For this week, I decided to read the short story by Margaret Atwood titled “The Martians Claim Canada.” Since this was not a long story nor did it have a lot of characters, the network graph looks a little underwhelming. It does, however, illustrate the frequency and direction of conversations between all of the characters. For example, the thickness of the lines between the Martians and the Mushroom character represent the frequency of the conversations between each other. It is clear from a glance that there were far more interactions between the Martian and Mushroom than with each other which took place only early on in the narrative. While informative in the sense that you, the reader of the network graph, can get a sense of the dynamic of the story, it does not reveal much about how these connections were formed.
I think it is important to take note that relationships are developed over time and the kinds of connections that take place building up to relationships determine the type of relationship that is formed. Thus it would be useful if the network analysis allowed users to illustrate the change in the network over time. This way, users can get a better sense of how a relationship develops throughout time. Additionally, having color coded connections would allow users to understand the nature of the relationships. In this case, it is only clear that a connection took place (conversation, statement, question, etc.), but the nature of that connection is unclear. Was the connection pleasant? Was it out of anger? Was it out of curiosity? The color coding these connections would do a lot to illustrate the nature of relationships but more importantly, the characteristics of the characters themselves.These features are particularly important for this story since there are few characters and the type of conversations are primarily expository and satirical which does not quite get captured in the current visualization.
Ultimately the biggest limitation to network analysis/graphs more often than not is not the tools themselves but the person/people behind the analysis. While I would like to think I am considerate of the perspective that is presented in this short story, the author or other readers may have a different take-away from the story and would quantify the connections between characters differently. To be more considerate of human limitations towards network analysis, I think I personally need to learn a more structured way of developing these network analysis to understand how others would view a visualization and how not to be unintentionally dismissive of other important perspectives.
Wow what an insightful comment! I really enjoy how you took a really critical look at the tool we are using in this week’s blog post. The limitations you address are clear and real and should be incorporated into future editions of this tool. In my interpretation of this story, I too had the issue of trying to mark up the transition in the relationship between the Martians and the Mushroom. I ended up adding features, but your idea would likely work better!