
Marking in Spotfire
In Chapters 1, Welcome to Spotfire, and Chapter 2, It's All About the Data, we covered the basics of marking but it now warrants a more in-depth explanation since it's such a key concept in Spotfire.
We created details visualizations by right-clicking on a main visualization and selecting Create Details Visualization. This is the standard method of enabling drill-in from one visualization into another. However, behind the scenes, the drill-in capability is implemented via a concept known as marking. In other analytics tools, this is often called brushing and linking. Similar terminology may also be used. The example that follows will show you how to configure a details visualization manually by explicitly specifying the marking scheme to use.
The concept of marking is that data is marked or highlighted temporarily on one visualization (by using the mouse or touch). The same data, related data, or linked data is then shown elsewhere within the analysis. Spotfire will either simply highlight marked data on multiple visualizations, or it will restrict data on some visualizations based on data selected in another. The concept of a details visualization is the latter concept. Data that is marked on one (main) visualization is used temporarily to limit the data shown on other visualizations to the subset of data marked in the main visualization.
There are a few key rules to be aware of with marking:
- Spotfire analyses can have multiple marking schemes
- Marking can be used to set up a parent-child relationship between visualizations
- Multiple visualizations can share the same marking schemes
- Multiple marking schemes can be used to restrict data on visualizations (it's even possible to use Boolean logic—data must be in all markings or just one or the other)
- Marking data does not affect the underlying data in any way
- Marking of data can work between multiple, related data tables (note that we have not discussed data table relations yet—this topic is covered in Chapter 8, The World is Your Visualization)