INFO VIS & AESTH
INFO VIS & AESTH INFO 424
Popular in Course
Popular in Informatics
This 75 page Class Notes was uploaded by Brennan Schmeler on Wednesday September 9, 2015. The Class Notes belongs to INFO 424 at University of Washington taught by Staff in Fall. Since its upload, it has received 41 views. For similar materials see /class/192219/info-424-university-of-washington in Informatics at University of Washington.
Reviews for INFO VIS & AESTH
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/09/15
nfo425 UW iSchool 10042007 Data amp Tasks Thursdav 4 Oct 2007 PoHe ZeHWeger Overview information visuaiizanon modei a process Data in uence on representation encodings Tasks amp data structurecompiexitv In uence on representation Visualization Components Human Abiimes DESign pnnnpies an v ai Maui Vaui pea Design mazess Flamew mus Dan WM Visualization Components Numan mimes visai reamquot mmquot in mean nah we mus The message is central The impact arid appeai of information quantitative or not ows naturaHv from tne signincance arid reievarice of tne message it contains H Winen Vou design tne dispiav ofquantitative information wnetner Vou use a tabie or grapn tne speci c tvpe of tapie or grapn vou use depends primariiv on vour message 42W Show Me me Numbers page 15 nah lypes pacepnm simciurei Cngnnmn quot 39quotquotY anr suns Aiso describes tne process for deueioping a Visuaiizanon Polle Zellweger MacZell Consulting nfo425 UW iSchool on new 10042007 Information Visualization Model my paced Revealed S Ng39m m mama 1 Fll ld or collect raw data 2 Cle amp structure data 3 Select approprlate data encodlngs 4 Present subset ot data Handle user ll ltel actlol l Data amp Tasks Basic Elements of a Data Model A data model represents sorne aspect ot tne Wor Objects are lterns of lnterest people plants cars lms etc Objects allow vou to de ne and reason about a dornaln Ecnsystem ponds strearns woodlands rnountalns plants anlm als etc ld Data Data rnodels conslst ot tnese baSlC elernents abject values also called armbures Few relatons cnapter 2 Numbers Wortn Kl loWll lg Basic Elements Objects Ba Elements Values Values or attrlbutes are propertles ot objects Two major tvpes quanttatlve categorlcal Approprlate Vlsuallzatlons often depend upon the tvpe ot tne data values Polle Zellweger MacZell Consulting nfo425 UW iSchool 10042007 Basic Elements Relations are part atapant Reatians reiate two or more Objects ieaves a department eansists MemoDyes Ecosvstern Types of quantitative information Quantitative aHDWs aritnrn et a aperatans Categorical gruup identity a arganize na artnrnetie n Types of quantitative Information Quantitative aHqu antnrn etie aperatians a 123 29 56 Categorical gmup identify a arganize na artnrnete Norninai narne aniv na argering e egi cm East Scum West Ordinai argereg lt uperatur 7 Time n Feb 2 intervai subdivide inta argereg ranges 7 177999 1017074999 5017079999 1170170719999 Hierarchicai sueaessive ineiusian Region North OR ID WestcA w AZ w a x DHS between streams and Lakes Omi ai predetDrprey netwark wahat eats wnat Ordinai intervai Hierarcnicai Nominal ordinal time series De rived Values Reiationsnips between quantitative vaiues Ranking Ratia carreiatan Sirnpie statistics Average mean me ian made In grange Distributinn range standard geviatian Carreiatan iinear earreiatian eaetneient a negative earreiatan u na mrreiatan 1 pasitive Ranking 1 r Correlation trend line 1 IIIIIIIIII C Polle Zellweger MacZell Consulting nfo425 UW iSchool 10042007 I I I ll Totalsalesvsmerlotsalesgrey Merlot sales as A oftotal sales Types of Data Values Other mfuvls authors use a srmpr eo set afbaslc types Nomme Categarlcal rm Urdermg are Dnly lm type Acnon Comedy Romance Ordma can be urdered m sequence lt nperatnr lm ranng a pa P6713 R Quentltetwe alluw arlthmetu nperatmns lm6an 120 mmutes Data Tables Examples of Data Types Visual Mappings Visual Structure Fllm Fmder U of Marvland Vldeo Quanmatlve AXlS Nurmnal Culur Ordlnal LlSt I Polle Zellweger MacZell Consulting nfo425 UW iSchooI Bertin s Graphical Vocabulary Position g X Ma rks Points Lines El l Areas Er1 Retinal vgams Color Size Shape I I I IEI Grayscale I I III III 0 O o Orientation I I Q 9 Texture 10042007 Tasks amp Data Structurecomplexity Shneiderman The Eyes Have It ATask by Data Type Taxonomy for Information Visualization