Data Visualization

upon evolutionary developed domain models



Generative model driven software development

We are currently researching in the field of data visualization upon evolutionary developed domain models. This research project has been started as a consequence to our previous one, which was about evolving domain models. It is bringing us a step closer to seamless transformations in generative model driven software development with Sclable.

The data visualization research project is based on following assumption: Your domain model should not only incorporate a full technically interpretable description of your business domain – and therefore be able to generate functionality from it that allows you to actually manage your data. In fact it should also cover additional semantics for data analysis and support your business with data visualization based on domain models.


What do domain models tell us?

Sclable Business Domain Models already provide a lot of information about the analyzability of the data managed by them. The simplest examples are Sclable workflows: Your Sclable Domain Model just knows what states your entities can be in. And an entity’s default overview screen shows the instance count per state. It is a fairly small step to visualize your business data. For example, visualize an entity’s state counts as a pie chart.

Now think of additional dimensions such as task responsibilities (agendas), workflow progress (history), selective relational information (selections) and more available in Sclable Business Domain Models. Without going beyond a simple COUNT() aggregation, your Sclable Business Domain knows so much about what’s going on!


How to ask the right questions?

The idea behind the current data visualization research project is to enhance domain models to get two benefits:
Number one is to derive meaningful questions for data analysis directly from your domain model.
Number two is to simplify the task of visualizing that information correctly.

What applies to the very obvious explorable dimensions such as the “state” of your entity’s instances, applies to any other dimension as well. Let’s assume a domain model is extended by a GIS domain providing latitude/longitude points as a geographic data type. From there it doesn’t take much to visualize an entity’s state count distribution across continents, states or cities on a geographic map.

The simplest thinkable domain model already offers thousands of explorable dimensions with only a handful being crucial and useful to the business. At Sclable we feel that developing business applications upon Sclable Business Domain Models requires modelling those too.


What is needed to analyze a business domain visually?

Data analysis and visualization itself can be a complex task. Applied correctly, it can require the combined skill set of computer scientists, mathematicians, geographers, system architects, business analysts, UX designers, visualization artists. Not to speak of programmers and developers of different sorts.

We’ll explore all those areas, aiming for concepts ready to implement into the Sclable Business Application Development Platform. Just like domain modelling itself, those implementations will allow any developer to make use of data analysis and visualization capabilities, while a great part of the complexity behind it is already taken care of. In addition, as one of the mentioned experts, you should be able to extend the Platform beyond any limit. Providing additional functionality for non experts.

Our data visualization research project was kindly supported by the FFG, the Austrian Research Promotion Agency.

ffg_logo_4c

LEARN MORE ABOUT DATA VISUALIZATION AT SCLABLE

Blogpost: Domain Driven Data Visualization

Blogpost: 3 Top Business Data Visualization Challenges and how we adress them at Sclable

Did you know?

This chart is from William Playfair’s “Commercial and Political Atlas", first published 1786. It contained 43 time-series plots and one bar chart, a form apparently introduced in this work. It has been described as the first major work to contain statistical graphs.

RADICALLY RETHINKING TECHNOLOGY.