Geographic information in a web-based world

I am spending the afternoon at a CASA conference (sponsored by the AGI) at University College London talking about use of geographic information, which promises lots of talk about mash-ups, re-use of data.

1.45pm: First up, Richard Milton demonstrating GMapCreator and MapTube – applications for publishing data to Google Maps, showing, for example, maps relating to London, including ones relating to building volumes across the capital, knife crimes, obesity, air quality, etc. You can build maps combining data from different maps. But not just Google Maps, Openlayers offers mapping with lots of map tools, eg: OpenStreetMap.

2.15pm: Alex Singleton talked about public engagement in collecting and disseminating public data. Using GMapCreator CASA took ONS classification data to produce London Profiler, Spatial-Literacy.org and publicprofiler.org. The latter includes the ability to mash-up images from Flickr with neighbourhood maps. Forthcoming projects include education mapping and looking at friend-names’ “poshness” using data from social networking sites.

2.35pm: Following on from Alex, Pablo Mateos talked about geography and ethnicity of people’s names in onomap, and WorldNames, showing how telephone directory and electoral list data can be used to show the distribution of people’s surnames across, so far, about 26 countries covering roughly 1 billion people.

3.30pm: After a quick (and crowded) coffee-break, the conference resumed with Andrew Crookes talking about Mapping people’s mood and crowdsourcing, overcoming the “aspatiality” of most surveys and instead relating data to respondents’ geographical location. Subjects included mapping the credit crunch, anti-social behaviour and the Manchester Congestion Charge.

4pm: Muki Haklay also talked about crowdsourcing – getting people to provide data – in the context of tools such as OpenStreetMap (now being commercialised via CloudMade and others), Flickr and the photographic project Geograph.org.uk. Muki’s focus was on OpenStreetMap, where he addressed issues of quality, the number of data gatherers, completeness, etc.

4.25pm: Jon Reades talked about his cellular census: drawing on data from telecommunications networks (huge datasets!), mapping telephone use against human activities across cities or regions, for example.

4.50pm: Duncan Smith talked about using mapping data in relation to urban forms with particular reference to creating communities – economically, residentially, urban sustainability, transport planning, etc. Need new methods and data (eg building profiles, real estate data, OS address data) to correspond to finer scale architectural or planning needs (London’s Isle of Dogs used as one example).

5.10pm: Andrew Hudson-Smith focused on ‘Web 2.0 and Neogeography’, including geo-caching and Second Life in his presentation, starting with a GIS fly-over map of the population density of the UK. No definition of Web 2.0, but a long list of buzzwords and hundreds of new companies reliant upon users, foremost among them: Google, which has been responsible for an explosion of activity in relation to mapping through Google Maps and Google Earth and various free software. All Andrew’s CASA work is on YouTube, but he is now switching some work to Vimeo. Web 2.0 in the real world, he says, extends to combining the Applie iPhone and Tupperware to undertake high-tech treasure hunts (geo-caching), using gaming environments to fly-though a virtual representation of the real world populated by free Google 3D models, and Second Life (3D representations of Chicago, importing geographical and real-time weather data, editing building models). (See digitalurban.blogspot.com.)

Update (14 January 2009): CASA event presentations here.