Schneiderman-Information Visualization

Annotated Bibliography
Schneiderman discusses the types of information visualization and breaks them down into categories based on the number of variables involved and the types of data. These include 1,2,3 and multidimensional data types as well as temporal, network and tree. The author also discusses the main tasks that are involved in data visualization which are Overview, zoom, filter, provide details-on-demand, showing relationships and history and allowing for extraction. Different examples of graphs are shown and the author discusses the purpose and usefulness of each type in different situations. The author says that the advantage of visualization is that humans are well equipped to process visual information. They can use graphs and visualizations to understand relationships between data points at a glance. The author also emphasizes use control when it comes to asking for details and viewing history or undoing history and exporting the information to be used elsewhere. Schneiderman describes some of the challenges of information-visualization. These include how to organize data so that the input is correct, how to combine visual and textual labels, how to allow the user to access deeper or related information, how to view large volumes of data, and how to integrate data mining. Other challenges are how to aid collaboration and achieve usability for a diverse group of users.

My Thoughts

What are excentric labels? They are mentioned on page 598 in the context of mapmakers and user-controlled approaches. Are they like pop up help?

In the context of building exhibits in Sony Wonder Technology Lab (SWTL) an interactive technology museum, information-visualization concepts are very important. The users are diverse not only in age but also in nationality and the educational concepts need to be clearly explained.

In the signal station exhibit, students are supposed to learn what a pixel is. The exhibit automatically uses a call out box to zoom in on a picture to show a pixel. It seems that it might be more effective for the user to zoom in to experience the details for themselves so they are better oriented in space. Although progressive refinement is meant to describe information refinement in this article. Allowing for progressive refinement of the picture zooming into a pixel in this case could teach the lesson more effectively.

Information-vizualization challenges including the question of how to input data makes me think that some form of standardization would greatly benefit science and other fields. There is probably a lot of research that has been done or connections that have been made that have fallen into disuse or been forgotten over time. If there were a standard that allowed information-visualization systems to make connections between discoveries and knowledge from disparate places and even languages we could probably make some great discoveries. I guess you could call it data mining, but the data needs to become more uniform so it can be mined more efficiently.