(Abridged) Typically large telescope construction and operation costs scale up faster than their collecting area. This slows scientific progress, making it expensive and complicated to increase telescope size. A metric that represents the capability of an imaging survey telescopes, and that captures a wide range of science objectives, is the telescope grasp -- the amount of volume of space in which a standard candle is detectable per unit time. We provide an analytic expression for the grasp, and also show that in the background-dominated noise limit, the optimal exposure time is three times the dead time. We introduce a related metric we call the information-content grasp, which summarizes the variance of all sources observed by the telescope per unit time. For seeing-dominated sky surveys, in terms of grasp, etendue, or collecting-area optimization, recent technological advancements make it more cost effective to construct multiple small telescopes rather than a single large telescope with a similar grasp or etendue. Among these key advancements are the availability of large-format back-side illuminated CMOS detectors with <4 micron pixels, well suited to sample standard seeing conditions given typical focal lengths of small fast telescopes. We also discuss the possible use of multiple small telescopes for spectroscopy. We argue that if all the obstacles to implementing cost-effective wide-field imaging and multi-object spectrographs using multiple small telescopes are removed, then the motivation to build new single large-aperture (>1m) visible-light telescopes which are seeing-dominated, will be weakened. These ideas have led to the concept of the, currently under construction, Large-Array Survey Telescope (LAST).