Getty Images

Getty Images is a for-profit digital collection that licenses both film and photography. I wanted to look at their site to see the differences in types and quantity of metadata they provide as opposed to archives and libraries. (The image on the right is what I found when I searched "duke" and narrowed it down to "Duke University, Durham, NC". I could have also narrowed it down to "John Wayne".)

1) Selection decisions

Selection decisions are based purely on what images are available for licensing. It appears the entire collection is digitized, with no interest in preservation. In fact, there is no mention as to what is done with the physical copies, if in fact there are physical copies. According to the About Us page, they have 3.2 billion thumbnails on their site and only a tiny percentage of their content is delivered non-digitally.

2) Metadata

The metadata for the collection includes descriptive (caption) and administrative (photographer – searchable), release information, availability, collection. The naming convention looks purely Unique ID-type, perhaps based on date of receipt, or at least a chronology numbered per collection as received. Extensive descriptive metadata is available in the search function – keywords have been added for each image so that the user can find other similar photographs. For example, if you find an image of a brown-haired white man in profile, you can search for other brunets, Caucasians, men, or profile images. On their new site you can search by concept, subject, viewpoint, and more.

3) Characteristics

The images are JPEG. Its persistence is pretty much assured as long as Getty Images is in business. The naming convention is not really intuitive, but there is arguably some system to it considering there are 3.2 billion images.

4) Audience

The audience is intended to be potential and current clients, most likely in advertising.