On to the ground-based (i.e. NSF funded) recommendations (for large, new projects -- i.e., not including on-going investments in ALMA; there are a number of interesting medium scale projects recommended, but I probably won't have time to get to them). First priority was the Large Synoptic Survey Telescope (LSST) -- a survey for a multi-color, multi-cadence survey of the sky with an 8m class telescope. As my colleague and LSST Project Scientist Zeljko Ivezic puts it, "LSST will make a movie of the sky," which, you have to admit, is pretty cool. When you think about discovery space in astronomy, the largest gains come when you move into new regimes. We've largely run out of new wavelength regimes, but the time-variable regime has not yet been explored in a large scale systematic way (although PanSTARRS and the Los Cumbres Observatory will certainly be making headway). In addition, the co-adds of all the epochs will produce an 8m telescope version of the 2.5m Sloan Digital Sky Survey (SDSS) imaging, which is a good thing. All data is non-proprietary, and can be used by anyone. Second priority is a "Mid-Scale Innovations Program" -- basically, a ground-based equivalent of the NASA Explorer program. The decadal survey committee reviewed a wealth of scientifically compelling medium size projects. These don't rise to the level of building giant new facilities, and are typically seeking funding for an instrument and a decidated multi-year survey on existing facilities. The report recommends that there be a review and funding mechanism for such projects, which have the capability of responding nimbly to scientific and technological changes. Third priority is contributing to the development of a 30m class ground-based optical/nearIR telescope (a "Giant Segmented Mirror Telescope"; GSMT). Such a telescope would be essential for carrying out spectroscopy of the sources found at the limits of 8m-class telescope imaging; basically, if you detect a source in an image, when you want spectra, you're spreading the light over much larger areas, requiring bigger apertures to reach the same signal-to-noise as when all the wavelengths are being imaged together. There are currently 2 large US programs that are well underway (TMT and GMT), using private funding. For these programs to have enough money to be built and operated, an investment of Federal money is required. This money would also guarantee some degree of access for the larger US community, but probably significantly less than 50%. The report recommends that involvement should be at least a 25% share. However, they argue that there is only money enough to invest in one, and the community had better pick one as soon as possible, rather than letting both go forward. The fourth priority is participation in the "Atmospheric Cerenkov Telescope Array" (ACTA), to detect and characterize the highest energy cosmic rays. Recent years have seen the detection of TeV cosmic rays, which places strong constraints on particle acceleration at the highest energy scales; a new array would greatly expand the chances of fully understanding the origin of these high energy events. Rather than funding a separate US initiative, however, the report recommends joining into an existing European project (CTA), in spite of the fact that the US would be a minor partner. Reactions to the Ground-Based Recommendations: Perhaps the biggest surprise was the drop in the GSMT from (1) its prioritization in the previous report, and (2) its prioritization in the actual optical/IR subcommittee (See Table B.1). The justification was that LSST was a much lower risk in terms of cost and technology, and, as in the space recommendations, pragmatism ruled the day. The committee was quite strong in their support for GSMT as a project, and pointed out that the combination with LSST is highly synergistic -- LSST provides the targets, and GSMT tells you what they are. However, the pie was simply not big enough to give everyone a slice. In addition, if you can only dish out one slice of pie, you want it to feed the most number of people -- LSST made a strong case that a much larger fraction of the US community could make use of the data. Personally, I'm very sympathetic to this view. There are scientific advances that come because you have new facilities pushing into new territory, and GSMT has this in spades. However, there are also scientific advances that come about because you have the largest number of very clever brains thinking about how to exploit a given data set. Taking SDSS as a model, a ridiculously large fraction of the ridiculously large number of SDSS-related papers had absolutely nothing to do with anything in the "black book" of science justifications used to obtain funding for SDSS. You take good data, you let smart people work with it, and you'll get science you never anticipated. I'm optimistic that LSST could work the same way, with the caveat that the scientific impact may well be blunted without a wide scale investment in spectroscopy (which SDSS had, and which LSST lacks). I very much hope that a 30m gets built, but not to the point where I'd be comfortable leveraging all public large ground-based investment over the next 10 years for a 25% share of a telescope. (Full disclosure: I am not at an institution that would have private 30m access, and am at one that has made early and ongoing investments in LSST. So, my perspective is undoubtedly shaped somewhat by viewing GSMT projects as a potential "outside" user. I do my best to be fair, but I've pretty much shaped my scientific research around the premise that I won't have exclusive access to large aperture telescopes.) I am also really pleased to see the "Mid-Scale Innovations" recommendation. I think this is a smart way to make sure we can take advantage of rapidly changing fields. When something like dark energy or extrasolar planets shows up on the scene, it's great to have a mechanism in place to take advantage of new opportunities. In addition, it's a smart way to skim the low hanging fruit, so that larger missions have a better understanding of what the scientific requirements really are -- for example, you'd design a very different dark energy mission if you know that w is nearly equal to -1, than if you had no idea of its value. The other noticeable lack here is a call for US participation in the Square Kilometer Array. (The panel did recommend some radio projects in the medium scale category.) However, if you look at Figure 4-8 (which I found fascinating and surprising) fewer than 10% of the members in the American Astronomical Society (ASS) categorize themselves as "Observational Radio" astronomers. I'd presume this would grow in response to investment in ALMA, but the community is clearly not enormous. So, my take on the ground-based recommendations, is that they did pretty well at making hard choices. And the choices were indeed hard, and are going to be rightfully hard to swallow in many cases.