Once we have an effective definition of abstract information, it simply remains to specify structures for presenting that data on human interfaces, including interactive web pages and windows, and various types of documents and reports. It should be noted that most data processing is associated with objects (LOD’s), the structural representation of data. There is no processing at all associated with reports, which simply place object data on a formatted document. Processing often associated with reports, such as derived data, is defined with respect to the object and is simply formatted on the report. In a manner similar to the W3C standard, XSLT and XSL FO, the structure of the data on an AOA report is driven from the object being formatted. The result of putting processing normally associated with reports in the object, simplifies that processing, but more importantly, increases its reusability by other interfaces.
There is significant work flow processing associated with interactive pages and windows, however. This processing controls the flow of pages or windows, activates and commits object data associated with that flow and often creates temporary data structures for the presentation of data. Again, however, the complex transformation of object data is typically organized with the objects involved to increase reusability.
With most data processing relegated to objects, the main task of creating human interfaces is the layout of data on an interface. Though the actual medium through which data is presented differs dramatically between windows, pages and reports and though there are significant differences in design between the three, the actual process of specifying the data components is much the same. In all cases, we have the much the same components: literal prompts, variable text, lists, tables, graphics, combo boxes, edit boxes buttons and the like, with only the interactive interfaces using the latter input or response components. In the top two diagrams to the right, we have a sample of a painted window and a painted web page. The third diagram shows a sample mapping window for either a window or a web page. The fourth diagram shows a painted report, along with a mapping window on top of it.
Also, in all cases, the components presenting or inputting variable data, all work off the principle that the data source is always an object and is defined by a combination of the three object data identifiers, 1) the object, 2) the entity and 3) the attribute. This greatly simplifies the definition of interfaces and creates a consistency when building interfaces across all environments. Note that this is true, whether the final, generated system ends up as JSP, HTML, XSLT or a group of internal MS Windows controls.