From People to Presentation
Having reviewed the people in digital libraries and the roles they play, now we turn to presentation and what global users experience when interacting with digital libraries, which they invariably do through a Web browser.
Recall the definition of digital library from topic 1. The definition begins by stating that a digital library is: a focused collection of digital objects, including text, video, and audio … and a good place to start is with the objects themselves (the documents) and how they appear on the user’s screen. The next section illustrates different document displays in digital library systems. There are many possibilities, and we include a representative cross-section: structured text documents, unstructured text documents, page images, page images along with the accompanying text, audio, photographs, and videos. A digital library might include objects that manifest themselves in different forms. For example, music has many alternative representations—synthesized audio, music notation, page images, recorded audio performances—and digital libraries must cater to these optional views.
In addition to documents, digital libraries include metadata like that used conventionally for bibliographic organization. The role of metadata is considerably expanded in a digital library, and we examine some examples to convey how useful it is in helping to organize digital library collections. (topic 6 provides a more detailed account of metadata.)
The definition goes on to say: along with methods for access and retrieval . and this topic illustrates different methods for access and retrieval. Conventionally these are divided into searching and browsing, although in truth the distinction is not sharp. We first examine interfaces that allow you to locate words and phrases in the full text of a document collection. Searching is also useful for metadata—such as finding words in title and author fields—and we look at interfaces that allow these searches, and combinations of them, to be expressed. It is often useful to be able to recall and modify past searches: search history interfaces allow you to review what you have done and reuse parts of it. Next we examine browsing interfaces for metadata—titles, authors, dates, and subject hierarchies—and show how to capitalize on the structure that is implicit within the metadata itself.
But searching and browsing aren’t really different activities: they are opposite ends of a spectrum. In practice, people want to be able to interact with information collections in different ways, some involving searching for particular words, some involving convenient browsing—perhaps browsing the results of searching, or exploring various facets of the information being presented.
The final part of the definition is: … and for selection, organization, and maintenance of the collection.
This topic does not address this activity, except insofar as browsing structures reflect the organization of the collection. In truth, this entire topic is about organizing and maintaining digital libraries, including organizing them in such a way that they are easy to maintain.
Most of the illustrations in this topic show screen shots from a Web browser. However, the browser does not show a static Web page: the digital library software constructs pages dynamically at the time they are called up for display. The navigation bar at the top of Figure 3.1a, the cover picture at the top left, the table of contents on the right, and the book title near the bottom, are all composed together with the text at display time, every time the page is accessed. The information is typically stored in some form of managed datastore, or else reconstructed on the fly from compressed index files. If you look on the computer that is the Web server for this digital library, you will not find this page stored there.
This is why in topic 1 we distinguish a digital library from a Web site, even one that offers a focused collection of well-organized material. The fact that documents are treated as structured objects internally enhances the prospects for providing comprehensive searching and browsing facilities. Furthermore, because all pages are generated dynamically from an internal representation, it is easy to change the entire look and feel of all pages associated with a collection without regenerating or even touching the content of the collection.
