The Internet first strengthened then destroyed the idea of a canon of art history.
Early Internet dreamers saw the possibility of the utopian virtual museum, drawing together all the world’s great masterpieces to present a coherent narrative of the history of art. But the very proliferation of images that appeared on the web demonstrated the pointlessness of focussing on a single set of definitive images – there are just too many images to choose from. The notion of a canon, intellectually battered during the twentieth century, was helped into its grave by the postmodern sprawl of the World Wide Web.
The more pragmatic approach to creating a portfolio of digital images was to let museums, galleries and other collectors release their own digitised artworks, with which users could then sift through to their heart’s content. Digitise enough high-quality painting, sculpture, architecture etc. and you can let the user define their own canon.
And yet we are nowhere near the tools that are needed to let the user do that. The Internet is a horrible hodge-podge of discoloured images, cropped slides, often without crucial attribution (or worse with the wrong attribution) or posted in flagrant disregard for licencing terms. Using Google search for ‘Bernini’, ‘Braque’ or ‘Ronchamp Chapel’ brings a random collection of images, that might be useful for the casual browser, might not sufficient for professional education.
Alternatives to Google Search exist. There is ArtStor, which has a healthy assemblage of digital images, though its subscribing libraries in the UK find it horrendously expensive (an early more altruistic form of ArtStor, AMICO, failed because it couldn’t find the right business model). Google Art Project gets closers and offers some favourite images, but it’s locked down proprietatry approach defeats the purpose for many art historians (see here and here. The Visual Arts Data Service is great but the collections are often quite niche. The Flickr Commons has had successes but is limited in scope. Wikimedia Commons offers an open alternative but it does not deliver very many artistic images at the moment and it would take a tremendous leap of faith from senior curators before they started using it to publish images from their collections.
Given the difficulty of assembling a centralised pool of digital images, the future is much more likely to see a pool of the related metadata. Content owners are reluctant to freely and openly share their colour corrected, professionally created, high resolution digital images, but they recognise the importance of sharing their metadata. A recent Europeana report (pdf) highlighted the work of the Rijksmuseum in flooding the Internet with high-quality metadata about the Vermeer painting The Milkmaid – with the intention that inserting open metadata in the right channels would lead users back to their high quality version of the image on the museum website. If content curators begin to get more savvy about how metadata is released, then it will become easier to construct the tools which lead people back to the images.
For those interested in art history via the Internet, therefore, it may not be the images but the words that describe them that are the most important thing.