Lesson 3: Working with elements using the Outliner frame

Concepts, Attributes, & Elements

A concept in Visage is a uniquely-named, persistant "object" (often contained in a database). Concepts have attributes ; a single concept may have one or many attributes. Each attribute has a single value.

For example, in the "cities" database, there are 600 or so "city" concepts. Each concept has a number of attributes such as "name", "state" or "population." So one "city" concept has attributes with values "Pittsburgh", "PA", and "1,000,000." Another "city" concept has attributes with values "New York", "NY", and "5,000,000."

[ insert diagram -- TBD ]

Elements are the visual representations of concepts. The type of element rendered depends on the frame that contains in. Not all attributes are rendered in all frames (for example, "crime rate" may not be illustrated in a Map frame). The city concept "Pittsburgh" might be displayed as a text string in the Outliner, or it might be displayed as an icon in a Map frame. In either case, what you see on the screen are is an element; the visual representation of an underlying concept.

Elements are "lightweight" objects. Creating multiple "Pittsburgh" elements in an Outliner frame does not cause a new city concept to be created in the "cities" database; it simply creates another instantiation of the concept. Similarly, removing an element does not cause it to be removed from the database. It simply removes the visual representation of the concept from your Visage workspace.

This ability to easily create and remove elements provides an easy way to look at data in a manner most suitable for a task. For example, you can use an Outliner frame to view hierarchical data in an outline form. You might use the Map frame to locate a "city" element on a map of the US. Or you might use a Plot frame to compare the populations of a number of cities.

Element Relations and Drill Down

Visage is also capable of showing relations between concepts. These "relations" are typically derived from the underlying data and can be explored in a number of ways. One of the most powerful features of Visage is the ability to "drill down" and view the relations within specific data sets. This is the most fundamental act of data exploration in Visage.

To drill down on a data element, move the mouse pointer over the element, hold the right mouse button down, and drag the mouse pointer to the subordinate unit that you would like displayed.

1. If you don't have an Outliner frame open with the "cities" element visible, then shift-drag an Outliner out of the Frames palette and drag the DB Root element from the Tools palette to the left side of the Outliner (see the Moving section in Lesson 2 for a review).
2. Hold the right mouse button down the "cities" element in the Outliner frame to see what dimensions are available to view.
   

Those items listed with a ">" or "<" are relations. The "cities" element has no attributes of its own; instead, there are a number of relations with concepts including "City", "Metro-Area" and "State". The arrow represents the origin of the relation. Selecting one of these options will drill-down or show new elements that are related to the original element in terms of the selected dimension.

For example, you can select ">State", and Visage will display elements representing the concepts in the "cities" database that are related to the "cities" element by the relation "State": a list of all 50 states.

1. Right-mouse down on the "cities" element and drag the mouse to the ">State" relation and then release the mouse button. Visage will display new elements that are related to the "cities" database element by relation "State."
   

1. Drill down on AL by right-clicking the "AL" element. This lets you view the "AL" element's subordinate units.
   

1. Select "State_City >" Visage will display new elements corresponding to the cities in Alabama.
   

Viewing Attributes

The Outliner frame has been designed to let you view attributes associated with each element. As you might expect, the type of attribute that you can view depends entirely on what attributes are available for each concept. "City" concepts have different attributes than "state" concepts.

Note: In the "cities" database, a "city" concept may have an attribute "state" that is a different entity than the "state" concept that corresponds to the attribute. In other words, there may be a "state" attribute for the "city" concept "Pittsburgh" that has value "Pennsylvania." However, there is also a "state" concept "Pennsylvania" that has its own attributes such as "land area."

The Outliner displays element attributes in a table form, with elements down the left side of the frame and attributes displayed in columns. If an element does not have the attribute displayed in a particular column, the Outliner will display "--" in the related table cell.

To choose an attribute for a column, move the mouse pointer over the <None> marker in the Outliner frame and hold the right mouse button down. You can then select one of the displayed attributes.

1. Right-mouse down on the <None> marker in the Outliner frameand choose "Population." Release the mouse.
   

Note that Visage displays population figures for the elements representing cities, but not for the elements representing states. This is because a state's population includes more than just the population of the major cities in the database. (Of course, population could be an attribute for a state concept as it is for the cities, but this sample database has not been designed that way.)

1. Right-mouse down on the <None> marker and choose "latitude."
   

Note that the value for latitude is not particularly helpful to look at in the Outliner frame (it is the city's latitude in seconds). However, this particular attribute will be much more useful when we view the same concept in the Map frame.

Removing elements from view

There are two ways to remove elements from the Outliner frame. You can drag elements to the "Remove" button in the Tools frame and "drop" them there.

Alternatively, you can remove whole groups of elements by double clicking on their "parent" element. Remember that elements are "lightweight" representations of concepts. Thus removing an element is not equivalent to deleting a concept from a database.

1. Drag the "Birmingham" city element to the "Remove" button in the Tools frame. Notice that the Outliner removes "Birmingham and its attribute listings from view.
2. Using the left mouse button, double-click the "AL" state element. Notice that all of the cities are removed from view.
   
   Note: in previous versions of Visage, this mechanism was referred to as "hiding." This is incorrect, as it implies that there is a quick operation to make these elements visible again. Rather, you'll need to go through the same steps to view these elements as you did originally.
3. Now double-click on the main "cities" element at the top of the outliner. Notice that all ofthe states are removed from view.

Summary

In this section, you learned that Visage renders the attributes of concepts on screen as elements. Since elements are simply representations of concepts, they are easily created and removed.

In the next section, we'll show how Visage's aggregation feature lets you compose your own elements by easily combining elements.