This guide documents the different data formats used by WorldWide Telescope software, as well as some of the tools used to generate or manipulate them.

This guide synthesizes several manuals that were written separately. Cross-references are still being updated.

The WWT ecosystem involves many more data formats and standards than you might think. Links that are internal to this guide are marked in bold.

  • Individual color images that can be placed upon the sky if they are tagged with AVM metadata. TODO: cross-reference to Layers Guide.
  • Scientific image data in the FITS format, whose visualization properties can be manipulated interactively. The most straightforward way to load and visualize FITS data with WWT is in the JupyterLab scientific computing environment with pywwt.
  • In order to efficiently display large images, they must be broken into pieces and downsampled, a process referred to as “tiling”. This can by done with the Python package toasty or the Windows programs known as the Study Chopper or the Sphere Toaster. Either toasty or the Windows Tile Pyramid SDK allow you to write custom software to drive the process, which is generally needed when the input images are gigabytes or larger in size.
  • In order to display images that cover most or all of the sky, one also must be careful about how to map between the spherical shape of the sky and the flat, 2D nature of computer imagery. The mathematics behind this mapping are described in the Spherical Projections section. The toasty and Sphere Toaster software packages implement the necessary mathematics.
  • The technology that WWT uses to manage all-sky imagery can also be used to process 360° panoramas, as described later in this guide.
  • Information about all of these sorts of data must be exchanged in order for WWT software to know how to locate and properly display the underlying data sets. The foundational way to exchange WWT metadata is WTML, an XML-based format for describing data collections. WTML files can be created and manipulated using the low-level wwt_data_formats Python package.
  • WWT uses additional specialized XML data formats that have not yet been documented, including:
    • The wtt format for serializing WWT Tours.
    • The wwtl format for serializing groups of layers.
    • The “File Cabinet” format for bundling up a group of files.