OpenCulture

Things are (finally!) moving on the Culture Grid front. If you missed it in previous posts, ‘Culture Grid’ is the working title for the MLA-funded project to connect cultural data and the online mass-market, mopping up the last 10 years of legacy services on the way.

Stage 1 is to invest in the Peoples Network Discover Service platform so that it can handle a variety of different content formats, supported by API, SOAP and OAI harvest in and a variety of web services to get the content back out.

This investment began with the BBC CenturyShare project, which we’ve just signed off with the BBC and Knowledge Integration. CenturyShare runs on the PNDS infrastructure, and allowed us to develop towards a new more extensible version (v3.0!)

Stage 2 is to implement a managed withdrawal from the legacy services - Cornucopia, MICHAEL and the MLA Institution Server. This is tricky, because about 7 individual services depend on Cornucopia for backend infrastructure, not to mention the SSNConnect platform, which draws on Cornucopia data.

As part of this withdrawal, we’ll be harvesting these datasets into the PNDS/Culture Grid architecture, and doing some testing to make sure they’re nice and stable.

Stage 3 gets really interesting. We’ve got 3 dependent projects which will be putting data into the Grid and drawing it back out again via web services to develop web interfaces. The Peoples Record is a London 2012/Cultural Olympiad project which will use the Grid as backend infrastructure. The Grid will also power our contribution to the EuropeanaLocal prototype in September (the last staging post before full Europeana integration). Finally, we’ll be working to get the Grid content exposed to GoogleBase for indexing via Google.

Alongside all of this, we’ve got a whole lot of development going on. Naomi Korn is going to be working with us on the Rights framework (and we’ve got a discussion going on with DACS and the UK IPO to set up a licensing scheme to make it work). We’ve been talking about the Digital Preservation angle and thinking about setting up a long-term preservation strategy for all this metadata.

Then we’ll need to sort out sector address data so we can start running location-based services through the Grid. All of which while we’re running the parallel redevelopment of Collections Link. Oh yeah, and we’ve got to make a proposal for the use of Persistent Unique Identifiers under Europeana.

Don’t say we never tell you what we’re doing!

A profoundly exciting conversation with Richard Light yesterday has got my head buzzing about Linked Data and the real ambition of the next phase of development in cultural services.

Collections Trust is working on a project called the Grid, which is basically a next-generation evolution of the Peoples Network infrastructure which aggregates metadata from cultural institutions and serves it up to mass-market services such as Google, the BBC and - interestingly - anyone else who wants to consume it via an open API.

I spent some time this morning looking at Tim Berners-Lee’s TED presentation about Linked Data and two thoughts struck me:

1) The guy is way too young to be the inventor of the Internet and

2) The net result of the last generation of publicly-funded digitisation is not services, it’scritical mass. It’s about the sheer volume of more or less catalogued data forming part of the global collective effort to extrapolate meaning from connected datasets.

Richard is a genius, in many ways, and at least partly because he just does stuff. So, he’s been working with the Wordsworth Trust’s dataset (with the permission of Jeff Cowton, another hero of museum information world) to dismantle it and express it as RDF triples…(sound of roughly half the audience for this post disappearing).

It means that the dataset is transformed from being a database of records into a database of references - every concept in the system is assigned a unique URL, and every assertion it makes about the world is then available to other systems as a reference.

What does it mean in practice? I’m still working that out. On the one hand, it means that the culture sector steps out of the dark web and is able to contribute to the collective global fund of knowledge and meaning. On the other, it means that our organisations can connect to, consume and repurpose the collaborative efforts of countless others, all over the world. In the same way that DBPedia is different from Wikipedia, the next generation of cultural services seems to be based on openness and interconnectedness.

Could it be that we are finally seeing the point of interoperability?