Visual Information Seeking Mantra Overview zoom ampfilter detailsronrdemand Overview zoom ampfilter detailsronrdemand Overview zoom ampfilter detailsronrdemand Tasks abstract Overview Zoom Filter Details on demand Relate History Extract Tasks Overview see overall patterns trends Zoom see a smaller subset of the data Filter see a subset based on values Details on demand see values of objects Relate see relationships compare values History keep track of actions amp insights Extract mark amp capture data Polle Zellweger MacZeII Consulting Sample user tasks concrete Find value of specific object Find object with given value See patterns trends outliers Compare attributes of two objects nfo425 UW iSchooI 10042007 Data structure complexity 1D 2D 3D Temporal MuItiD class Oct 18 Trees class Nov 1 Networks class Nov 1 Data complexity sample tasks 1D find of items see items with certain Values 2D find adjacent items counting ltering 3D adjacency abovebelow insideoutside Temporal find events before after during M Lllti D find patterns clusters outliers Trees how many levels childen Networks shortest path between points U nivariate Data Tukev box plot can a m M mleEDDa hm Mean U nivariate Data M ABEDE D ABE Bivariate Data c Scatter plot is common Trivariate Data Scatter plot with retinal variable Polle Zellweger MacZeII Consulting nfo425 UW iSchool 10042007 Trivariate Data Temporal Data 3D scatter plot ls posslble Exerclse Under What condmons7 Perspective Wall Robertson 1993 Summary Informanon Vlsuahzatlon model amp process Data In uence on representenon encodmgs Questmns Tasks amp data structurecomplexth In uence on representanon Polle Zellweger MacZell Consulting IN F0 424 UW iSChool Case Studies in Interactive Infovis Design Spence Information Visualization Chapters 2 amp 61 6 65 Thursday 8 Nov 2007 Polle Zellweger A simple usercentered interaction design model werme n dse BS ZbIEh 39 requlremems iRelDesrgn Evaluate Eulld an interactive version Final product 1 182007 1 Today 5 Lecture Usercentered design process Overview Applied to project Case Studies E Chooser Baby NameVoyager I Metro KC oSkope FishCalDateLens z4All Identify needs Establish requirements enlW nee eslauls Flnal product Requirements Data tvpe volatllltv slzearnount perslstence accuracv values US user39s abllltles skllls lnterests novlceexpert casualfrequent Tasks functional w attne svstern should do Y learnabll rv ef clencv memorabll rv errors earlsracrlon Context of use enwronmental clrcumstance s m m ch the svstern Wlll operate pHVSlCal soclal organlzatonal techn cal Polle Zellweger MacZeII Consulting Scenarios Describe human activities or tasks in a story that allows exploration and discussion of contexts needs and requirements Concrete stories that concentrate on realistic and specific activities IN F0 424 UW iSChool Scenario for shared calendar A professor ls trying to schedule a meeting ofthe faculty search Commitlee She selects the search Committee lo entethe names ofmeeung ham lnani quot rnee ing and roughly when rtneeds to take place The system checks the lndlvlduals calendars and the central departrnental calendar for matching lmes Because ofthe slze ofthe committee the system also checks for an avallable meeting room ls free atthe same time and a meeting room ls avallable She selects one of those dates and the system adds the new rneetrng into the member s calendars in a specral color as an extemallyrscheduled rneetrng The system also ernarls each member an alert to request contrnnatron torthe rneetrng and adds the meeting to the calendar for the meeting room ll 1 182007 Project Identify needsestablish requ39 e Data nd or create a dataset explore find ranges outliers patterns questions issues create static vlsuallzatorls User identify and describe the target users ks lderltl quest ons that target users may ask about the data descrlbe the speclfc tasks that your system Wlll support Scenaros document and pro de insight into requirements may demonstrate some usabll ty requirements ReDesign 39M Flnal pruuuct Using scenarios in conceptual design Express proposed or imagined situations Used throughout design in various ways scripts for user evaluaton of prototypes concrete examples of tasks as a means of coroperaton across profess onal boundaries Plus and minus scenarios to explore extreme cases Using