Movies + Politics (Aaron and Ramsey)
Posted: October 9th, 2009 | Author: Aaron | Filed under: Assignments | Comments OffHere’s our pitch:
Here’s our pitch:
Wordpress won’t let us put html code into the html tag, it removes it, so we can’t put the presentation here, we have to link to it
by Yoon, Seung
This visualization is a tag cloud of the newswire sections. The frequency of the section determines the weight in the cloud.
Parsing the sections to get the frequency was the toughest, but more rewarding part of working on this project.
I had some trouble adding urls for the tag cloud to the project, to make it more interactive. It’s something that I would want to add
in another iteration of the project.

Tag Cloud
I think of this as a widget that could be in the corner of a webpage.
Interesting library of design patterns for data visualization here:
http://www.patternbrowser.org/code/pattern/pattern.php?4,1,0,1,0,6.php
Based on Christian Behren’s Masters’s thesis — his pattern library is here:
http://interface.fh-potsdam.de/infodesignpatterns/patterns.php
The idea behind the initial design was to create a fun, interactive visualization which would encourage people to look at the news/headlines without the current restrictions of “sections.” The headlines self-organize into their sections eventually, but initially they roam the screen freely, allowing people to see connections between stories that might not end up associated with each other in the paper or on the site.
Code, images, and application



OK I’ve spent WAY too much time on this since class and I don’t really have a lot more to show than some interaction, so I’m calling it quits for now and here is the final project. Getting used to the New York Times API’s was fun and not too difficult, however creating visualizations in C++ is proving to be a very big challenge. I need to accomplish this for my thesis though so as painful as this was I really learned a lot and built a lot of things that I can use in the future to (hopefull) speed up the process significantly
The blog won’t let me embed vimeo video, so here are the screenshots and you can check out the video here

http://parsons.justinblinder.com/datavis09/timeswire/justinblinder_timeswire.html
My visualization initially focused on the “terms” and “catagory tags” associated with each article. After realizing that most of these terms were very specific, I instead chose to use the section that each article was a part of. I created circles for the articles and made variables for their size, color, and position, that corresponded to the section frequency and the title. I hoping to give a basic idea of the section frequency within a certain period of time, however partially randomized positioning is still a bit chaotic.
My original design plan was to have a radial positioning for each section that slowly emanated from the center. I used this process in my second visualization mockup, and showed the frequency of articles for each section within the last 24 hours.

My idea was to pull in the headlines from the NYTimes Newswire API in order to count and visualize the total headline length and relative word lengths of each worth in the individual headlines. I ran into the technical limitation of needing more hours than were available to wrap my head around and make sense of the code that would enable me to visualize my idea. I have experience in using Processing, but it has become apparent that I need to devote more of my free time to becoming more familiar with the language and/or look into developing with an alternate language.
The applet & code are linked below. A look at the commented code will reveal a number of setbacks that I encountered on this project and I hope to address those technical back-end issues prior to launching into the upcoming midterm assignment. Applet at http://a.parsons.edu/~linc184/workaround/hw_applet/
Sorry for the delays. After some consternation about what to do with the TimesWire API, I determined my best approach would be to construct a wrapper and built out some basic tools for my use with it. As I went about this process, one thing that I kept coming back to was the ‘Section’ field for each article. As I worked throughout the night, I saw that over time, different sections were dominating the last 20 responses I was pulling (this is before I found out in class that we could pull results over a longer period of time). My prototype and mockup both deal with the of being able to track which section of the paper was currently submitting the most articles, as a way to get a sense of the ebb and flow of the production schedules of each section. I thought this would provide an interesting look at a hidden side of The Times, and I may pursue it further just to see what the actual results are like. My second idea which I unfortunately could not find a meaningful way to produce was the idea of an article-generated game as a way to combine the growing popularity of casual (especially mobile) games and a desire to have them be educational.
Here’s a screenshot and link to my SWF that returns back The Times section with the highest frequency of articles within the last 20:
[ SWF ]
Here’s a screenshot link to a PDF I made of the mockup for a proposed project that would include this:
[ Mockup (PDF) ]