There is a large difference in usability between the print environment and the screen environment. The limited size of the screen and the limited options for movement within and out of that screen make the screen environment inferior to the print environment for reading text.
In existing online services, there is no use of hypertext.
Prodigy uses a frame structure that breaks a text into single screens that are linked to each other in a strictly linear fashion. Users can jump out of the linear sequence, and pop-up windows provide various escape routes, but a hypertext style is not employed. Thus a text is fragmented in one sense, but no more than a steel chain is fragmented. One must either read the entire text, in order, or not finish it. The user who wants to read an entire ten-frame text has to keep pressing the "Next" button; that user is more restricted than a reader of printed texts, who can always choose to skim ahead or skip to the end.
A similar difficulty is present in CompuServe's format, which (when used with the most popular access or shell software) is window-based. An entire text, perhaps of considerable length, is contained within a window, and the user scrolls the text upward to continue reading to the end. This format is slightly more flexible than Prodigy's, because at least scrolling goes more quickly than screen-by-screen paging, and also because the user can perform a keyword search to find out whether something is mentioned further on in the text (e.g., to see if robots will come up again after they are referred to in the first paragraph). But the scrolling-window reader is still deprived of the print reader's ability to skim and scan. America Online also employs the scrolling model.
The flexibility provided by in-text links in World Wide Web/HTML documents perhaps offers the best of both worlds: by judiciously placing links to subsequent sections in a text, an author can empower the reader to skip ahead instead of scrolling or paging incrementally.