HCMC Journal

Project DB Requirements from Corinne

26 June 2024: Deniz Aydin
Minutes: 60

Currently, the best option to go forward with Braided Narratives seems to be to use the database, which accepts discrete input data and needs to export discrete output data. To that end, I’ve asked Corinne what their current bottlenecks are, and came up with the following list of requirements that could presumably be easily rectified in the AdaptiveDB codebase:

• In the database, there are multiple entries for the same novel. Out of these, we need to back up and delete the following: Emma’s History of Love, Rowan’s History of Love, Birnan Wood, The Island of Missing Trees, The Overstory, Walden, and The Sound and The Fury (2 copies, dete rp_ and also the one with id = 9).
• Mention and Exchange tabs in the DB need to go.
• We want to keep the character tab, and make narrator a trait of characters. Then we can also do away with the narrator tab.
• Formal structure, time graphs, and possibly also network graphs need to have the narrator information. In the first two, this is a displayed trait; in the latter, this trait allows us to infer relationship types between characters and narrators.
• Not all narrators are characters. There is a 3rd person narrator that should be a global variable for every novel (instead of us having to generate one for each book).
• Who can be a narrator should be populated by the list of available characters + the 3rd person narrator option in the drop-down menu, instead of displaying every character in every book in the database.
• Unchosen option should always auto-select to a default, instead of requiring manual entry.
• There are knows of, knows, and exchange type relationships available between characters, in increasing levels of closeness. The first is a unilateral relationship, while the latter two are reciprocal. Maybe set the colours of the edges based on categories of relationships?
• There should be an option to specify the novel you’re dealing with, which should prompt a paring-down operation of the information presented in the front-end.
• In general, only characters and other information from a chosen novel should show in the AdaptiveDB dropdown lists, instead of all the information from every novel in the database.
• Narrator/chcaracter should go away in the MenEx tab.
• MenEx tab: characters 1 and 2 should be able to be input via names, and exported via character IDs for plotting.
• If a narrator describes two people, they necessarily know of them.
• Delete container and use page number in finding relationships.
• If there are two narrators for the same page ranges, allow for manual conflict resolution.