Archive for November, 2010

Richard Wallis (at lower left) presents a map of the open data systems available now. The next version of this chart will likely fill up an entire page!

I sat in on a session of the track on exploiting linked and open data.

Richard Wallis, Technology Evangelist, Talis, UK began with a presentation entitled “Demystifying open and linked data”, which reviewed government open data projects.   Governments have taken the lead in opening their data and linking it together.  For example, the UK government’s data site, appeared early on.  OpenlyLocal aggregates data from local governments.

The first step in opening data is to collect it and get it into an application.   The collection activity is frequently spurred by directives from government ministers to open the data.  Formatting and preparation may involve considerable effort because of varying data formats.  Talis is helping local governments in this effort and provides a “toolkit” which helps them to open their data.  It provides data aggregation, ontology creation, and visualization tools and others, all constructed from open source software.

Chris Taggart, developer of OpenlyLocal, has been opening up local government information since 2009; OpenlyLocal now has data from150 councils.  He wondered if we still need to answer the question, “Why open data?” and concluded that we do because all aspects of our lives are affected by data, and we need information when we want it, where, and in the format in which we want it.  A data democracy allows people to look at the same data and arrive at different perspectives on it.  Open data is relevance, and by opening up their data governments increase their relevance.  Users bring a fresh approach to the data.  But what if nobody uses the data?  This is inevitable for some data, but even this is a useful feedback mechanism.

Taggart discussed several problems with open data, including proprietary data, messy data, perceived threats by producers, and personal information in data sets (probably the most difficult problem).

Some lessons learned:

• How you approach open data is critical.
• Give people what they ask for.
• As data moves to the web, it takes on the web’s characteristics: incremental change, connections, etc.

Don Hawkins
Columnist, Information Today and Conference Circuit Blog Editor

Katherine ("Kat") Allen (L) and Val Skelton (R)

Here are my Information Today Europe colleagues who will be bringing you news from the conference on their blog, with a European slant.  You can see their latest posts at right.

Don Hawkins
Columnist, Information Today and Conference Circuit Blog Editor

Stephen Dale (L) and Dion Hinchcliffe (R), Opening Keynote Speaker

Lorna Candy, Conference Manager, opens the conference

In his opening keynote address, Dion Hinchcliffe, Vice President, Dachis Group, said that this year’s hot topic is social media, analytics are the future of social computing.  Social is how we function today, and it is now the dominant form of communication.  About 850 million people are using social computing today–more than use e-mail.  This is the fastest migration we have ever seen, and its aftereffects are still being felt.We have always had more information that we can handle, and we are creating even more every month.  Social tools make information abundantly observable.  Static portals and intranets are changing to social environments to deal with that.

Your influence in the world will be determined by the kind of social capital you have built up.  Social capital is tied to us as individuals, not to organizations.  There is no good way to build capital around organizations–it’s the aggregate of all an organization’s employees.  The Dachis Group has recently acquired the 2.0 Adoption Council, a collection of managers in large enterprises working on Web 2.0 adoption, and will study the effect of social media in large organizations.  Knowledge is increasingly visible in social channels.  It no longer “evaporates” or is hidden, and we now realize it is part of a single continuum and the one consistent way we consume information and collaborate.

Social Business Power Map

Hinchcliffe has developed a social network map (see photo above), which has 4 stages:  buzz, experimentation, adoption, and maturity.  The goal is collective intelligence:  getting value from what’s observable.   But is all this information valuable? It appears that it is, at least enterprise channels. Some of the issues with observed information are:

• Knowledge Workers spend 20% of their time looking for info needed to do their jobs.
• 42% of the economies of developed nations is tacit interactions
• Organizations that adopt social tools widely see the amount of observable work skyrocket, which becomes a management and search concern.
• Between 80-90% of the information that organizations have collected over the last 30 years is inaccessible by most workers.

We must recognize that the web is very different from the enterprise.  We have tools to find the needle, but analytics let us see the shape of the haystack, understand what we have, and know what it means. We don’t need to directly see all the information, but we need to see the “shadows” it makes.  According to Hinchcliffe, many organizations are using “shadow IT”: informal or unauthorized channels.  The new flow of business will be to put engagement in context.

How will we listen in the future?  Strategic engagement tools now exist but are in their infancy.  They focus primarily on the outside world, favor new social environments over older style and vertical communities.  they don’t connect well to existing reporting tools and data warehouses.  They are relatively expensive, and exist where you don’t expect them.

Most tools are focused on listening; analytics are harder to resolve.  We need tools to understand network graphs, power laws, information architecture.  Unfortunately, real results always seem to entail taxonomy and classification, which has been one of the very few classically stubborn problems on the web.  Social analytics are almost as difficult and require insight into people–it’s about sociology as well as networks.  Unlike listening, real analytic tools are hard to find.  One example analytic system is nodeXL.  It’s more than just social network visualization.  We need to understand what we have.

Social networks are being used to solve real problems. For example, KatrinaList was widely used to help survivors find each other.  Similar uses were made of social systems in the San Diego fires and Pakistani floods.

In the future, almost all social computing will be mobile.  According to a recent report, most computing devices sold next year will be mobile.  Digital, social, and mobile are all part of the same strategy.

Analytics are the future of social computing.  The Google of social computing will arrive–it will be free multichannel, and will allow us to see more of the shadows than ever before.  There will continue to many niche tools but the social landscape will ultimately look like the search industry does today.  We will know we have arrived when social analytics tell us why, not just what.

Don Hawkins
Columnist, Information Today and Conference Circuit Blog Editor

Roger Bilboul (L) and John Bryans (R) set up the Information Today Booth

The Exhibit Hall is Ready

On Monday, exhibitors were busy unpacking and getting ready for Tuesday’s opening.  Here are Roger Bilboul, Information Today Chairman, and John Bryans, Publisher, setting up the ITI booth.

Don Hawkins
Columnist, Information Today and Conference Circuit Blog Editor

Look at the notice on the bottom of this sign.  Londoners had to use creative means of travel yesterday.  I saw a headline in the paper, “Prepare for Chaos!”

Don Hawkins
Columnist, Information Today and Conference Circuit Blog Editor