prototypes in conceptual design oAllow evaluation of emerging ideas oLowfidelity prototypes used early on highfidelity prototypes used later Polle Zellweger MacZeII Consulting Project ReDesign Consider alternative visualization designs view examples from lectures z4All other class Resources try different encoding schemes for data attributes e representaton poslton color texture connectlvltv e presentat on space and time constraints 7 ll ltel act Ol l modes Proposal versions create lowrfdelity prototypes 7 paper sketches screen composites 7 sklp v through scenarios Final version refine amp iterate based on class amp user feedback INFO 424 UW iSChool 1 182007 Build an interactive version Finaiviadud Lowfidelity Prototyping Uses amedium whichis unlike the nal medium eg paper cardboard 1s quick cheap and easily changed Examples sketches ofscreens task sequences ete Postit notes story oar s WizardofOz Storyboards oOften used with scenarios bringing more detail and a chance to role play all is a series of sketches showing how a user might progress through a task using the device oUsed early in design Paper prototyping Separate page tor each screen Posteits transparent overiavs can heip With interactive areas Hownto Video avaiiabie ii i undergrad iibi ai v Highfidelity prototyping IUses rnateriais that vou Wouid expect to be in the hnai product Prototvpe iooks rnore iike the nai svstem than a iown deiitv version For a highnfideiitv sottware prototvpe cornrnon environments inciude Macrornedia Director Visuai Basic and Smaiitaik manger that users think they have a fuii svstem see compromises Project Build an interactive version Proposai versions sirnpie starybaards 2 designs 2 saenanas eaen Finai version dnaase a design a re ne iterate irn pmve e based upan eiass a martianai user feedback rnare detaiied rnare paiisned starybaards a paper Phdmshnur Pawerpaint Fiash Tabieau TreemaDr neues i s reai data natjust sketchedappmxim ated additianai features ta suppart user tasks rnare seenanas 375 aiternatwe paths Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1 182007 Evaluate Flnalvladud Evaluation Argument for testing WIth only 5 users 5 1x 3 Testlng u use s uluwuleswlsarnlsnjllvvrx Testlng 5 users nds Eu Dfusablllty prnblems u e add tanal budget ta terate a run rnare tests Nlelsen 2mm Project Evaluate proposal verslol ls dass at lnstruetar feedback an task users desldns seenanas VlEWS aptlanal tardet users auts de elass e test usabllltv and udlltv fur selected tasks Fll39lal verslorl evaluate nal prntntype Wltn 2 2 partlelpants e dart anal auts de elass test a terate lnternredlate esldns e ln lab NDV 3D test usabll y and utlllty fur seleeted tasks P4 Presentation ln lab formal presentatlon 2n rnlnutesdraup lneludlnd dlsmsslnn DPT Dr slmllar Past the PPT Dr slmllar Presentatlon ll39lcudes Ova Vnur analysls thhe enedlveness thhe desldns Scenarlos ser and task steps taken ta adnleve task Overwew and detallsaanadernand plus earnpare nlter Case Studies Polle Zellweger MacZell Consulting INFO 424 UW iSChool Definition of Information Visualization Spence 1 182007 VlsuaI29 to rm a mental mude Dr image at something inrorrnation Visualization t ss arrarmind a mental madai arinrarmatian Important cnaracteristic a nun uters may involve senses other tnan seeing sound tadan Spence s Information Visualization Model wuwi i lntmmatiari msunlllnllnn F uiel a niaaimsstiiiitmiimi W man Bf ahirgllyenrwvn m isms in o aa a m a mental we a m m Principal Issues in Information Visualization Representation data i naw the s usually in Visual at encoded form naw suitably enmded data is iaid But in e aailable diSDlav aaa am e includes anaasina m an data ta dispiav lriteractio actions partarm ad by user ta move from Brie VlEW artna data ta anatnar Recalling Shneiderman39s Tasks Overview see overall patterns trends see a smaller subset of the data Filter see a subset based on values Details o and see VEILIE of objects Relate see relationships compare values History keep track of actions amp insights Extract mark amp capture d Desi nin a Sam le 39 g g39 p EZChooser Vlsua zatlon Tool Selecting a car to bu imiw choosing Brie tern from many based Uri attributes Lavered design studv Note selection criteria often impreai nat known at the outset Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1 182007 Car data 0 