Social Science Week: Using ICT to Improve the Visibility of Research
Created by Catherine Howell (University of Cambridge) on March 20, 2006
Last Friday, I attended a London seminar on "New Technologies for the Development of Research Knowledge in Education", part of the UK's Social Science Week events programme. Sponsored by the Economic and Social Research Council, this week is a key outreach opportunity for education and social science researchers, and represents a major chance to communicate research findings and engage with members of the general public.
The morning focused on Sakai. My colleagues Richard Proctor and Patrick Carmichael from CARET / TLRP, and Sanna Rimpilainen from Strathclyde / AERS reported on their experiences of using the Sakai platform to support diverse learning communities in the UK -- ranging from an anonymous support group for students with mental health difficulties in Scotland, to a UK-wide professional research community. We also got to hear about Richard and Patrick's work using an open standards protocol (OAI-PMH) to perform a network analysis of the TLRP research community. (Essentially, this involves analysing co-authorship of research papers to identify patterns of collaboration).
The afternoon was devoted to showcasing new developments in the British Education Index, a major resource for UK-based educational research and a very important gateway through which research findings get "exposed" to a wider audience. Despite its "retro" database design (to me, this was appealing; however we were assured it was subject to change!), the "new BEI" looks to have some very useful tools. However, I still think there is some distance to go before it becomes truly user-friendly.
To categorise resources, BEI continues to use the British Educational Thesaurus (or BET), an 8000+-term vocabulary which is used by various organisations in the UK to describe educational resources (including the British Education Internet Resource Catalogue and Education Online). As a formal vocabulary, BET is not necessarily an intuitive system. Users will not necessarily be familiar with its terms, or with the way in which it divides resources into a hierarchy of "levels". The designers have addressed this in the usual way, by incorporating a few more "user-friendly" interface options: for example, the database will enable users to "surf" between related results within a single category.
But we're getting fussy these days. Outside academia, the information-finding and -sorting tools we use are increasingly sophisticated. And increasingly they rely less on structured vocabularies and more on what I sometimes call "natural language plus", by which I mean, natural language, plus the special affordances the Internet brings us; namely, hyperlinking, fast resource discovery, and the ability to share (and aggregate) folksonomies and personal ways of organising information.
It might seem cheeky to suggest further developments for BEI when we've only just been exposed to the latest version... But ideas generate ideas. Iterations need more iterations. Resource permitting, what about incorporating some of these "Internet-derived" or -inspired affordances into a future instance of BEI? Things like: an intelligent search agent that learns about my searches as I use the tool. Or a tagging function that lets me tag resources and then expose my interest to other users of the database in some way (even just at the level of "36 users tagged this item"; "The Top 10 items tagged today were...").
At the moment, we tend to think of folksonomies and tagging as minority interests, but I believe their popularity will increase exponentially as more users discover the way that they can use and benefit from them. Some integration with database design would be a major step in the right direction. I am not suggesting any less need for formal vocabularies; user-generated categories may easily sit alongside a professional ontology. The semantic web should enable both to be of use.
The morning focused on Sakai. My colleagues Richard Proctor and Patrick Carmichael from CARET / TLRP, and Sanna Rimpilainen from Strathclyde / AERS reported on their experiences of using the Sakai platform to support diverse learning communities in the UK -- ranging from an anonymous support group for students with mental health difficulties in Scotland, to a UK-wide professional research community. We also got to hear about Richard and Patrick's work using an open standards protocol (OAI-PMH) to perform a network analysis of the TLRP research community. (Essentially, this involves analysing co-authorship of research papers to identify patterns of collaboration).
The afternoon was devoted to showcasing new developments in the British Education Index, a major resource for UK-based educational research and a very important gateway through which research findings get "exposed" to a wider audience. Despite its "retro" database design (to me, this was appealing; however we were assured it was subject to change!), the "new BEI" looks to have some very useful tools. However, I still think there is some distance to go before it becomes truly user-friendly.
To categorise resources, BEI continues to use the British Educational Thesaurus (or BET), an 8000+-term vocabulary which is used by various organisations in the UK to describe educational resources (including the British Education Internet Resource Catalogue and Education Online). As a formal vocabulary, BET is not necessarily an intuitive system. Users will not necessarily be familiar with its terms, or with the way in which it divides resources into a hierarchy of "levels". The designers have addressed this in the usual way, by incorporating a few more "user-friendly" interface options: for example, the database will enable users to "surf" between related results within a single category.
But we're getting fussy these days. Outside academia, the information-finding and -sorting tools we use are increasingly sophisticated. And increasingly they rely less on structured vocabularies and more on what I sometimes call "natural language plus", by which I mean, natural language, plus the special affordances the Internet brings us; namely, hyperlinking, fast resource discovery, and the ability to share (and aggregate) folksonomies and personal ways of organising information.
It might seem cheeky to suggest further developments for BEI when we've only just been exposed to the latest version... But ideas generate ideas. Iterations need more iterations. Resource permitting, what about incorporating some of these "Internet-derived" or -inspired affordances into a future instance of BEI? Things like: an intelligent search agent that learns about my searches as I use the tool. Or a tagging function that lets me tag resources and then expose my interest to other users of the database in some way (even just at the level of "36 users tagged this item"; "The Top 10 items tagged today were...").
At the moment, we tend to think of folksonomies and tagging as minority interests, but I believe their popularity will increase exponentially as more users discover the way that they can use and benefit from them. Some integration with database design would be a major step in the right direction. I am not suggesting any less need for formal vocabularies; user-generated categories may easily sit alongside a professional ontology. The semantic web should enable both to be of use.
Bookmark/Search this page with:
