While the present situation is extremely rosy for inflation, which stands as the favoured model for the origin of structure, there is a sense in which the present is the worst time to be considering inflation models. A quick survey of the literature suggests that there are perhaps of order 100 viable models of inflation, the most there has ever been. At the original Inner Space, Outer Space meeting in 1984, there were only a handful. It's true that some models devised in those 15 years have been excluded, such as the extended inflation models [31, 23], but model builders have for the most part had quite a free hand operating within the given constraints.
Further, this is likely to be about the most viable models there will ever be, because observations are at the threshold of significantly impacting on this collection. Experiments such as Boomerang, VSA and MAP are capable of ruling out inflation completely, by one of the methods outlined in this article. If inflation survives, they will have significantly reduced the number of models, and then a few years later PLANCK should eliminate most of the rest. Hopefully, by the time of Inner Space, Outer Space III in 2014, we will be back once more to a handful.
Thanks to Ed Copeland, Ian Grivell, Rocky Kolb and Jim Lidsey for collaborations which underlie much of the discussion in this article, and to Anthony Lasenby for stressing to me the importance of the absence of vector perturbations as a prediction of known inflationary models. I thank the Royal Society for financial support.