of the networks we're competing with. That a network that has a big value using this scale has a lot of power over networks that have a small value. And it's not just linear, it's quadratic. So a network like Facebook, being somewhere between here and here, has an extreme amount of power over a solution that we use in one of our communities. Just by getting all of our communities, so Debian and Fedora and Ubuntu and FOSAsia and other groups using the same technology, would make us all a lot more powerful in comparison to that competition. But ultimately, we have to start thinking about solutions that are built for the whole world. Because we have to think about that big number, that 36 billion billion billion. Is that what the corporations are after? That's what Facebook wants, that's what Google wants, that's what Skype wants. This is why we see people like the head of Google going over to visit North Korea recently. Because that's a few million people that he's not counting in the value of his network just yet. Doing this seriously means thinking about solutions that scale to the whole world. At a technical level, that means we need to think about scalability when we design our solutions. We need to think about making them easier to install so that more people will install them. People that don't know as much as we do about how to operate and change their computer. Is Ring using a distributed hash table? Is that aiming for that level of scalability? You're going to find out. Okay, well, let's help them find out, okay? It's a scalability test. It's not a denial of service attack. So this is a frog. If you put your frog in a pot of cold water and you just heat up the pot gradually, your frog is going to sit there until he gets warm. He's going to get warmer and warmer and he's going to get a little bit uncomfortable. And your poor frog, and no frogs were hurt in the preparation of this slide, your poor frog is going to boil to death. If you take your frog and you boil your water first and you drop your frog into the water, he's going to jump out. He's not going to stay there. He's not going to put his life at risk. Most users are like this. They're like the frog in the water that started out cold that every year more and more of their privacy is being stripped away. And they're not changing their behavior. They're signing up for more and more services. I mean, who's familiar with Tinder? And people have just rushed into it and you can't sign up for Tinder without signing up for Facebook first. If you tried to launch a service like that 10 years ago where you insisted that to use one service you needed to be signed up to another, many people would have been extremely skeptical. But today, this is the new normal. This is the water getting warmer and warmer to the point where it's going to boil with many of your friends and maybe even some of yourselves boiling in there with it. Now, going back to some of the things we can do at a technical level, I mentioned scalability already. There are three main areas that we need to address in order to engage users. The ease of installing and configuring free software solutions, both the free software operating systems like Debian and Fedora, and also individual apps or soft phones. How do we install them? How does the user set them up the first time? How difficult is it to configure their account properties? Making all of that as easy as possible makes a big difference. If you cut the number of setup questions by half, you double the number of users. If you have 10 fields in your setup form and you reduce that to five fields that they have to fill in, you double the number of users. Convenience of dialing. This is one way that Viber and WhatsApp have become big very quickly is they integrate with the address book in a smartphone. So that's made them very convenient. So Skype was very dominant and then WhatsApp and Viber came on the scene and they took a lot of market share very quickly through that convenience of dialing, of using the existing phone numbers to dial. And reliability and quality. So when people make a call, they need to be able to hear the other person. They don't want these one-way audio calls. They don't want the video cutting out for extended periods of time. So we need to address those types of problem or at least we need to be able to give reliable messages to say, look, your bandwidth is not good enough. We have protocols that can monitor the bandwidth and give feedback. And at the moment a lot of soft phones don't use that effectively and the user just has a bad experience. Just telling the user, look, your line is not good enough shifts the blame from the soft phone to the thing they need to fix, which is their internet connection. But if they don't know that, they blame the soft phone. So what next? I've gone through a number of soft phones and related products over the last year and I've been rating some of the issues in each product that would make a really big difference to this goal of replacing Skype and those other proprietary solutions with free software. And I've been collecting that information at freertc.org, which is the website mentioned there on the slide. I've also been building the RTC Quick Start Guide. So that's an online book. It has the diagrams and a discussion of all the different issues in deploying free real-time communication. It's aimed at communities and businesses who want to run their own SIP and XMPP server. So that's atrtcquickstart.org. You can download the PDF or you can browse it online. And the aim of that is that people who want to set this up that already run their own mail server or other infrastructure can set this up themselves. The book focuses on federated solutions and that tends to be more interesting for corporate users and enterprises. There are also peer-to-peer solutions like Ring and those are great for families and friends because they have no central point of control, no central servers. So look at both of these things. That's all we've got time for today. I'd invite you to come back tomorrow for Alex's session about Ring that will focus in more detail on one possible solution. Have a look at these websites. Come and join the free RTC mailing list. There's a link from the websites. And I'm more than happy to answer questions afterwards over drinks. So thanks for coming.