I'm sorry, marcus, but as you well know, strings continue to dominate high energy theory, despite your efforts to fool people into somehow doubting this. In fact, marcus, strings have gone through a number of (relatively) barren periods, the last one ending when D-branes and dualities where discovered ushering in the so-called second string revolution. However, these tools may have taken us as far as they can on their own, so we need to find new ones that will help us see more deeply into a vast, complex and profound theory which has already changed utterly and permanently the way we think about quantum theory, but whose underlying principles have yet to be discovered or fully understood. One high profile issue at the moment is how to accomodate in strings the observation that the universe is in a period of accelerating expansion. The obvious approach (although not necessarily the correct one; strings could - and in my view probably will - surprise us) is to invoke the idea of a positive cosmological constant and hence study within string theory the simplest solution to einstein's equations that has one, namely de sitter space. However, there are problems to solve in relation to treating de sitter space by string theory in it's present form, and string theorists by and large do not find any of the current approaches to this issue convincing. In fact, we feel not that strings have "crashed" but that we simply haven't learned enough about what string theory actually is to clear this very interesting hurdle (that is, if it is a hurdle). The general level of faith in string theory really hasn't been shaken very much by any of this since, really, string research is still widely viewed as being in it's infancy, with this by now familiar cycling between periods of consolidation and discovery not being viewed as reason to change fields. In fact, university physics departments are as keen on hiring string theorists as they've ever been. This would of course change if strings could be shown to be inconsistent, but so far - unlike every other approach - it has passed everyone of the many and nontrivial consistency checks put to it.