I hear you & appreciate that prose is an art & craft in itself. Prose, iconography, pictures, spoken word, voice inflection, body language, etc were the only g

I hear you & appreciate that prose is an art & craft in itself. Prose, iconography, pictures, spoken word, voice inflection, body language, etc were the only games in town before the advent of computers. Now we have algorithms, graph data structures, set logic, as tools to work with information. Prose can be analyzed using NLP, but it is verbose, full of throwaway words, & the expression of scoped intent of meaning can be improved. One major weakness of prose is the difficulty of gauging breadth & completeness of expression. When writing software, it is easier to perceive if a program is incomplete. Compilers, type systems, automated tests, & runtime errors guard against bugs or issues with syntax or logic. It is more difficult to parse the prose to understand if all of the essential pieces of information (i.e. logical premises followed to their conclusion) is present in the document. If a conversation is started about a document pointing out missing logic or a fallacy, it is difficult to counter a chain of reason other than quoting phrases & responding to those phrases, possibly taking the phrases out of context. There is also a trend toward rivalrous attention dynamics in online conversations. If you look at Hacker News, many of the threads repeat each other & the disagreements quickly turn into bickering. Such patterns of conversation could even be identified with simple algorithms with can be open. A graph data structure allows the context to be transmitted so if a phrase in a graph is quoted, the context comes along with it. If the phrase is separated from the contextual graph, the meaning changes. An assertion or counter assertion could be tested using Set logic. Phrases are less likely to be taken out of context, diminishing strawmen. The full breadth of unique points can also be aggregated. I think tooling with expressions being in a graph would also disincentivize bickering & rivalrous attention dynamics. The cool thing about writing software is that one participates in the trend of evolved tooling of expression. Whether it's writing a library, creating a program or app, writing a language, an experiment of expression is conducted.