Originally published on Digital Curation class blog, 9/23/09

Two days ago Vimeo demonstrated the “Series Browser”: a visualisation of approximately 65,000 series from the National Archives of Australia’s collection. Mitchell Whitelaw created this as a prototype and part of his Visible Archive – “a research project on the interactive visualisation of archival datasets.”

I am very excited about the prospects these kinds of visualization technologies can have on digital scholarship, particularly in the areas of the arts and history. The browser is interactive, with visualized archival series (represented by squares) arranged both chronologically and in terms of relations to other series. When one scrolls over a series a bubble appears with general scope and content information. The size of the series is illustrated by how big the square is. An inner square represents the volume and the outer square represents the physical shelf space that the series occupies. By selecting and zooming in on a series – lines demonstrating relationships with other series appear (different kind of link relationships are categorized by different colors).

One can also browse and research by agencies outside of series to gain broader context. All this is handled through interaction with the visualized data.

Three hours ago Mitchell launched a video demo describing the A1 Explorer, a similarly interactive visualization of the A1 series in the National Archives of Australia. In this we are looking at what appears to be a kind of tag cloud – as a word-frequency visualization of the contents based upon the item titles. When you hover over one of the words, lines linking it to related words appear. The thickness of the connecting lines is also an additional layer of information. Selecting a text item displays a list of items belonging to that category.

Another aspect of this example is the histogram which displays visually the number of items appearing in the collection per year. This demonstrate the distribution over time but also works as another port of entry into exploring the archive. The text cloud and histogram are also linked in that hovering over a term will display how many of such items appear in the collection per year.

An additionally cool functionality of Mitchell’s visualization is that if there is a term that is dominating the text cloud – if you select and click on it you can remove it from your view and the cloud will adjust accordingly, allowing you to see and explore the remaining items in the collection. If you select an item from the list it will load a digital image of that item from the archive.

I couldn’t help but think about how cool this application would be for navigating my links in delicious.com, but really – the applicability of this for all manner of archives is immense. I hope to see this technology catch on!