Ross Parry drew the conversation away from funding and back towards considering what the Semantic Web could mean to museums. Nick Poole led on this, describing the museum’s role as mediator and presenter of both the objects themselves and the complex layers of information that sit around them. Nick’s atom-like structure, described by Jennifer as a beautiful Japanese package, led Ross to then question which layers need to be collaborated on and how much of it needs to be standardised for the visions of the Semantic Web to be realised.
To Dan Zambonini, the ideal would be for each resource to have a single URI, which could then later be built upon with additional information or URI’s that are also consensual, agreed and co-ordinated.
To Jennifer definitions and ontologies must stay flexible and open in order to acknowledge fully that knowledge can come from all different kinds of places, although she acknowledged the difficulties with this as museums are ‘giving up something that is critical to [their] success, and not controlling it’. It may be necessary, however, for museums to standardise some of their information.
To this Ross responded:
‘We all recognise that there is that core business function, and with it that moment when we standardise and co-ordinate. And then there’s that fecund, flexible, dynamic, externalised world of interpretations that affect this function. Now, in terms of a UK sector, a publicly funded sector, the first function – which may involve establishing ontologies and taxonomies – is incredibly important and that alone is going to be really hard for us to negotiate in these discussions. The second function – the ancillary bit, the letting go – is difficult for other reasons. Not least because it requires a completely new model of authority and voice for a museum.’
To Nick and Jennifer, much of this has already been established through MODES and Spectrum, leading Ross to pose the question:
‘Is there a fundamental difference in logic between the data models and the ways we’re describing our collections and the ways that that ontology and that taxonomy need to be constructed for the Semantic Web?’
Mike saw that the current well-formed wealth of the information that museums hold should be relatively easy to transpose into a framework useful for the Semantic web. However, Mia questioned how universal the taxonomies would have to be prompting Paul to suggest that something should just be attempted to see what works and that agreements could be made later.
Brian suggested seeking agreement at an appropriate level, albeit fairly local, however, Mike and Nick both felt that a sector level agreement would be important and finally Areti Galani suggested that perhaps one suitable outcome of the Thinktank might be to generate a description of what an ontology might be, leading Jon Pratty to joke in his final statement ‘We’re not going to do nothing’!
- Workshop One – Leicester, July 2006