Session B301 - Mashups & Data Visualizations: New Breed of Web Applications October 31, 2007

Darlene Fichter, Head, Indigenous Studies Portal, University of Saskatchewan

Web 1.0

• realm of the HTML coders

Web 2.0

• don’t need to know HTML
• citizen journalists, stock photo sites
• DIY Programming

What is a mashup?

• uses content from more than one source to create a new service.
• uses an API or RSS feed
• sent in an XML stream

Mashup Ecosystem

• Open Data
• Open set of services and applications (APIs)
• Us

The library could be one of the sources of content, as well as the place that creates the mashups that use content.

• Programmableweb.com is a great website for sources of mashups.
• Mapping, photos and shopping mashups are really hot right now.

Implications

• Fastest growing ecosystem - Most of the sites we’re going to right now are mashups
• Questions of authority and province - where is the data from? Is it reliable?
• Don’t have to ask anyone’s permission to use APIs

Mashup Tools

• Yahoo Maps
• Frappr
• Google Maps
• Yahoo Pipes
• Popfly

Google Maps

• Chocomap.com
• McMaster Aerial Photos (GIS Data)
• Western Springs History

Yahoo Pipes

• A little more complicated than some of the other mashup tools
• Except that it uses types of advanced database queries - boolean terms, etc. which librarians are pretty good at, so it’s not impossible to use
• Cambridge Public Library - top book covers

Unintended Consequences

• Garbage In, Garbage Out. If you don’t know the real source of the data
• Client side scripts that modify the pages that you look at.

These days there are so many information rich data visualizations in newspapers and magazines. There are some really good tools out there to help you do that kind of thing.

Tools:

• Web site structure
• Newsmaps
• Facets as Elastic Lists

Social Sites for Data Visualization

• Important new medium
• an individual should get value from their contribution
• these contributions should provide value to peers
• the organization that hosts the service should derive aggregate value and be able to expose this back to the users

Tools

• Many Eyes - from IBM, will analyze text and display tag clouds. View and discuss visualizations, view and discuss datasets, create visualizations from existing datasets, upload your own data, topic hubs, select items to watch, track contributions, see comments
• Swivel - tasty data goodies. Shows you teh source, gives lots of graphs. Very cool.
• LivePlasma - shows relationships between actors, directors and movies
• Gapminder / Trendalyzer - bought by Google

This was an amazing session - I learned a LOT about some really cool tools for data visualization and mashups. I’m really excited to explore some of these tools and see what could be useful for my library.