Spreadsheet 2.0 - User-Oriented Tools for Analyzing Complex-linked Data

Problem Statement

In today’s enterprises, not only the amount of data, but also the data’s complexity is continuously growing over time by linking data objects to each other yielding to dependencies and relations between these objects. Hence, the analysis of an enterprise’s data becomes a challenging task, which becomes even harder by taking into account the information demands of various stakeholders at different times.

In practice, stakeholders are using Spreadsheets (e.g., Microsoft Excel) in order to interactively analyze and visualize certain data sets by their own. However, prevalent Spreadsheets suffer, i.a., from the following shortcomings [Re15b]:

• Lack of transparency of a Spreadsheet’s design: In this context, the design of a Spreadsheet refers to the information of how the data cells, calculations, and visualizations depend on each other, i.e., “the flow of the data”. As a consequence, the impact of changes of data, calculations, and visualizations is not comprehensible by users.
• Lack of support for complex linked data: Spreadsheets are focusing “flat” data structures, i.e., two-dimensional table structures, whereas each table’s cell contains a simple object, e.g., a number, a date, or a string. However, the content of a cell – and hence the result of calculations – cannot be a complex linked data structure (e.g., nested trees).
• Lack of collaboration support: While typically an enterprise’s data is managed collaboratively, its analysis and visualization is done in a non-collaborative Spreadsheet. However, especially the analysis of the data could benefit from a collaborative environment, e.g., by considering the rights, responsibilities, and information demands of users on the one hand, and by accessing the data’s history on the other hand.

Ambition

Therefore, the goal of the Spreadsheet 2.0 approach is the development of innovative Spreadsheet concepts addressing the identified drawbacks of existing Spreadsheet applications, e.g., by making the information flow through the spreadsheet transparent:

The Spreadsheet 2.0 prototype is built based on the chair's SocioCortex platform, and furthermore consists of two clients, namely the Spreadsheet 2.0 DataTable as part of the SocioCortex Content Manager and the SocioCortex Visualizer. The Spreadsheet 2.0 DataTable component is an end-user-friendly and spreadsheet-like web application for the collaborative management of complex-linked data, while the Visualizer component is a flexible dashboard solution. Through the analysis of data functions and visualizations within the DataTable and Visualizer components respectively, the Spreadsheet 2.0 application can foster the transparency of the information flow from the data's origin to the consuming user.

By the implementation of a Spreadsheet 2.0 prototype, these concepts can be evaluated based on requirements gathered from an industry partner [Re15a, Re15b].

Research Design

Based on identified observed use-cases of the spreadsheets usages at the industry partner (e.g., Capacity Management, Financial Reporting, and Risk Analysis), we derive concepts addressing the shortcomings of those spreadsheets. Subsequently we implement a prototype, which will be evaluated in close cooperation with the industry partner.

SocioCortex Visualizer

There is already a first prototype of the SocioCortex Visualizer, whose goal is to enable end-users to interactively and collaboratively define and design dashboards as an assembly of multiple configurable visualizations (see here). The bindings of the individual visualizations to the respective data are defined by using the base plattform's model-based expression language (MxL) [Re14]. 

The analyzability of the data bindings and underlying data transformations enables the system to automatically generate an information flow graph, whereas its nodes represent different artifacts of the system including various meta information, e.g., roles and responsibilities of users [Wa15b].

Publications