Monday, July 21, 2008

Data Visualization--new, but adaptive?

What makes a good website? We all know what a bad website looks like, but why is it so difficult to define what makes a good one? Because a good website must fit what the mission, scope, objectives, or users need in a website—which is completely different from site to site. There is a fine art to creating a website that is intuitive for users and includes everything essential to the company and the web-surfer.

Data visualization is the new tool web design companies such as Stamen have been using to create innovative and interactive web sites for large corporate companies. Check out some of the companies working with Stamen in this news article: http://www.businessweek.com/innovate/content/jun2008/id20080623_831514.htm?campaign_id=rss_tech.

The idea behind virtual data is that workgroups and end-users can have access to shared data and can independently work on large projects using one or several shared data sets. Data sharing has been growing in popularity with open web projects such as DISC-UK at http://www.disc-uk.org/datashare.html where researchers can share open access data sets to have other researchers use the same datasets to create new studies, work with the data in different ways, or peer review other researchers. More than just researchers will benefit from open data sources—as data sets grow, they will provide a broader picture for researchers which will then translate into better programs or solutions for the general public.

The Stamen group is working on innovative web design for a more commercial market. They want to create interactive and customizable data maps or other collections of data for the general public to use and manipulate. “It's just one example of Stamen's attempts to nudge Web site visitors to explore data on their own, interactive terms. Rodenbeck [the director of Stamen] calls it "exploratory navigation." The hope is it will encourage users to take an intuitive, adventurous approach to finding information, rather than following a prescribed path.” A current project they are working on includes mapping incidents of crime over a real estate company’s property listing map so that both realtors and house-hunters can “investigate” the crime rate in area neighborhoods. Other projects include visualizations in Digg websites that highlight news stories that are more popular or more read by other online news readers. Users manipulating data encourages them to explore data relationships in new ways.

Librarians are always encouraging user manipulation of data—in essence, this is at the heart of what librarians are after—creating access to information so that individual users can manipulate it and explore. One thing librarians consider which may not be addressed by commercial situations is adaptive technology. It is great that users can see interactive, 3D worlds of information, but is it accessible to people using assistive technology? Can a screen reader tell that one area of a map is 10x more likely to have a murder or other violent crime? Do screen readers differentiate between the red headline which is more popular than the blue headline at the bottom? Data visualization is an amazing new way of exploring information, but lets not leave the disadvantaged in the dust. A great website would be accessible to all users.

1 comment:

DataShare said...

Hi Claire - just to clarify DISC-UK DataShare - we're a JISC-funded project looking at technical, cultural and legal issues surrounding 'orphaned' research data sharing in the UK tertiary education community with a view to investigating tools, methodologies and mechanisms that will enable researchers to share data over the web in an institutional repository environment. One of the areas that I'm particularly interested in is 'open' data visualization (both numeric and spatial) using such tools as Swivel, Numbrary and Many Eyes (numeric) and the open geobrowsers such as Open Layers, Google Earth, NASA World Wind and how they can be utilised and explored in research departments and organisations. I've part 2 of a briefing paper dealing with Spatial Data Visualization in Web 2.0 Environments out this week or next. Have a look if you get the chance and feel free to provide comment.

Best wishes
stuart macdonald