Imaging and spectroscopy with the next generation of optical/near-IR telescopes

In this paper, I hope to do a little crystal ball gazing into the future of imaging and spectroscopy with ground-based optical and near infrared telescopes. Predicting the future is tricky at best, but I hope to convince the reader that the way forward ties not just in the development of instrumentation, nor in the development of telescopes, but rather in the effective development of both telescopes and instrumentation together. Any self-respecting telescope or instrument builder must, of course, begin with the science. Specifically, he or she should ask what astronomers really want. The answer to this question has not really changed much over the years: astronomers want an instrument-telescope combination with great sensitivity, large field of view, excellent spatial resolution, wide wavelength coverage, and a variety of available observing modes. In other words, a machine that does everything. There is a wonderful expression for this type of device in the German language, namely Eier-legende-woll-milch-sau, or "egg-laying, wool-bearing, milk-giving, pig." This simple stereotype is not entirely appropriate, however. I believe that there is a good type Of Eier-legende-woll-milch-sau, that is, an instrument-telescope combination that offers a variety of options to the observer, allowing multiple science programs, as well as rapid reaction to changing observing conditions. On the other hand, there definitely is a bad type of Eier-legende-woll-milch-sau. This is a device that attempts to do so much and is so complex that it is impossible to build, afford, and maintain.