IS 289 Information Visualization Spring 2014
Professor Johanna Drucker Tuesday 1:30-5
[email protected] Office Hours: Tuesday 12-1 Thursday 10-12:30
Overview
Information visualizations have become increasingly prevalent as digital tools have made their creation easier and more popular. Visualization applications make it possible to give graphical form to data, text corpora, networks, behaviors, and any other aspect of cultural material, behavior, or activity that can be structured or parameterized. But do we have a good critical language for understanding the intellectual foundations on which these visualizations perform their rhetorical arguments? How do the historical roots of graphical forms inform their semantic value, or, in other words, how is graphical form part of the content of an information visualization? What are the ways in which the organization of visualization presents arguments about knowledge? What kinds of software tools and applications are available for visualization? How can we analyze the organization of GUI interfaces to understand the ways these depend on visualization to structure assumptions about knowledge and user behavior? What historical and critical tools can be brought into useful dialogue with contemporary visualizations? To answer these questions, some historical background, critical literature, and hands-on problem solving are essential.
Objectives: This class will provide an introduction to some current visualization tools, but it will also provide a critical vocabulary for reading visualizations and the semantic content of graphical organization. It will engage with the issues embodied in relations between data and visualization and will familiarize students with some of the classic images and current literature in this field. The class will also address assessment criteria for information visualization – appropriateness of the graphical forms and models for specific research problems. At its more theoretical edge, the class will engage with experimental visualizations for showing affect, experience, or for creating non-representational presentations.
Format: Seminar with discussion, workshop activities, and presentations.
Prerequisites: None, but familiarity with statistics or GIS will be useful.
Learning Outcomes: At the end of this class students will:
1) Be able to chose between different modes of information visualization for the display of quantitative and qualitative phenomena;
2) Have a basic knowledge of the history and development of graphical forms of information visualization and their epistemological underpinnings;
3) Be able to discuss the rhetoric of information visualizations, their argumentative principles and persuasive force;
4) Have a preliminary familiarity with the literature of information visualization as a foundation for future study;
5) Have an acquaintance with some of the commonly used platforms for basic information visualization.
Projects, Assessment, and Grading Criteria:
Short Projects: (10% of the grade) Short projects will be started in class as exercises and will need to be submitted by the beginning of the next class for credit. They will not be graded, but they are each worth 1-2% of the total grade. Students will do small, short-term projects in many of the following areas:
Graphs, charts, networks, diagrams (Many Eyes, Tableau, Google Fusion)
Processes, lifecycles, instructions
Timelines (Neatline, Simile, or other tools).
Maps (Google maps)
Network diagrams
Final Projects: (50% of the grade)
Students will develop a project by focusing on its historical, critical, creative, or practical dimensions. Each project will have a visual aspect and a written essay. The visual part of the work may be done in any medium – digital platforms or analogue media. Projects will not be graded on the basis of technical skill, but on intellectual engagement with issues, research, and amount of development in the presentation. Students may work individually or in teams, but if they work together, each student will submit a separate written document that includes a statement about who did what signed by all members of the team. Final papers should be between 1200-1500 words, approximately.
Here are some examples:
Historical: Track the visualizations used for epidemiology, animal habitat, the census, financial markets, or any other kind of information; mine the archive for useful ways to organize and work with information in graphical terms. The archive of examples should contain at least five graphs from different sources, and the essay should analyze the approaches. Work is research based, analytic, bibliographic, and descriptive.
Critical: Take a principle of graphs – scale, change, facets, movement, complexity, comparison, growth, temporality etc. – and look at it across a set of examples. Propose a set of alternatives, or a set of best practices or examine the philosophical issues of knowledge production in the visualizations. Pull together a set of visualizations using similar properties for different purposes or information types and address the ways the visualizations make an argument. Work is analytic, theoretical, and conceptual.
Creative: Take a theme and/or topic and create a set of visualizations for it. This might include creating asset of maps that give a sense of the affective character of places at different times of day, or a mood calendar, or a way of tracking illness, or of looking at crucial decision points in policy workflows or emergency rooms. The “data” can be speculative or based on real situations, but the visualization is open to invention. Work is original, imaginative, risky, and aesthetic.
