deletions | additions       diff --git a/untitled.tex b/untitled.tex  index 3c930e2..1401514 100644  --- a/untitled.tex  +++ b/untitled.tex 
...
\textit{\textbf{K.W. Willett et al.}}  The Galaxy Zoo project uses crowdsourced visual classifications to create robust, large catalogs of detailed galaxy morphology. We present initial results for the Galaxy Zoo: Hubble dataset, which includes 169,944 images of galaxies selected from the AEGIS, COSMOS, GEMS, and GOODS surveys. The galaxies span a redshift range of $0includes distinction distinguishes  between bulge-dominated and disk-dominated galaxies, as well as kpc-scale features including galactic bars, spiral arms, irregular structure, and mergers. The data also measures the geometry and orientation of clumpy structures unique to high-redshift star-forming galaxies. Visual classifications are calibrated using $z\simeq0.05$ SDSS images that are processed to appear at a variety of simulated redshifts; this measures the bias in feature classifications for galaxies with changing apparent size and brightness, without overcorrecting for effects such as an evolving $L^\star$. We present a new technique for debiasing the morphologies; we demonstrate its effectiveness for bulge/disk separation, and discuss its limitations for correcting detailed structures due to insufficient sampling of the simulated redshift space. All the above data will be included in the upcoming release of the full Galaxy Zoo: Hubble catalog.