The advantages of the tabular layout are that it is familiar, flexible, easily configurable, and excellent for interactive comparison tasks. These advantages are evident in numerical spreadsheets, and translate easily into visualization spreadsheets. Users can construct their own configurations in situations that programmers cannot foresee. Because users are familiar with tables, they can immediately start organizing their data in this spreadsheet metaphor. This flexibility is what contributed to the success of the numerical spreadsheets. It can be tailored to multiple situations in a single tool that is both easy to understand, as well as easy to configure.
For example, comparison tasks are commonplace in numerical spreadsheets, therefore these tasks are easily supported by the visualization spreadsheet. For easy comparison in numerical spreadsheets, users often put two numbers next to each other or load two sets of numbers into adjacent columns. Similarly, in the visualization spreadsheet, users layout two datasets next to each other, or compare two groups of data using adjacent columns. By allowing users to enter data into cells in various configurations, the spreadsheet supports a variety of different tasks. In the following examples, we show how this technique can be adapted in a visualization spreadsheet to show different datasets, different visual representations, and several time steps in the time dimension.
The tabular organization of the spreadsheet enables the user to immediately detect differences between the visualizations of different datasets. For example, even viewers without experience with molecular biology can see that the general structure of the datasets in Figure 1 are similar, but that some alignments that are present in cells A2 and A3 do not appear in A1. Users can now take advantage of their visual comparison abilities to detect differences between datasets.
Users of the spreadsheet can also use it to compare different visual representations. In Figure 2, the tabular layout is used to show different visual representations in different columns. Across each row, the values in the cells are the same. The visual representation in each cell of a row has been changed to bring out different features of the dataset.
The columns and rows of the table increase the number of dimensions we can see simultaneously. In Figure 3, the columns show several snapshots of the steps of the 3D Delaunay algorithm. So in this case, the columns are used to represent the time dimension.
As the above examples show, the tabular layout is one of the reasons why spreadsheet-based environments are so powerful. The organization is familiar to users, and simple direct manipulation operations can be used to rotate contents in the cells. It can be custom tailored to individual situations on-the-fly.