Practical: Take a project, data sets, materials for an actual problem or situation, and figure out the best solution to visualizing it. Analysis of the decision making process and documentation of various attempts should be included in the final presentation. If the project is beyond the skill level of the students involved, then a proposal for how it looks, works, and relates to the information on which it is based should be included. Work is pragmatic, realistic, and oriented towards a real-world outcome.
Assessment criteria: All final projects should demonstrate a connection to readings, images, and principles discussed in class.
Research: Are the references of good quality and well-used and cited?
Argument: Is the argument clear, well-presented, and demonstrated?
Presentation: Are the writing and images carefully presented, proofed, finished?
Quality of thought: Is the work developed, reflective, and thoughtful?
Final presentations will be in class on the last day, June 3rd, and should be turned in with changes or corrections by June 6th.
Attendance and participation, 40% of the grade.
Week by week
Week 1: Introduction to issues, problems, scope and the rhetorical approach (vision, cognition, representation, presentation, visualization, graphics)
In class: a. overview, intellectual questions, terminology
b. work on identifying basic visualizations and formats
c. resources for the history of information visualization
Exercise: Look at best and worst examples and describe their virtues and faults.
Propose an alternative for one of the “worst” examples.
Links:
http://www.informationisbeautiful.net/2013/over-300-of-the-best-data-visualization/
http://eagereyes.org/blog/2008/ny-times-the-best-and-worst-of-data-visualization
http://www.visualisingdata.com/index.php/2013/02/best-of-the-visualisation-web-january-2013/
http://gizmodo.com/8-horrible-data-visualizations-that-make-no-sense-1228022038/all
Discussion: What are the types/categories of information visualizations?
Exercise: Create a spatial data set in class based on the room working in groups of three-four. Make use of (some of) the seven basic graphic variables in creating a visualization of the data using whatever means you have available among you in your group. We will examine and discuss these.
Readings:
Paul Mijksenaar, Visual Thinking (hideous link below)
Alan MacEachren, “Information Processing Model,” from How Maps Work
CCLE
Paul Mijksenaar in Google Books
http://books.google.com/books?id=-j7JcB2al7sC&pg=PA27&lpg=PA27&dq=Paul+Mijksenaar+%2B+articles+online&source=bl&ots=7BkwaDXJVz&sig=DD1Z-S_tU80okQDoSYBStoSCcaE&hl=en&sa=X&ei=-tQoU9b3DNXZoATw2YDQAQ&ved=0CHAQ6AEwCQ#v=onepage&q=Paul%20Mijksenaar%20%2B%20articles%20online&f=false
Week 2: Charts and graphs: visual display of quantitative information
a. what works and what doesn’t, basic skills and insights
b. historical materials, aesthetics, development
c. applications available and their use
Exercise: Generate a data set and then create an information visualization with a short statement of justification for using that particular format.
Exercise: Locate a resource focused on historical material or a specific set of contemporary examples and/or applications related to the theme or topic of your data or your visualization type.
Exercise: Make a connection between historical materials and contemporary problems.
Readings:
Edward Tufte, “Graphical Excellence,” from The Visual Display of Quantitative Information CCLE
Harold Wainer. “Why Playfair,” from Graphic Discovery CCLE
Karin Knorr-Cetina and Klaus Amann, “Image Dissection in Natural Scientific Inquiry” CCLE
Franco Moretti, “Graphs, Maps, Trees” CCLE
http://www.mat.ucsb.edu/~g.legrady/academic/courses/09w259/Moretti_graphs.pdf
Calvin Schmid, Statistical Graphics: Design Principles, excerpt, CCLE
Week 3: Processes, lifecycles, instructions
While many visualizations are concerned with giving a graphical expression to data, others address the rendering of non-visual systems or processes. Certain conventions for showing lifecycles, for addressing pressures/systems/flows of energy, influence, or ideas, for instance, can be put into this process. In this class we will look at:
o Making complex processes visible
o Lifecycles and lapses
o Making invisible processes visible
Exercise: Take a seemingly un-visual process and describe it in steps. Sketch the stages of a process. How are relations among stages depicted? How are the temporal unfoldings of the processes shown? Try to do this exercise twice, once with a complex process and once with a process that seems inherently non-visual.
Exercise: Look at the work of Anton Stankowski, a pioneer in the area of visualizing non-visible processes.
Exercise: Find examples of lifecycle graphics and use one as a model to describe/graph the lifecycle of one of these: a) a news event; b) a relationship; c) a political campaign; d) a crisis management situation in a natural disaster.
