When talking about the foundation necessary to build smarter, digital cities, the focus is often on ubiquitous broadband, wireless access or sensors – in other words hardware. However, another important ingredient is data, ideally of the open flavour. In other words, open data is one of the key enablers for better services, improved opportunities for citizen participation and new business models (have a look at lokaler for a startup making use of open data). In this discussion, open data is often named synonymously with open government data, i.e., public sector data from local, national or even international government bodies. However, public sector data is not all there is – open data can also include data from businesses (opening hours, details of product and service offers, locations, etc.), from non-governmental organisations, or even from individual citizens.
Data portals such as daten.berlin.de are an established solution for providing a central place to find relevant public sector data. What is missing is a way for other local (regional, national, etc.) organisations to make use of the same portal. Of course, the data itself is published de-centrally, but a single channel for finding and announcing relevant datasets from all sectors would make a lot of sense. Even data aggregated through embedded metadata à la schema.org could be contained – e.g., the Berlin data open portal could make the subset of all schema.org information relevant to Berlin available.
So, what about “digital graffiti”? I have stolen the label from a talk on the Data Foundations for Digital Cities recently given by my former Talis/Kasabi colleague Leigh Dodds at the Open Data Cities event in Brighton, UK. Basically, digital graffiti is the traces that citizens can leave in their digital cities; it’s a way of contributing to their city’s web of data. This would usually happen through applications such as Fix my Street or any of the similar sites that followed in its wake. In an ideal world, even commercial services such as Qype or Amen could contribute by opening up portions of their data, particularly when they have data with a geographical context.
As a final requirement, all these different data sources would have to be linked to a larger whole, rather than remaining isolated data islands. There is no easy solution for achieving this, but I believe applying Linked Data principles – publishing 5-star Open Data with URI identifiers and a graph data model – is the way to go here.
I have given a presentation with a more elaborate version of these thoughts a while ago at the Xinnovations 2012 conference in Berlin, where Open Data (and Open Government Data in particular) was one of the major topics that surfaced again and again throughout all three days. As this was a German-language event, the slideset is in German as well.