What Sections Does the Glue Documentation Cover?

The Glue documentation at docs.glueviz.org is organized into a small set of top-level sections that map directly to how people use the tool: getting started, using the graphical application, driving Glue from Python, domain-specific workflows, customization and development, and reference material. If you are looking for a specific topic, the section names below are the fastest way to locate it.

The main documentation sections

The documentation's table of contents lists these sections:

Which section to open for a given task

• You have never used Glue and want to try it. Start with Getting started. The docs describe Glue as designed for "data-hacking" workflows and note you can use the graphical application "as is, and never type a line of code."
• You want to explore data through the GUI. Go to Using the Glue application. This is where the linked statistical graphics — scatter plots, histograms, and 2D/3D images — and the brushing-and-linking paradigm are documented.
• You want to send data from Python into Glue. See Interacting with data from Python. The docs describe two routes: the IPython terminal built into the Glue application, and sending NumPy arrays or Pandas DataFrames to Glue from a Python or IPython session.
• You work in a specific field. Check Domain-specific guides first, since it may already cover your data type or workflow.
• You want to change how Glue behaves or extend it. Use Customizing/Hacking Glue and Advanced customization. The docs list config.py files, automatic data loading and cleaning before Glue starts, custom file-format parsers, custom dataset-linking functions, and custom data viewers as things you can do here.
• You want to build a plugin or embed Glue in your own application. See Developing Glue and The Glue architecture. The docs note you can develop distributable plugin packages and use the Glue framework in your own application to provide data-linking capabilities.
• You need to cite Glue or find related papers. Use Acknowledging glue and Publications.
• You want the programmatic reference. Go to API.

What the documentation says Glue is

Glue is a Python library for exploring relationships within and among related datasets. Its stated main features are linked statistical graphics (scatter plots, histograms, and 2D/3D images, focused on the brushing-and-linking paradigm where selections in any graph propagate to all others), flexible linking across data (user-specified logical links between datasets that overlay visualizations and propagate selections), and full scripting capability (written in Python on top of Numpy, Matplotlib, and Scipy, so you can integrate your own code for data input, cleaning, and analysis).

The docs also point to a comparison table for how Glue compares with other open-source and commercial data visualization solutions, and to a videos page for demos.