Keynote: Digital Innovations #ILI2011

Friday’s keynote was split into two. The broad topic of both talks was Harnessing the Power of Technology and Social Media. First up was Kevin Anderson talking about overcoming information overload.

Kevin started his talk with a public information announcement, offering support to sufferers of IOS – Information Overload Syndrome. He played us this video to help us identify the signs among our friends and peers:

We have moved from a world where information is scarce to one where it is abundant. This abundance of information leads people to shut down and the psychological impact of that is that people can no longer process complex stories, e.g. about the Eurozone crisis. They turn instead to simple stories e.g. about celebrities.

To counteract this we need to move from mass media to relevant media. The mechanisms we can use to do this are the application of filters and the development of context.

Filters
In early days of the Internet we went to Google, or other search engines, to solve the problem of finding information. Now we go to Twitter and other social networks for filtered information from networks that we have created.

Context
Linked Open Data can be used to apply context and related information to information. Kevin gave examples of this:

• Gnosis – a Firefox plugin for finding more relevant related information
• Poligraft – for analysing political news stories to identify campaign contributions.

Following this came Suw Charman-Anderson with some tips for tapping into the wisdom of the crowds.

In her talk Suw guided us through the what and why of crowdsourcing. I think most of us use the term relating to small scale things, like finding the answer to a question or looking for examples of something using our online networks. Suw’s examples however are large scale and I think they wowed the crowd.

There are two types of tasks that can be crowdsourced:

1. Computational tasks – using spare computer processing time e.g. SETI at Home.
2. Human tasks – brain driven easy for humans but difficult for computers e.g. recognising and describing things. Does not rely on 100% success by individuals. Using people’s spare time.

Examples

• Your Paintings – users tagging oil paintings to generate metadata
• Old Weather – transcription of ship’s logs and identification of weather and genealogical data. Creating a semantically rich database.
• Ancient Lives – transcription of fragments of papyrus identifying greek letters for later translation.
• Herbaria@Home – transcribing specimen sheets
• What’s the score
• NYPL what’s on the menu?

Why crowdsource?

• Complete large, expensive tasks quickly and cost-effectively.
• Involve and engage the public with your collections – important for researchers.
• Build a community
• Opportunities for education – don’t require existing knowledge but you can’t help learning about the topic of the project as you take part.
• Improving accessibility through the addition of metadata

Is it successful?
Yes – see GalaxyZoo and FoldIt.