Writing Stretchtext

After writing two stretchtexts—three, including the present one—it is my very clear recommendation to begin with the long version of the text, writing it out in all its details. Little is different from writing for a paper journal in this phase. Only when a first draft of the whole text is finished is it time to select how to compress it into a shorter version. Some paragraphs will be assigned to the long version, others may be kept intact, always visible.

When the sorting of paragraphsnto the long and medium versions is finished, the author must check that all the paragraphs in the long version is summarised in the shorter. If there is no hint to what will appear when expanding the text, a summary sentence or two must be written.

My experience dictates that the stretched version must always be summed up in the shorter level above. Otherwise, the reader will have no idea of what is in the longer version. It is difficult to decide whether to follow a link or not, if you have no information to base that decision on. If insufficient information is given, the reader will follow links to many nodes she or he is not really interested in. For most genres, that is a less pleasurable reading experience, that may cause the reader to spend less time with the text.

In some literary genres, the text may be puzzle-like, where searching for the nodes that will explain what one has read is much of the experience. I doubt that this is the right poetics for a research article, however.

All references to literature and prior art are linked to a separate page, where the referenced passage is cited in a longer quote than is usual, to provide context to the citation for readers who do not know this text. The passage is also linked to the full text, if it is available online. This is an obvious heritage from Nelson, especially his Xanadu project.

Nelson dreamed of a system where all the world's literature was included, and could be linked to or copied into (transcluded was his term) other work. Today's World Wide Web has not yet reached that point. Depending on your area of research, quite a bit of the reasearch literature is online—and Google has really increased the amount in recent years. Still, I have had to type at least half of the quotes myself from books. This is time-consuming, but I do believe it is worth the effort. I see it as a great advantage for readers to have the original quote available in a bit of context.