In the News Focus article “A singular conundrum: How odd is our universe?” (28 September 2007, p. [1848][1]), A. Cho perpetuated misunderstanding of science with the statement, in part from James Gunn, that “'Cosmology may look like a science, but it isn't a science' because it's impossible to do repeatable experiments.” In the truly natural sciences (such as geology, oceanography, atmospheric science, and ecology), rigorous observation and interpretation are commonly used, rather than “repeatable experiments” a la Karl Popper—except in those few cases where a small-scale experiment is meaningful ([1][2]). It would be better to say that cosmology is science—it just isn't Popperian physics. 1. 1.[↵][3]1. C. M. Condit, 2. L. B. Railsback , The Transilience Project [www.gly.uga.edu/railsback/Transilience/Transilience.html][4]). # Response {#article-title-2} With all due respect, I think that L. B. Railsback misses the point that I was attempting to make. The pursuit of understanding the cosmos is certainly a scientific pursuit and makes use of many of the most powerful tools of science. Unfortunately, there is only one observable universe, and while it is quite possible in principle, and probably in practice, to formulate theories that describe its observed behavior perfectly on the largest scales, those theories could well be unverifiable by any doable experiment. In geophysics and astrophysics, the experimenter is nature, not the scientist, but repeated experiments can be done and the results can be observed. This is not so in cosmology for phenomena on the largest scales. Further confusion stems from our belief that the structure we are observing is stochastic on scales up to and beyond the current particle horizon. As discussed in the News Focus article, we may be unlucky enough to live in a volume in which some large-scale quantity assumes a very unlikely value within the framework of some otherwise seemingly successful theory. It then becomes a very subjective matter of whether this observation does or does not rule out the theory in question. Whatever one's view on the Popperian definition, verification by whatever technique is a cornerstone of science; I am merely saying that this can be impossible for crucial and interesting aspects of cosmological inquiry. [1]: /lookup/doi/10.1126/science.317.5846.1848 [2]: #ref-1 [3]: #xref-ref-1-1 View reference 1. in text [4]: http://www.gly.uga.edu/railsback/Transilience/Transilience.html