 competition. So of course perfect competition is approximated by free software. No participant in a free software market has market power specifically when you look at free software projects. Anyone can contribute and not talking about the market where people are paying for stuff because that is not the market of free software production. There are no barriers to inferior exit. You can use any software that's there. You can leave and change to any other software that's there. The price and quality of things are known to all, known again in the economic sense of being available. Even if you directly can't read the software it's available for you and you can get anyone else you want to read it. And there are no transaction costs because you can get that software or return it back in. And what's interesting is that free software projects are extremely competitive. There is much more competition in the structure of free software development than there is at the average team in a company because there are no employee-employee relationships. So everyone is trying to get their own idea their own piece of code accepted. And so a long time ago I started writing about the economics of free software and did an interview with Linus Dorvas. And one of the studies looked at the pervasive structure of competition that exists within the cooperation. There are very, very dense links and very complex links between every single module in a free software project. We actually looked at, we chose, we looked at the Linux kernel. The structure of competitive cooperation in the Linux kernel back in 2002 looked like this. There are 169 modules. The links between those modules show whether there's a technical dependency between them and whether they are common authors. So there's social links and knowledge flow as well as technical links. As you can see, it's a philosophic ed because it's extremely complex and dense. But there is an enormous amount of competition going on between modules, between different versions of modules, between forks of modules before something eventually gets taken up and goes into the main branch. And all these other things are available. Every single competitor is equally available and has equal access more or less to the market of ideas. But the best one ends up often winning. So to conclude, we shouldn't choose openness because it's good. The elevator pitch now that I'm used to doing this with VCs, the elevator pitch for openness is not, it is good and close it bad. We should choose openness because it promotes fitness. Openness is economically efficient and socially just. Open ecosystems, open economies and open societies succeed. As the 20th century showed us, although it may take painfully long, closed economies and closed societies eventually do fail. Thank you.