Festival2012/Submit/Creating visualizations using Linked Open Data

From MozillaWiki
Jump to: navigation, search
  • Title of session: Creating visualizations easily using Linked Open Data
  • Your name and affiliation: Alvaro Graves, Poderopedia, Tetherless World Constellation
  • Session format: Learning Lab

What will your session or activity allow people to make, learn or do?

People will learn how to use multiple sources of Linked Open Data to create visualizations that later will be used in blogs, articles, etc.

How do you see that working?

Using tools like SPARQL and LODSPeaKr people will learn how to use Linked Open Data available on the Web to create visualizations. Some familiarity of the participants with HTML, SPARQL and programming languages is highly recommended but not necessary to know all of them.

I will explain how these tools work for 40 min. or 1 hour. After that an individual/group project will put in practice what has been learned.

Since the tools used are Open Source, we expect collaboration and feedback from the audience on how to improve them and makes them easier to use for others.

How will you deal with 5, 15, 50 participants?

Every attendee will create its own set of visualizations using the tools provided and data openly available. These visualizations can later be embedded in their own blogs, personal or work home pages, etc.

For 5 people, I expect everyone to create a one or more visualizations that can be embedded in their blogs or shared via twitter. The visualization must be meaningful, in the sense that it should make a point about an issue important to the attendees. It will be encouraged to discuss, share and reuse code done by others while creating a project, to avoid people getting stuck in technical details.

For 10, 50 or more people I expect to have groups of no more than 4/5 people each that will create visualizations. The project will be to create a set of visualizations that will support a story (about economy, politics, etc.) that the group wants to tell.

Group formation

Ideally each group should consist of members with different backgrounds and different levels of technical skills. Thus we can maximize the number of areas of expertise in every group. Also, we can learn other people's point of views that we are not used to know :-)

Pre-cooked projects

For those groups who don't have people with strong technical skills and do not want to get stuck in one specific step, I will provide "pre-cooked" projects, i.e., visualizations at different levels of development that can be used for groups who do not want to start form scratch.

  • Full SPARQL queries: The queries are ready, the group will have to create a visualization from it.
  • Half-made SPARQL queries: The queries need some modification to fit their needs
  • No queries: The groups is assigned a specific visualization (and told where they can find the information) but they have to create the queries from scratch.

How long within your session before someone else can teach this?

If there are not enough people, probably 90 to 120 min aprox. With more people, we will be able to have something to show to someone else in 30-45 minutes.

What do you see as outcomes after the festival?

People will be able to install create useful visualization based on data available on the Web that can be used in their work. If there is enough time, I will show how to install the software on a Linux machine so they can have a "visualization server". Otherwise, I'll leave the documentation to do so available online.