There is no text that is not parallel. That is to say, there is no text that does not admit context, depend upon context, or become subject to recontextualization. Therefore, there is no text that cannot be made more meaningful through the addition (con-catenation with) of or juxtaposition with some other -text. An author may have an expectation of shared experience with the text’s audience — in other words, implicit context — but the work of connecting context with text is always the reader’s own: the reader must recognize parallels, determine their nature, project intent or assume accident, assign meaning, and refactor their own mental model. A text is not a record of meaning but a set of constraints that restrict possible projected meanings (and, an inattentive reader may project internally inconsistent meanings as well). The work of meaning-making remains personal and internal.
Hyper-text (what is now called transliterature) is different. Visible connection (both transclusion and transpointers) makes lines of flight explicit: context (whether intended by the author or accreted by waves of critics, scholars, and provocateurs) is connected, documented, and its relation categorized and color-coded. Detournment and commentary stands equal in status to an original text. Where the web supports only links-as-navigation, translit centers links-as-association.
Software-mediated communal meaning-making can be dangerous when it fails to implicitly acknowledge the universe of context. After the Boston Marathon bombing, redditors marked up collages of surveillance camera stills with red circles and arrows, creating narratives and arguments. But, each of these artworks stood alone, as static images: the format did not allow for a viewer of one such collage to see the rest of the video, or to see all of the other interpretations of the same still image. Instead, one had to manually trudge through endless disconnected static analyses & collate similarities, or ignore context entirely & accept the conclusions of the most shallowly convincing-looking narrative. The absence of a certain kind of context (the fact that, at any large event, there is enough variety to produce numerous ‘suspicious’ behaviors and coincidences) led to extremely high confidence in a wholly incorrect interpretation. Merging all interpretations and commentary on a given piece of media into a single extended universe of discourse, on the other hand, admits the primacy of no single narrative, and encourages further meaning-making in a shared space — even partial & incomplete ideas, which may be developed further by others.
In other words, the death of the author & the opening of his grave to vandals acts as a counter to private mythologies. Limiting in-line commentary to authorized users, or pushing commentary to disconnected spaces (which bear little or no overlap with each other), perpetuates information bubbles. The scale, scope, and variety of potential and actual interpretation remains invisible, and interpreters thereby retain their foolish certainty that theirs is the only viable model.