said Tony Hirst as he took us on a mini journey of exploration of just a few of the mashups he has been creating with the OU OpenLearn content and (generally) freely available tools at the Mashup Market session at the JISC-CETIS conference yesterday. From creating the almost obligatory google map to mini federated searches to scrapping content for video, audio, urls to daily feeds of course content, Tony showed just some of the possibilities mash-up technologies can offer educators. He also highlighted how (relatively) simple these things are now and how little time (generally half an hour) it takes. He did concede that some half hours took a bit longer than others A number of the tools Tony talked about are listed on the session conference webpage.

Of course, having well structured, open content has helped enormously to allow someone like Tony to begin to experiment. In terms of reusing content the content scraping that Tony has been doing was really exciting as it showed a simple way to get at the stuff that people (I think) would want to re-use – like videos, urls etc. Also, using an embedded iframe just now allows you to display just the video, not any surrounding advertising. However this may well change over time as advertising becomes more embedded into actual content.

So if it’s so simple to remix, reuse and republish content now, why aren’t we all doing it? Well partly I guess it’s down to people (teachers, learning technologists, students) actually knowing how and what they can do this. But also, there are other wider issues in terms around getting people/institutions to create and open up well structured data. Issues of privacy and our conceptions of what that actually means to us, students etc – particularly relevant given the current government debacle over lost data – and (as ever) IPR and copyright were discussed at length.

Clearly this implications of this type of technology challenges institutions not only in terms of what IT services for users they support, but also how and to whom they open their data to – if at all. Paul Walk suggested that institutions and individuals need to start with the non-contentious things first to show what can be done, without risk. Brian Kelly pointed out that there could be a tension between a mash-up based approach and a more structured semantic approach. Unfortunately this session clashed with the semantic technologies session; but maybe it’s a theme for next year’s conference or something we can explore at a SIG meeting in the coming months.

There was a really full and frank discussion around many issues, but generally there is a clear need for strategies to allow simple exposure of structured data, allow people to get to small pieces of data and easy tools to put it back together and republish in accessible ways. Again the need for clear guidelines around rights issues was highlighted. Some serious thought also needs to be given to the economic implications for our community of creating and sustaining truly open content.