my.Hero

In 1949, Joseph Campbell wrote The Hero With a Thousand Faces, a survey of world folk stories, myths, and religions that ultimately seeks to distill the massive manifold amount of narrative into a single ‘mono-myth’ – common throughout all human endeavors. This monomyth, put shortly, describes a hero almost ubiquitously male, who leaves a state of normality, passes through a barrier or set of trials into the unknown, conquers both women, himself, and his elders, and then passes back through that barrier or trial to a normal state with the ability to bestow boons on his fellow men. Campbell likened such figures as Christ and Buddha to following the features of this mono-myth, and his theory found an audience in the mind of those such as George Lucas, who based the Star Wars saga on his writings and the idea of monomyth.

As detailed the exposition on each facet of this monomyth is within Campbell’s classic, his influences include the then-popular Carl Jung and Sigmund Freud. Much of Campbell’s reasoning for the existence of a monomyth has to do with their work in the pioneering of psycho-analysis. Campbell reasoned that if there were common threads inherent in the thought process of a person, there would necessarily be common threads in the thought process of a culture, and thus, they would find or invent heroes that either spoke to or were derived from their deepest drives. It is for this reason that Campbell dwells on the sexual in his description of the conquering of women by the hero, despite the fact that many of his exemplar heroes such as Christ, Buddha, and Odysseus did not focus on or even make mention of conquering their own sexual desire.

Since psychoanalysis, Jung, and Freud have not only fallen out of style but have also been damningly questioned in effectiveness and veracity through clinical trial, Campbell’s work is connected to those failing theories in a way that is almost impossible to sever. A monomyth may still exist, and it may very well draw upon innate wishes or desires prevalent to humanity, but it will not come from our id or superego. This may not be a condemnation, if it can be shown that there is still some correlation between heroes, that there are common threads strong enough to tie them together into something resembling a cohesive narrative.

As a student who works in the humanities and is fascinated by the notion that some characters are inherently more attractive than others, and also as a Christian who believes that there is some common thread of a creator within man, Campbell’s idea that there could be some sort of internally derived notion of a hero is immensely interesting. While in 569, I will, architect and develop a web site that will undergo several iterations, allowing a collective whole of people to tell a single story about a hero.

The first iteration of the project will be a bare-bones, database driven application that collects anonymous submissions and displays them on a front page in a style similar to grouphug.us, or any number of blogging sites. These submissions may be allowed commenting or rating, to be indexed for later use. A second iteration will introduce indexing and ‘tagging’ of these stories with sensitive words in the text. A list of different heroic types, tropes, and stories will be devised, and an algorithm created to parse the raw text of a submission and tag it with at least one of these attributes, also available for later indexing.

A third iteration will introduce a fuzzy association map of these tags and terms, to be stored in a database. Each term will be given a randomly generated x and y value after its first appearance in a submission. Where these terms appear together, they will be ‘attracted’ by an attraction function that will pull them together across x and y space. Since these values are simply two unsigned integers, it will be trivial to write this function. Every n hours (to be determined), an expansion function will be introduced that will remove by pairs values from the database, stretch them out an average number of times a compression has occurred, and then re-insert them into the database. The result of this algorithm is that where at first there will be an even distribution of terms across Cartesian space, after an interval of time these terms will float together and collect into clumps, where certain terms appear together, and can easily be processed and ‘read.’ Terms that do not belong together will naturally drift apart, and terms that appear seldom or never will drift off of the bounds of the map and be lost and reclaimed for memory’s sake. It may come to pass that there are both ‘humble and meek heroes’ and ‘courageous and gallant’ heroes, all based on this impartial algorithm.

A fourth iteration will work on separating the causal relationships present in the submissions. Looking for keywords and paragraph breaks that separate the text, there will be as many as five ‘acts’ pulled out of each submission, each one containing its own fuzzy association map of terms. After enough submissions, an overarching story arc will be created – a hero may be meek and mild in the first act, strong and courageous in the second, failing in the third, redeemed and victorious in the fourth, and once again meek, mild, and wiser in the fifth. At this point, this seemingly meaningless set of attributes and values can be read as a collaborative text that all submissions have worked to shape. At this point, the site could be packaged and hosted online after a suitable designer is found.

The project’s audience would be all those who would encounter it online, and who have a story to tell about a hero. These same people that come to the site to tell stories about the good they have seen in the world will also have the opportunity to experience the manifold experiences of all other viewers. If I have time at the end of the quarter, I will also add a system where users can register and keep track of all of their submissions – developing a journal of their experiences with those they respect and admire. In this sense, the site would be one part social network, one part blog, and three parts collaborative storytelling experience.

The challenge would of course be developing a wide enough user base to allow for the content to have any meaning – however the technical algorithms have far-reaching implications in the world of artificial intelligence and semantic mapping, and thus should find some readership when published to Slashdot.org – which generates millions of link views per day. As part of the project, I will also keep a development blog on Blogspot, which will not only place the site high in Google’s rankings (Google runs Blogspot) but will also serve to increase my own ‘Googleability’ which will open up prospects for jobs down the road.

I will code and deploy this application in Ruby, with either of the web frameworks “Ruby on Rails” or “mERB.” I will experiment with both framework’s stability, but both use a model – view – controller architecture and the ActiveRecord library for database interaction, so the fundamental code would be the same. Once I have a working application, I will buy some hosting online for two or three months to attempt to create that readership base. Should it be well-read and used after that time, I’ll attempt to sell advertising space on the page to recoup the cost of web hosting.

I am already well versed in the technologies required for the project, the only learning I have yet to do would be differentiating between mERB and Rails – which is a couple of hours worth of background reading.

You would be tempted to think that this is an ambitious project for the rest of the quarter, and it is, but most of the work is doable in a very short amount of time – I will have to spend a lot of time reading up on semantic parsing techniques to get the sequential order right for submissions, but the basic framework of the web application should be easily deployable, I’ll just have to develop clever solutions for some tricky problems that may not as of yet have solutions.