Here we take a broad view of annotation, in particular viewing it as any additional information used to describe or categorise a digital object. Note that this also includes the specific annotation of digital objects (since this provides additional information relating to them), but is not limited to this (it may also be termed labelling, tagging, or categorisation).

At the most general level, annotations of digital objects can be described as ‘descriptive metadata’, although other forms of metadata may be relevant (see the following for an introduction to metadata types http://marciazeng.slis.kent.edu/metadatabasics/types.htm).

One very common form of descriptive metadata is the subject categorisation of the object – for example, using a standard, such as Dewey Decimal or Library of Congress Classification, typically applied to books. More information on these traditional systems can be found here (https://en.wikipedia.org/wiki/Library_of_Congress_Classification), although see also the following interesting article which discusses the ‘Googlization’ of cataloguing, whereby key words rather than a formal codes are used (http://www.wsj.com/articles/SB118340075827155554).

In addition, more specific annotations may be made to the object, with these providing information more relevant to a particular user (or group of users) and their perspective(s) (http://www.digitalhumanities.org/dhq/vol/6/2/000121/000121.html).

See also the following First Monday article, which gives a nice overview of the topic as well as some of the differences present between approaches in museums, libraries and archives:

http://firstmonday.org/article/view/1628/1543