Annotation (Art History)

Annotation in this context does not refer to textual markup procedures (i.e XML and its variants), but should be understood as providing a way for scholars to relate a thought or an idea to a particular area of an image or a piece of writing, in a number of different ways. This might be in a very informal way using simple free tools that are capable of superimposing arrows and circles over an image or, more interestingly, might provide researchers with ways of collaboratively building up threads of information about objects which can be attributed and archived and overlaid as the item becomes subject to alternative kinds of scrutiny and different methods of analysis.

To take an analogue example, the Witt Library at the Courtauld Institute of Art is a collection of images, extracted from a huge variety of sources that is mostly composed of printed matter depicting works of art, pasted onto thin cardboard mounts, with handwritten, typed and latterly, printed metadata that describes and identifies the image. One fascinating feature of this library’s holdings is that annotations in pencil on the surrounding mount have not only been tolerated but actively encouraged over the years and in certain sections of the library, there is every chance of tracking down initialled comments from very eminent art historians who have had cause to dispute the information relating to the image; and indeed on some occasions to dispute the veracity of the image itself.

In the digital realm, one initiative from the discipline of musicology that provides a model to accommodate this form of collaborative annotation is the Online Chopin Variorum Edition (OCVE) project (http://www.ocve.org.uk). The delivery of this system is via the web (which usefully echoes how most users of art historical information will browse relevant data) and shows what can be achieved within the parameters of a web browser. The pilot version features clickable anchor points over images of digitised sheet music (see fig.1) along with the ability to superimpose and juxtapose different versions of specific bars of music against each other.

The interest of this system lies not with the fact that a point on an image can be made to provoke a an additional hyperlinked window, but in that the reference (or anchor) points on these images can potentially be added by multiple users via the browser, and as much or as little attributed scholarly annotation can be applied to very precise parts of the image. For the purposes of the pilot project, the ability to add annotations to the material was reserved to members of the project team and as such, the version that is currently accessible to the public does not allow third-party annotations, but this is a managerial rather than a technical limitation.

Fig.1  Screenshot from OCVEFig.1 Screenshot from OCVE

In terms of how easy or difficult it is to make web applications truly responsive and interactive in ways that approach the experience of using desktop applications, the growing interest in using a group of technologies that are collectively referred to as Ajax (Asynchronous Javascript and XML) may mean that designers will be able to build in increasingly sophisticated and responsive functionality into web pages. Certain functions within FlickR, Google maps and other Google applications rely on the Ajax approach to enhance their usability and this approach is scalable down to much simpler systems that might simply involve being able to attach virtual post-it notes to web pages that will reappear the next time the page is accessed. In simple terms, the Ajax methodology uses client and server side scripting to replace some of the transactions that would otherwise have to go backwards and forwards from the client to the web server every time the user invoked an action on the web page. Communication between the Ajax engine and the web server happens independently of any actions on the part of the user – hence the designation of ‘asynchronous’.

The DIAMM website (Digital Image Archive of Medieval Music- http://www.diamm.ac.uk) is a current example of a resource that allows users to create notes that will remain persistently attached to an image whenever it is accessed. Once registered, the user can search for an image and then has the option via a ‘tools palette’ to create a public, private or transcription annotation that will be attached to that image and stored on the web server. This is a more common use of an online note facility and in this sense, it is not dissimilar to features already in use by some of the larger art history systems such as the Art and Architecture system (http://www.artandarchitecture.org.uk) based at the Courtauld Institute and ARTStor (http://www.artstor.org/info/). One of the defining features of the DIAMM system however is the incredibly good quality of the images and the functionality of the viewer that allows the user to research the images in very fine detail, thereby maximising the usefulness and scope of the annotation function on offer.

Following on from work on the OCVE project, John Bradley at CCH, King’s College is currently developing an annotation tool called PLINY that will work as a standalone desktop application. Built on an Eclipse platform it is meant to be a flexible and extensible tool and should allow the user to create annotations in a number of different environments including web pages, pdf documents, images and simple text documents.

Fig.2  Screenshot from PLINY (with kind permission from John Bradley)Fig.2 Screenshot from PLINY (with kind permission from John Bradley)

This approach may more accurately reflect the complexity and density of the type of analysis work that some art historians require when working with images. As Bradley has stated, it may also address a need voiced at the University of Virginia Digital Tools Summit, and again at a recent Methods Network Workgroup on Tools Development, for a computing tool that can support the scholarly act of interpretation across formats at a truly useful level of detail. Pliny is still currently (July 2007) a work in progress.

Syndicate content