Consuming Linked Data Tutorial
At the Consuming Linked Data tutorial I attended on Monday 25 April 2010 I heard details of a number of Linked Data applications which could be used to process the Linked Web of Data. Of particular interest to me was the sig.ma search engine. Below I discuss the implications of discovering that personal content (perhaps provided using Social Web tools) becoming surfaced in a Linked data environment through semantic search tools such as sig.ma.
sig.ma is described as a “semantic information mashup” service. I used this Web-based service for vanity searching: a search for “Brian Kelly UKOLN” provides me with an interesting view of how data about me is freely available in the Web of Linked Data. A screen shot is shown below.
The service demonstrates how information from disparate sources can be brought together in a mashup and, as such, is worth trying out to see what information the Web of Linked Data has about you.I found, for example, many blog posts which I was unaware of which referenced my work in some way such as a summary of an event in Southampton I spoke at last year;a reference to a post of mine in a post on FriendFeed: where the conversation happens and a link to one of my briefing documents in a list (in the Czech language, I think) of Library 2.0 resources. In total it seems there were 152 sources of Linked Data information about me.
This service is of interest to me not only for the information it contains but also to understand incorrect information which may be held, the reasons for such information and the risks that personal information you may not wish to be shared has already been gathered and is available in Linked Data space.
Linked Data and New Media Literacy
As can be seen in the above image two data source think that my surname is ‘UKOLN’. Further investigation reveals that this is due to the slides from a talk I gave at the Museums and the Web 2009 conference, which were uploaded by the conference organisers, having incorrect metadata.
As well as information gathered from Slideshare, sig.ma has also gathered information from Twitter, the Archimuse Web site (which organised the Museums and the Web conference), this blog and various resources I maintain on the UKOLN Web site. And on asking the service to retrieve data from additional services I discover that Linked Data about me is also available from data held on Wikipedia, YouTube, Ning, Scribd, Blip.tv, VOX, Blogspot and Tumblr as well as a number of individual’s blogs e.g. posts on Stephen Downes, Dulwichonview, Daveyp, the Openwetware and no doubt (many?) other blogs. It would appear that if you are a user of these popular Social Web services your information may be available as Linked Data.
I also noticed that sig.ma knew my date of birth. I have also tried to conceal this information from public display and was puzzled as to how it came to be known. I discovered that I had included my data of birth in a FOAF file which I created in about 2003 – before I decided to conceal this information. I have removed the data of birth from my FOAF file – but how long will it take for sig.ma to refresh this data source, I wonder?
The large amount of information about my professional activities which can be found using sigm.ma is pleasing – and it is good to see how RSS feeds, RDFa and other structured data sources which are accessible from various Social Web services is being used. But what if the information is wrong, misleading, embarrassing or is confidential? I have recently read that Syracuse University [is] to Provide Online Reputation Management to Graduates. We all need to have the skills to carry out such reputation management activities, I feel. And search engine which explore Linked Data sources should now be part of the portfolio of tools we need to understand. Have those involved in New Media Literacy appreciated this, I wonder?