Last updated on July 25, 2020
The basic function of citations is to give the reader the means through which they can locate a cited source and a superficial means through which they can evaluate the appropriateness of the source by the author’s name, the journal the paper was presented in or the date it was published.
Citations can provide a much richer function however, by providing more information about the cited sources—more meta information—to allow the reader more insight into the source and to better see how the sources connect to the document and to each other. This is what I will present later in this document as a core attribute of Visual-Meta.
As it is today, academic documents have a few special fields for metadata (Abstract and Keywords) but they are not included in Reference Section of the documents which cite them, hence they are one step removed for analysis and not available to the reader. Visual-Meta can easily accommodate such extra metadata without interfering with the fashionable cosmetic layout preferences of the academic field, institution or journal.
This can allow documents can take on hypertextual node characteristics. The benefits can be profound.
1: The author of a document fills in additional fields to their document when exporting/publishing (in addition to Author, Title, Publisher, Keywords etc.), such as ‘Finding: The result of this work is that….’, ‘Introduces: …’.
This is stored in the document’s Visual-Meta and is included when someone cites this document, thorough Augmented Copy or other means, same as all the other fields, including author, title, date and so on. These are similar fields which services such as Scholarcy.com currently provide and which can be employed in such a system but it is clear that the benefit to the author to add this information will increase their ability to communicate the intent of their paper by ‘forcing’ clarity. This is about adding more explicit means for the author to communicate what the document is about. A series of initial fields could be included, with options for the user to add both fields and contents easily.
A possible implementation is mocked up here, in the Author Export Dialog:
and the full dialog:
2) When citing the initial document the user can tag the relationship with the cited source, such as the current document ‘refutes’ the cited document, or ‘enlarges’, or ‘updates’, ‘was inspired by’, ‘uses methods presented in’ etc.
There is considerable research in the hypertext literature about methods of accommodating such link types and other metadata (Trigg & Irish, 2004) (Halasz, Moran, Trigg, 2003) (Zhang, Hu, Zhou, 2008) (Noirhomme-Fraiture & Serpe, 2004). A way to start this could be simply to have fields in the document’s citation dialog for the user to specify relationships, of which a possible implementation is mocked up here, in the Author Citation Dialog:
and the full screen:
3: These advanced fields can then be used by the Augmented Reference sections at the back of the document to organise/sorted based on explicit statements about the source documents and Dynamic Views can also use this connective information for layouts of citations: Augmented References. Keep in mind that this additional metadata is carried forward so that the newer document ‘knows’ what the previously cited documents present themselves to be and how the previous author tagged them.
Thus a document becomes a powerfully hypertextual node in a dynamic network.
If it’s important then write it down. If it’s important that it be understood, write as explicitly clearly as possible.
I would like to credit Mark Anderson for helping clarify the thinking for this work.
And here is Edgar interrupting my writing right now to tell me he went to the toilet all by himself without telling anyone he was going!