UK Web Focus

Innovation and best practices for the Web

Archive for April 28th, 2010

Linked Data and Lessons From the Past

Posted by Brian Kelly on 28 April 2010

The Buzz at WWW 2010

I’m current in Raleigh, North Carolina attending the WWW 2010 conference. And the buzz at the conference so far seems to focus on Linked Data (although I should add that I am writing this during the two days of pre-conference events and I have been out socialising with Linked Web developers, so perhaps these conclusions are quite subjective and premature!).

This excitement reminds me of previous WWW conferences – and makes me reflect on the extent to which the passion felt be many developers and researchers at these conferences actually results in significant changes in the Web landscape in the short term or whether the enthusiasms simply result in a failure to engage the mainstream community and a failure to address a bigger picture. So here’s my reflections of the excitement I felt after the WWW 2003 conference.

Reflections on WWW 2003

I remember returning from the WWW 2003 feeling inspired. In part this was after seeing how a communications infrastructure (WiFi and IRC) could be used to support a conference – this was my very first ‘amplified conference’ (although Lorcan Dempsey hadn’t coined that term at the time). The interest in this approach was described in an article by Paul Shabajee entitled “‘Hot’ or Not? Welcome to real-time peer review” published in The Time Higher educational Supplement. So inspired was I by the potential I felt that use of WiFi technologies to allow event participants to engage more actively in discussions that I wrote a paper entitled “Using Networked Technologies To Support Conferences” together with Paul and my colleague Emma Tonkin.

But despite my interest in this area, the topic that really inspired me at the conference in Budapest was the Semantic Web. The Semantic Web was not new to me – but instead having to listen to talks about low level protocol issues for the first time people were talking about – and, more importantly, demonstrating Semantic Web applications. The buzz at the conference, especially amongst a group of people I knew from about Bath and Bristol, focussed on FOAF – the Friend of a Friend vocabulary and associated applications developed initially by Dan Brickley and Libby Miller who then worked at ILRT at the University of Bristol.

So inspired was I by this lightweight approach to what subsequently became commonly referred to as social networking software that shortly afterwards I created my own FOAF file. And a few months after the WWW 2003 conference Dave Beckett (a Semantic Web researcher based, at the time, at the University of Kent) and myself gave a plenary talk on “Semantic Web Technologies for UK HE and FE Institutions” at UKOLN’s IWMW 2003 event – raising awareness of the potential of the Semantic Web to members of institutional Web teams in UK universities.

The following year myself and Leigh Dodds (a Semantic Web developer who then worked at Ingenta in Bath) explored ways in which we could seek to engage a wider community in this early example of a Semantic Web application. In a paper on “Using FOAF To Support Community Building” we described lightweight FOAF authoring tools which could be used to create FOAF files and viewers which could provide tangible evidence of the benefits. These tools were promoted to participants via a resource on Use of FOAF which was promoted at the IWMW 2004 event which summarised the potential of FOAF and described tools for creating and viewing FOAF (e.g. FOAFnaut, FOAF Explorer and Plink) and possible concerns people may have with this technology.

What Happened?

What happened after the identification of the new big idea at WWW 2003 was followed up by talks to the Web management community by a respected Semantic Web developer and the provision of simple authoring and viewing tools and a context for use provided? The answer was ‘not much’. A few people created their own FOAF files, but most seemed to have no interest, and this lack of interest continued despite the Use of FOAF being encouraged the following year at IWMW 2005, But it was quite clear that FOAF had failed to take off within this community. And in November 2005 I gave a talk on “Lessons Learnt From FOAF: A Bottom-Up Approach To Social Networks” which “describe[d] some of FOAF’s apparent failings to live up to its initial potential and discuss possible reasons for this“.

Lessons

In 2005 I was speculating on FOAF’s ‘apparent’ failure to fulfil the excitement I felt in 2003. The reasons I gave included people’s concerns regarding privacy, concerns regarding the term ‘friend’ and the perception that maybe the marketplace to legitimise the area.

But if we move on a few years we find that many people are now ready to share information on Facebook and ‘befriend’ people, even those they may have not met.

For me this example illustrates that back in the early to mid ‘noughties’ there was too much of a focus on the underlying technologies (how the Semantic Web would be build) and a failure to understand whether user’s real needs and requirements were being addressed.

What Next?

Despite the efforts of some researchers who are currently attempting to put a damper on my enthusiasms for Linked Data (:-) the failures of the Semantic Web to deliver on its initial promises shouldn’t be regarded as a reason to be sceptical regarding the promise of Linked Data today.

Rather we need to welcome Critical Friends who are willing to provide constructive criticisms on questionable claims of Linked Data and to help identify appropriate areas for deploying Linked Data approaches; the deployment models; the skills and other resources which are needed and the associated risks.

As well as the critical appraisal, which is particularly important at a time in which investment in the public sector is decreasing, we will also, however, need to continue the advocacy in order to ensure that the benefits of Linked Data are not being ignored. I will be publishing posts on the relevance of Linked Data for individuals and institutions shortly.

Posted in Linked Data | 2 Comments »