Published: 2012-11-20
ISBN: 978-91-7519-723-4
ISSN: 1650-3686 (print), 1650-3740 (online)
Most information practitioners will have implemented an interactive bar chart at some point in their career. In a large enterprise like IBM; we have hundreds of bar chart implementations available; across a number of different platforms. This represents a lot of duplicated work; and a potential maintenance nightmare. Ensuring correct and consistent design across the enterprise becomes much harder as a result. Adding to this problem is that our customers sometimes require new visualization types to be added to our products; or want to existing types tweaked to suit their needs.
In this talk I will talk about some of the practical problems we face in this area. I will also present an IBM rendering and mapping framework that represents a solution to some of these problems. By abstracting two key phases of the traditional information visualization pipeline; we can flexibly insert new mappings into products and facilitate deployment of previously created visualizations across different target platforms. I will outline some of the high level design constraints and present samples where possible.
