This section describes how the browser allows the traversal of additional links introduced by cross-references in the document. Command read-follow-cross-ref (bound to g) is executed after interrupting the rendering. By default, this renders the next cross-reference in the document; a prefix argument[+] results in the previous cross reference being rendered. This is necessary -typically, the rendering will have been stopped close to, rather than, on a cross-reference. AsTeR also allows cross-reference links to be traversed while rendering is in progress. When a cross-reference is spoken, AsTeR plays a short sound cue that acts as a prompt. If the user responds with y, the cross-reference is rendered before continuing. The presence of this functionality almost obviates the need for the user to call command read-follow-cross-ref interactively.
How a cross-reference tag is rendered depends on the object that is cross-referenced. For instance, rendering a cross-reference tag to a section results in the section number and title being spoken. This is more useful than just speaking the section number. Typically, the renderings of cross-reference tags are designed to give as much information as possible without being unduly verbose. Renderings of cross-reference tags are customizable by providing methods on generic function read-cross-reference. One such method is shown below.