Readings and Links:
Martin Gardiner, Logic Machines and Diagrams CCLE
Bruno Latour, “Vision and Cognition” CCLE
Robert Kosara, “Visualization Criticism—The Missing Link between Information Visualization and Art,”
https://viscenter.uncc.edu/sites/viscenter.uncc.edu/files/CVC-UNCC-07-07.pdf
Information is Beautiful, http://www.informationisbeautiful.net/
http://www.dundas.com/blog-post/a-brief-history-of-data-visualization/
Visualizing the invisible: http://architizer.com/blog/wind-maps-visualizing-an-invisible-ancient-source-of-energy/
Week 4: Networks, behaviors: visualization of relations
Categories and properties are central to the first and categories and relations characterize a network. Networks are based on nodes and edges, weights, directed-ness, and qualities of “betweenness” and “centrality” that have precise mathematical definitions, but can also be understood in vernacular terms.
a. examples and samples
b. applications and visualizations
c. what works and what doesn’t
Exercise: Looking at Bernhard Rieder’s analysis of a network or any other example, create an on-paper diagram of a network with which you are familiar. What are the limitations of and/or difficulties with the diagram?
Exercise: How are data structures different for a network than for a graph? Generate a network visualization using Gephi, Cytoscape, or other platform, and/or a dynamic system, set of behaviors or activities, using existing platforms or a representation of your own devising.
In class exercises with Googe Fusion tables and network visualization
Readings:
• Skye Bender-de Moll and Daniel A. McFarland, “The Art and Science of Dynamic Network Visualization,” Journal of Social Structure, Vol. 7
http://www.cmu.edu/joss/content/articles/volume7/deMollMcFarland/
• James Moody, Skye Bender-de Moll, and Daniel A. McFarland, “Dynamic Network Visualization,” http://www.soc.duke.edu/~jmoody77/ajs_online.pdf
• Martin Dodge and Rob Kitchin, excerpt from Mapping Cyberspace, CCLE
• Bernhard Rieder, Interactive Visualization and Exploration of Network Data with Gephi (n.b. contains one offensive image)
Week 5: Dynamic Visualizations and Complexity
In a digital environment, dynamic visualizations add movement, rate of change, and emergent/adaptive properties to those of static images. The data requirements of these are different from those of static data sets, but so are the graphical requirements.
Exercise:
http://images.businessweek.com/ss/09/08/0812_data_visualization_heroes/21.htm
Here are 21 heroes of data vis. Can you define their projects – what is the “information” and how is the form expressing it—what are the transformations, what are the distortions, etc. Think about the distinction of image vs. interface: set up and see if the opposition holds.
Exercise: http://www.umass.edu/molvis/francoeur/index.html
Can you put together a similar gallery/analysis for a domain in which you are interested? Or a discipline – like archaeology: http://www.vizin.org/
To what extent can visualizations be generalized and repurposed across disciplines and to what extent are they specific to the field?
Exercise: What is revealed and what is concealed:
http://infosthetics.com/
Look at HubCab for instance and think about what dimensions of human experience are concealed/revealed. How could this change, or not?
Links:
http://www.visualcomplexity.com/vc/
http://blog.visual.ly/12-great-visualizations-that-made-history/
Week 6: Timelines
Standard conventions for showing temporal sequence are so predominant that imagining temporality in alternative formats is very difficult. This unit looks at timelines and also considers some alternatives.
Exercise: Look at various timeline representations and then pick one to inflect with affective qualities. What graphic languages would you use.
Exercise: Consider the problem of representing the difference between the “telling” and the “told” in a film editing software environment or a news analysis. How do “stories” use time in these two different ways?
Exercise: Here is an example of a timeline used as an interface for historical information. How well does it work? http://www.bl.uk/learning/histcitizen/timeline/accessvers/1200s/index.html
Readings: JD, with Bethany Nowviskie, Speculative Computing, http://www.digitalhumanities.org/companion/view?docId=blackwell/9781405103213/9781405103213.xml&chunk.id=ss1-4-10&toc.depth=1&toc.id=ss1-4-10&brand=default
James Allen and George Ferguson, “Actions and Events in Interval Temporal Logic,” University of Rochester Computer Science Department
http://web.mit.edu/larsb/Public/16.412/pset%204/allen94actions.pdf
Week 7: Maps – point of view and structure, basic conventions
The field of cartography is old, complex, and filled with theoretical and historical and visual riches. Lifetimes have been devoted to its study. This session will not deal with the construction of maps in the larger sense, but with their analysis and use for projects.
Exercise: What is thematic mapping? Create a legend for a specific group, experience, theme, or realm.
http://www.datavis.ca/milestones/index.php?page=varieties+of+data+visualization
Exercise: Look at the maps in the Strange Maps archive and consider the ways they offers alternatives spatial systems and metaphoric schemes: http://strangemaps.wordpress.com/
Readings:
Alan MacEachren, “How Maps” CCLE
Dennis Wood, Maps CCLE
Koch, CCLE
Daru, CCLE
Links: Lief Isaksen, Ptolemy’s Geography and the Birth of GIS, https://lecture2go.uni-hamburg.de/konferenzen/-/k/13960
Week 8 Faceted queries and searching: Interface to data; Data mining and text
Much data is so complicated that it needs a faceted interface in order to be useful. What are the properties of these visualization and how do they relate to the data?
Exercise: Look at the “Digging into Data” project and identify projects that successfully translated complex data through a useful interface.
Ben Shneiderman, “Dynamic Queries” CCLE
Jessica Helfand, Reinventing the Wheel http://www.brainpickings.org/index.php/2014/01/09/reinventing-the-wheel-jessica-helfand/
Readings:
Weeks 9: Non-representational issues, experimental visualization and, also, art and information visualization….
How might we push the conceptual envelope of visualization to address theoretical and conceptual issues in knowledge production? This session will not be more speculative than the rest of the class, and is meant to show the critical dimensions of a field that has many useful pragmatic dimensions that are generally its main focus.
Exercise: Make a proposal for a visualization project that draws on non-representational principles.
Exercise: Analyze the graphical rhetoric in a project where aesthetics take precedence over communication.
Readings/links:
Michael Whitelaw, “Art against information…” http://eleven.fibreculturejournal.org/fcj-067-art-against-information-case-studies-in-data-practice/
Visualizing the invisible, https://www.youtube.com/watch?v=n72e2o79lW0
Nigel Thrift, Non-Representational…
Lynch Woolgar, Representation CCLE
Week 10: Presentations in class.
Books and references beyond those in the syllabus:
A great starting point for finding tools and platforms, DiRT, Digital Research Tools:
http://dirt.projectbamboo.org/
A more commercial/corporate approach:
http://www.creativebloq.com/design-tools/data-visualization-712402
Calvin Schmid, Statistical Graphics: Design Principles and Practices (Wiley 1983)
Nathan Yau, Visualize This (Wiley, 2011)
Jacques Bertin, The Semiology of Graphics (Esri Press, 2010, English Translation of 1967
Jessica Helfand, Re-Inventing the Wheel (NY: Princeton Architectural Press, 2006)
Brian J. Ford, Images of Science (Oxford University Press, 1993)
Daniel Rosenberg and Anthony Grafton, Cartographies of Time (Princeton Architectural Press, 2010)
Katy Borner, Atlas of Science (MIT University Press, 2011)
Edward Tufte, The Visual Display of Quantitative Information (Graphics Press, 1983)
Frank Jacobs, Strange Maps (Studio, 2009)
Peter Turchi, Maps of the Imagination (Trinity University Press, 2007)
Scott Christianson, 100 Diagrams That Changed the World (Plume, 2012)
Robert Klanten, Sven Ehmann, Nicolas Bourquin, eds., Data Flow (Gestalten, 2010)
Manuel Lima, Visual Complexity (Princeton Architectural Press, 2011)
Katherine Harmon, You are Here: Personal Geographies and other Maps of the Imagination (Princeton Architectural Press, 2003)
http://www.historyofinformation.com/
http://www.visualcomplexity.com/vc/
Bryan Connor, The Why Axis, http://thewhyaxis.info/
Kaiser Fung, http://junkcharts.typepad.com/
http://mbostock.github.io/protovis/
People whose work is worth looking at:
• George G. Robertson
• Hans Rosling
• Stephen Few
• Pierre Rosenstiehl
• Ben Shneiderman
• John Stasko
• Jean-Daniel Fekete
• Sheelagh Carpendale
Also: Ben Fry, Stefanie Povasec, Mark Hansen, Lisa Jevbratt etc.
