Ideas for MJP Visualizations
On this page we'll assemble ideas for different kinds of graphics that, we believe, especially suit and enhance MJP data. Eventually we'll try to get to all of them, but in the meantime we'll add links to whatever visualizations we manage to create. We'll also post here ideas that people send us for visualizing MJP data, along with any MJP graphics they create.
At the bottom of this page, we spell out some of the categories we're using to organize the following ideas. And we also list there some different resources available on the web that we've used or that we can recommend for creating visualizations.
Idea 1: exhaustive lists of a magazine’s contents. We should be able to build tables that list all authors who published in a magazine, first sorted alphabetically and then ranked by the number of each author’s contributions. We should similarly be able to build tables that represent all texts of a certain genre (like poetry) that appeared in a magazine. [▘authors, genres; ▘entire journal; ▘MODS files; ▘table; ▘Google Docs charts; ▘new info]
- Click here for tables representing contributors and genres in Others magazine.
Idea 2: comparison of selected author contributions. Maybe using a bar chart, we might compare the same author’s contributions (by their number, and frequency over time) in multiple journals, or the contributions that multiple authors made in a single journal. [▘selected authors; ▘one or more journals; ▘MODS files; ▘bar chart; ▘Google Docs charts ▘new view]
- Click here for charts that compare multiple authors' contributions to The Egoist, The Little Review, and Others.
Idea 3: comparison of texts from a selected genre. Same as idea 3 above, but with genre: we can compare the number of items from the same genre (e.g., letters) published in multiple journals, or the number of items from multiple genres (e.g., poetry, fiction, and articles) that appear in a single journal. [▘selected genres; ▘one or more journals; ▘MODS files; ▘bar chart; ▘ ? ▘new info]
Idea 4: comprehensive representations of entire journals. We should also be able to provide new and exciting ways to see, in a single instance, an entire run of a journal, thus amalgamating the information that's currently segregated in the contents pages for the various issues of an MJP journal. [▘authors, titles, page lengths; ▘entire journal; ▘MODS files; ▘nodetree, sunburst graphics; ▘Protovis ▘new view]
- Click here for three examples of such visualizations of Others.
Idea 5: journal size. We might compare multiple journals in terms of their total pagination (page length of all issues added together) or their total word counts, and then display the results as relatively-sized circles—and perhaps represent each journal further as a set of smaller circles, each representing the length of a single issue. [▘word or page lengths; ▘one or more journals; ▘MODS files (for page counts), TEI files (for word counts); ▘?; ▘?; ▘new info]
Idea 6: network analysis. Using a dataset drawn from three (e.g.) journals, we should be able to sort out which authors published in all three, which authors published in any two, and which published in only one—and then represent that visually (maybe using some sort of Venn diagram, with overlapping circles). We could also additionally calculate/represent the number of each author's contributions to each set of journals. [▘authors, titles; ▘multiple journals; ▘MODS files; ▘?; ▘?; ▘new info]
Idea 7: gender of contributors. By identifying the gender of contributors against a list of gendered names, we should be able to list the contributors to any MJP magazine and then group the names by gender. Then we might count up the contributions (item by item, or page by page) that authors of either gender made to each magazine, and represent those percentages with a chart. [▘gender of authors; ▘one journal; ▘MODS files; ▘pie chart; ▘ Google Docs charts; ▘new info]
- Click here to see some pie charts visualizing the gender of journal contributors.
Every visualization draws together a variety of materials, which means it can be thought of in several different ways. Here are those categories we think are essential for fashioning MJP visualizations, and that we've alluded to [within brackets] at the end of each idea above.
- ▘The subject matter visualized (which may or may not be tagged in the MJP files): e.g., author, editor, genre, page length
- ▘The extent or breadth of the subject matter (which likely corresponds to the number of MJP files used): e.g., the contents from one issue, multiple issues, an entire journal, multiple journals
- ▘The type of MJP files used: MODS files (catalogue records of each journal issue) or TEI files (text transcripts of each journal issue)
- ▘The generic type of visualization: e.g., table, pie chart
- ▘The software or tools used to create the visualization: e.g., Protovis, Many Eyes
- ▘How the visualization supplements the MJP's search pages: i.e, does it offer a new view of already-available MJP information or (more strongly) new information not available through the search pages?
- DIRT: The Digital Research Tools Wiki
- Information Visualization at the Open Directory Project
- "List of information graphics software" on wikipedia
- Visualization and Datamining Software
- Chronos Timeline at the Digital Humanities Hyperstudio at MIT
- Google Chart Tools
- Graphviz: Graph Visualization Software
- Many Eyes at IBM
- Protovis: A Graphical Toolkit for Visualization
- SEASR: The Software Environment for the Advancement of Scholarly Research
- SIMILE Widgets: Free, Open-Source Data Visualization Web Widgets, and More
- VUE: Visual Understanding Environment at Tufts