My name is Matthew Turk, and I am the primary author of yt. I designed and implemented it during the course of my graduate studies working with Prof. Tom Abel at Stanford University, under the auspices of the Department of Energy through the Stanford Linear Accelerator Center and Los Alamos National Lab. It has evolved from a simple data-reader and exporter into what I believe is a fully-featured toolkit.
yt was designed to be a completely Free (as in beer and as in freedom) user-extensible framework for analyzing and visualizing adaptive mesh refinement data. It relies on no proprietary software – although it can be and has been extended to interface with proprietary software and libraries – and has been designed from the ground up to enable users to be as immersed in the data as they desire.
Originally, yt was written as a frontend to HippoDraw, an extensible and comprehensive plotting package built at SLAC by Paul Kunz. Over time it has been generalized to rely on commodity Python packages (mostly), its dependencies have been reduced, and thus its installation made significantly easier.
In some sense, yt is also designed to be rather utilitarian. By virtue of the fact that it has been written in an interpreted language, it can be somewhat slower than purely C-based analysis codes, although I believe that to be mitigated by a cleverness of algorithms and a substantially improved development time for the user in the case of a desire to expand the functionality.
The primary goal has been, and will continue to be, to present an interface to the user that enables selection and analysis of arbitrary subsets of data.
yt has evolved substantially over the time of its development. Here is a non-comprehensive list of features: