book

Beautiful Visualization

by Julie Steele, Noah Iliinsky

June 2010

Beginner

415 pages

14h 7m

English

O'Reilly Media, Inc.

Read now

Unlock full access

How This Book Is OrganizedConventions Used in This BookUsing Code ExamplesHow to Contact UsSafari® Books OnlineAcknowledgments
What Is Beauty?NovelInformativeEfficientAestheticLearning from the ClassicsThe Periodic Table of the ElementsThe London Underground MapOther Subway Maps and Periodic Tables Are Weak ImitationsHow Do We Achieve Beauty?Step Outside Default FormatsMake It InformativeIntended messageContext of use.Make It EfficientVisually emphasize what mattersUse axes to convey meaning and give free informationSlice along relevant divisionsUse conventions thoughtfullyLeverage the AestheticsPutting It Into PracticeConclusion
Question + Visual Data + Context = StorySteps for Creating an Effective VisualizationFormulate the QuestionGather the DataApply a Visual RepresentationSizeColorLocationNetworksTimeUsing multiple visual presentation methodsHands-on Visualization CreationData TasksGathering the dataSorting the data: The discovery versionSorting the data: The technical versionFormulating the QuestionGrouping the dataApplying the Visual PresentationA note about area and circlesPresenting the data with country mapsBuilding the VisualConclusion
Wordle's OriginsAnatomy of a Tag CloudFilling a Two-Dimensional SpaceHow Wordle WorksText AnalysisFinding wordsDetermining the scriptGuessing the language and removing stop wordsAssigning weights to wordsLayoutWeighted words into shapesThe playing fieldPlacementIntersection testingIs Wordle Good Information Visualization?Word Sizing Is NaïveColor Is MeaninglessFonts Are FancifulWord Count Is Not Specific EnoughHow Wordle Is Actually UsedUsing Wordle for Traditional InfovisConclusionAcknowledgmentsReferences
Why Use Color in Data Graphics?1. Vary Your Plotting Symbols2. Use Small Multiples on a Canvas3. Add Color to Your DataSo Why Bother with Color?If Color Is Three-Dimensional, Can I Encode Three Dimensions with It?Luminosity As a Means of Recovering Local DensityLooking Forward: What About Animation?MethodsConclusionReferences and Further Reading
The Need for a Better ToolLondon CallingNew York BluesBetter Tools Allow for Better ToolsSize Is Only One FactorLooking Back to Look ForwardNew York's Unique ComplexityGeography Is About RelationshipsInclude the EssentialsLeave Out the ClutterColoring Inside the LinesSweat the Small StuffTry It OnUsers Are Only HumanA City of NeighborhoodsOne Size Does Not Fit AllConclusion
Techniques and DataColorMotionAnomalies and ErrorsConclusionAcknowledgments
Early Social GraphsSocial Graphs of Amazon Book Purchasing DataDetermining the Network Around a Particular BookPutting the Results to WorkSocial Networks of Political BooksConclusionReferences
Building the VisualizationGathering the Raw DataComputing the Voting Affinity MatrixVisualizing the Data with GraphVizThe Story That EmergedWhat Makes It Beautiful?And What Makes It Ugly?LabelsOrientationParty AffiliationConclusionReferences

The Visualization TechniqueYELLOWPAGES.COMQuery LogsCategorical SimilarityVisualization As a Substrate for AnalyticsThe VisualizationAdvantages and Disadvantages of the TechniqueThe Netflix PrizePreference SimilarityLabelingCloser LooksCreating Your OwnConclusionReferences
Visualizing Social NetworksWho Wants to Visualize Social Networks?The Design of SocialActionCase Studies: From Chaos to BeautyThe Social Network of Senatorial VotingThe Social Network of TerroristsReferences
Depicting Group EditingThe DataHistory Flow: Visualizing Edit HistoriesAge of EditAuthorshipIndividual AuthorsHistory Flow in ActionCommunicating the ResultsChromogram: Visualizing One Person at a TimeShowing All the DataWhat We SawAnalyzing the DataConclusion
Categorical DataParallel SetsVisual RedesignA New Data ModelThe Database ModelGrowing the TreeParallel Sets in the Real WorldConclusionReferences
Briefing and Conceptual DirectionsUnderstanding the Data SituationExploring the DataFirst Visual DraftsThe Visual PrincipleThe Final ProductAll SubmissionsBy PrizeBy CategoryBy CountryBy YearBy Year and CategoryExhibitionConclusionAcknowledgmentsReferences
The More, the Better?Databases As NetworksData Model Definition Plus EmergenceNetwork DimensionalityThe Matrix MacroscopeReducing for ComplexityFurther Matrix OperationsThe Refined MatrixScaling UpFurther ApplicationsConclusionAcknowledgmentsReferences
Getting Data: The Article Search APIManaging Data: Using ProcessingThree Easy StepsFaceted SearchingMaking ConnectionsConclusion
Collecting Some DataLet's Clean 'Em FirstPython, Map/Reduce, and HadoopThe First Pass at the VisualizationProcessingThe Underlay MapNow, Where's That Data We Just Processed?Scene 1, Take 1No ScaleNo Sense of TimeTime-LapseScene 1, Take 2Let's Run This Thing and See What Happens!The Second Pass at the VisualizationBack to That Scale ProblemMassaging the Data Some MoreThe New Data FormatVisual Scale and Other Visualization OptimizationsGetting the Time Lapse WorkingSemiautomatingMath for Rendering Time-Lapse VideoSo, What Do We Do with This Thing?ConclusionAcknowledgments
Our Multimodal ArenaOur Roadmap to Creative ThinkingBeauty and SymmetryThe Computational MediumInterpretation As a FilterProject DiscussionAllobrainArtificial NatureHydrogen BondHydrogen AtomHydrogen Atom with SpinCoherent Precession of Electron SpinConclusionReferences
BackgroundImpact on Forensic WorkThe Virtual Autopsy ProcedureData AcquisitionComputed tomography: Use of dual energy CTMRI: Use of synthetic magnetic resonance imagingVisualization: Image AnalysisObjective DocumentationAdvantages and Disadvantages of Virtual AutopsyThe Future for Virtual AutopsiesConclusionReferences and Suggested Reading
Principles of AnimationAnimation in Scientific VisualizationLearning from CartooningThe Downsides of AnimationGapMinder and Animated ScatterplotsToo many dots?Testing Animated ScatterplotsExploration with animation is slowerAnimation is less accuratePresentation Is Not ExplorationTypes of AnimationDynamic Data, Animated RecenteringA Taxonomy of AnimationsStaging Animations with DynaVisPrinciples of AnimationConclusion: Animate or Not?Further ReadingAcknowledgmentsReferences
Visualization: It's an Elephant.Visualization: It's Art.Visualization: It's Business.Visualization: It's Timeless.Visualization: It's Right Now.Visualization: It's Coded.Visualization: It's Clear.Visualization: It's Learnable.Visualization: It's a Buzzword.Visualization: It's an Opportunity.

Content preview from Beautiful Visualization

Chapter 15. This Was 1994: Data Exploration with the NYTimes Article Search API

Jer Thorp

IN FEBRUARY OF LAST YEAR, the New York Times announced that it was giving away the keys to 28 years of data—news stories, movie reviews, obituaries, and political statistics, all for free. Staring at such a huge pile of information—about 2.6 million articles—we're faced with three important questions. How do we get the data we need? What can we do with that data? And, perhaps most importantly, why should we bother in the first place? In this chapter, I'll try to answer those questions. We'll see how to access information from the NYTimes Article Search API (http://developer.nytimes.com/docs/article_search_api), look at some practical visualization examples, and discuss how the new era of open data is opening doors for artists, entrepreneurs, designers, and social scientists.

Getting Data: The Article Search API

"API" is one of those three-letter acronyms that means very little as a collection of three letters, and even less once you find out what it stands for: application programming interface. While this rather generic term can be applied to all kinds of things within the world of software development, an API typically exists to allow one piece of software to talk to another. If we imagine a database as a physical warehouse that stores information, an API is the shipping and receiving department, and it's open to the public.

In general, interaction with an API is fairly straightforward. We send ...