Computer Science has been in crisis for several years. Interest in studying it has dropped dramatically. We can wring our hands, or we can do something about it. Computer Science needs to engage with pure outreach: selling the subject (for free). Our approach has been to do this by just going out and having innovative fun. This is more effective than selling specific courses or institutions. CS is after all a naturally exciting, innovative and thought-provoking subject (Oh, and by the way there are good jobs at the end too). We see it as a Renaissance subject (so who cares what it's called or which variation the future depends on). It sits in a unique position, centrally connected to all of the sciences, arts and humanities. We are passionate about science generally, so we go out and spread that enthusiasm about it all. What have we been doing? There is cs4fn (www.cs4fn.org) a website and magazine that we've been writing for the sheer enjoyment of it; Sodarace (www.sodarace.net) where over 130 000 registered humans and computers compete in an online Olympics; Brain Academy (www.brainacademy.qmul.ac.uk), the Compute-Ability competition with career enhancing prizes; research talks for kids on Artificial Intelligence, women in technology, disability and mutant super-hero powers (actually its about search algorithms) and so on; a Computer Science Magic show (you have to promise not to tell anyone the secrets); and exhibits at the Royal Society Summer Exhibition on mathematical illusions and their link to computer science (www.cs4fn.org/illusions). We build brains that play Snap from rope and toilet roll, and introduce a piece of paper more intelligent than anyone in the room. Kids act out Turing Tests (can you tell the human from the robot?), and we challenge them to solve puzzles with CS twists. We do real research too of course: a spin-out of our EPSRC funded research project on Human Error and Interactive Systems includes an online SpaceInvaders game-experiment (www.cs4fn.org/humanerror) the data from which helps us understand the causes of human error: can you consistently avoid making the mistake that will blow up your ship and lose all your points?. What age group should we be targeting? Sixth formers? A major issue is that school ICT, vital as it is, gives a poor impression of how being a computer scientist is about being an innovative, creative, computational thinker. By the sixth form it is too late. It is the younger kids we have to get the message to. An early experiment has been teaching a version of a graduate level course on usability evaluation to primary school kids (www.cs4fn.org/manorside). They proved to be very innovative and "amazing", "will stay in my mind forever", "I want to teach others in the school what we did" were some of their comments. Our approach works: teachers, industry and the International Review of ICT have commended us & and we have seen an increase in undergraduate applications of over 130% in 2 years. Not bad when we are just having fun: serious fun.