I recently gave a lecture on creative possibilities in data visualisation at the tera-scale, so it’s nice to find IBM’s Many Eyes — a personalised ‘social’ data visualization site I hadn’t seen before. It seems aimed at (eventually) tapping into social tagging and social networks to refine visualisations of smaller data sets (less than 1 million items of data), in the form of… “Excel spreadsheets or tab-delimited” files. At the moment it doesn’t seem to be able to handle live feeds, and commenting is limited to writing a blog-like comment below the visualisation.

ibmdata.jpg

The main examples they have at the moment (graph, map, tag-cloud, pie-chart, etc) don’t seem very inspiring or innovative, but I guess they have to reassure the first-time visitor that this is stuff they can understand and use. Although the word-trees seem rather interesting, and could be useful for academics visualising a long paper or thesis.

But it’s running in Java, which may mean that Proce55ing may be powering it (?) — in which case it may offer all sorts of piped eye-candy in future (it seems to be in alpha at present), opening the door to creatives to do creative things with really large data sets, and perhaps even live streaming data.