The paper with Kallosh, Linde, and Trivedi was the first that gave a concrete recipe for starting with 10d supersymmetric strings and producing pseudo-realistic 4d models with no moduli fields in a controlled approximation scheme. (Silverstein had given a different recipe for stabilizing moduli in compactifications of noncritical string theories, which are nonsupersymmetric theories that start in more than 10 dimensions). We made the observation that in our construction, it was easy to envision dynamical mechanisms that allow one to add a small positive vacuum energy (roughly because the models naturally incorporated exponentially small physical scales relative to the Planck scale). Of course by "small" we just meant a vacuum energy small compared to the Planck scale (and comparable to the scale of supersymmetry breaking).

The mystery of how our world has such a small vacuum energy, compared even to the apparent scale of supersymmetry breaking, was one we could only "solve" by invoking an idea of Bousso and Polchinski: that the fluxes give rise to so many possible vacua, that some of them will also have this additional cancellation.

We also made some observations in our work which were relevant to conceptual questions being debated at the time, about whether string theory could truly have vacuum states that are infinitely long-lived and have positive cosmological constant (i.e., give rise to de Sitter space).

We gave a simple and general argument that, based on very reasonable assumptions, de Sitter models in string theory would always decay with a lifetime short compared with a natural time-scale of de Sitter space, the recurrence time.

My main further work in this area has been along three lines. I have tried to develop ever more explicit examples of moduli stabilization (both as suggested in the original proposal and in other string theories), since the original construction was very complicated and had many moving parts.

I have suggested several ways to embed pseudo-realistic inflationary models into string theory. And I have been thinking about whether these ideas and techniques can help us build particle physics models from the stringy perspective, or suggest new kinds of particle theory models as alternatives to the current front-runner, the supersymmetric Standard Model.

This work (along with work by many others) helped to motivate research in several further directions by many groups. In one direction, the recipe we suggested for moduli stabilization was more fully fleshed out and realized by various groups. The number of such flux models is indeed very large, and this has led to the development of a statistical theory to classify the crude distributions of properties of these vacua (developed mainly by Douglas and his collaborators).

In a second direction, people have worked intensely to really convincingly embed inflation into string theory, since the problems associated with runaway moduli (which previously hampered such attempts) can be solved in such scenarios.

A third result of our paper, along with the work of Bousso and Polchinski, Silverstein, and Susskind, was that a picture of the string "landscape" in which many de Sitter vacua coexist and are populated by quantum tunneling (giving rise to eternal inflation) has been widely discussed in recent years. This picture is very controversial. It provides a natural home for anthropic arguments, which people view with suspicion. It also suggests that many of our observed physical laws may be contingent or environmental (that is, not unique possibilities dictated from the top down by string theory, which instead makes a very large set of laws possible in different inflationary bubbles).

  Where do you see this work going in 5 years? In 10 years?

We are set to learn a great deal about both the correct particle theory model of the weak scale, and the correct model of inflation, in the next 5-10 years. The Large Hadron Collider (LHC) and the Planck satellite will both provide new insights in that time span. This data could be as large a perturbation on our field as the discovery of a cosmological constant was.

If LHC finds evidence of low-energy supersymmetry, the burning question will soon become to understand supersymmetry breaking and its mechanism of transmission to the Standard Model. Here, the ideas about flux compactifications and moduli potentials may well play an important role in stringy attempts at understanding. Lack of supersymmetry (or some other natural mechanism to protect the weak scale) would be very puzzling, and might be interpretable in a landscape scenario by invoking environmental arguments.

If the Planck satellite sees inflationary gravitational waves, it will mean that inflation occurred at a very high scale and with an inflation field rolling over more than a Planck distance in field space. Such models are the hardest ones to understand in string theory; one can very tightly constrain their existence. So this would be a big hint, and would reduce the number of reasonable inflationary proposals (in the string context) to a handful, at present.

Other surprises may happen. The observation of cosmic superstrings from lensing or gravity wave signatures could provide a different kind of hint, supporting models of inflation based on D-brane dynamics.

At a more conceptual level, the understanding of the landscape that I mentioned is in its infancy. Our understanding of string vacua after supersymmetry breaking is not nearly as precise as we would like. There are deep confusions associated with eternal inflation that arise in this context. There are hopes, but no convincing ideas, that vacuum selection may be dynamical instead of anthropic. Given the large landscape of possibilities, it also becomes natural to ask, "What exactly does string theory allow and what doesn't it allow?" So attempts to really define what is and is not characteristic of string vacua have been renewed, with some interesting suggestions in recent months. I think conceptual developments in all of these areas will be important, and are less tied to the direct results of the exciting forthcoming experiments.