cars attributes rows columns Creating an overview of prices 0 bargram 0 length proport onal to number of cars in each 95 Price 211 Car appearance 0 object vector of car icons ilh l a Showing individual cars on bargrams o o m Color coding to link car icons to bargrams waasaxsmgs Works for multiple bargrams was a coono m u Iii ar I mammalian Showing an ideal car Tagging a car for future reference i Interactive selection of a bargram price range 0 O O I I I O I Selecting a second bargram range shows cars satisfying both criteria ouuon uvs Scrolling bargrams through a window allows display of many bargrams or use on a smaller dev ce Handling limited space or many bargrams 2 0 user selects wh ch bargrams to view o 1 quotPG 10000 m3 III M n u Focu5ing on a Single price range MP5 I I D I 0 Price Ck Providing sensitivity information a a 60 Mi Mi 5 m Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1182007 Baby Name leard Name Voyager Adding functionality r rimH n rawa H Supporting comparison 4m 39 iquot one Idea Ways a seiect up a 2n names pernaps by dragging a eaen ane a a saved names are draw a cuiurrcnded iine drapn fur eaen saved name aii dispiayed in the standard tim ersenes IQDEIVZEIEIE other Ideas Onerstop snopping for prospective parents derivatinns DfnamES muntry nfnrigin meanings Dfnames reiated names nckna mes Ether ianddade farms famnus peepie W n that name mare PocketPC calendar tool FishCalDateLens nyvlew Agendavlew mumwa Appgmmenl m m M 5am maxlmlse maxlmlae maxlmlse minimise mlnlmlse minimise mea ap aim Scrolling amp searching Evaluation results Tzsls cnmpleled aideerisreair aeedekawe Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1182007 Elements of Information Interactive Examples Visualization U of Mar land Vlz4AH examgles Su mmary Links Userrcentered deslgn process EZChooser w httu b sa marl cum anammvszcmussrmmatasss 15mm tDrVZShnwcase A Media m m pp F J Babv NameVovager HttF WWW babvnamew zard cumnamwnanErlnvm HUM Men39ch httu Utenslt mmkc qu Case Studles EZCHDDser Baby Nam evnyager oSkoDe Httu Huskuue cum I MetrDKC FlshCalDateLens http WW 5 mm Eduhmldatelens nskupe VleAH I lshCalDatELens http WWW cs umd Educlass5pHHQZDDScmscESEswmalsz Vlz4All 4alLa hlml Polle Zellweger MacZell Consulting 8 INFO 424 Lecture 12 November 4 2007 Lecture 12 Layering and Separation In this lecture we continue our exploration of Tufte s design principles Our overarching goal is to explore the different ways of visualizing complex multivariate data This lecture on layering will combine design principles with the visual and perceptual functions that de ne lightness and grayscale In his Layering chapter Tufte shows the importance of separating the elements of a complex visualization into its primary and supporting components Supporting elements should be visually separated using a combination of relative lightness and color The poster my colleague Lyn Bartram and I presented at the IEEE Visualization conference concerns the discovery of algorithmic ways to define layering specifically for supporting structures such as grids Light falling on the retina stimulates the cones which transmit signals along the optic nerve to the brain One of the primary signals describes perceived lightness a quantity called luminance Accurately modeling the perception of lightness or brightness is fundamental to modeling color vision Goals for the lecture By the end of the class you will be able to 0 Describe the key characteristics of Layering and how it is applied 0 Give an overview of how the visual system perceives lightness and describe the following quantitative models intensity luminance and L Describe what is meant by 1l3 Describe what is meant by Whisper Don t Scream Reading Assignment Tufte Envisioning Information chapter 3 Things to consider as you read 1 What is 1l3 and how does it relate to layering 2 How can these principles be applied to your project Reflection questions The questions below are to help you think more broadly about what you ve read and its relationship to the class It is optional but strongly encouraged that you answer them and email your answers to info42 A quot mm to aid in J39 39 in class Email must be received by 7 am on the day ofthe class 1 Find an image or a link that illustrates layering lnfo425 UW iSchool 9282007 TA Marilyn Ostergren Students Name major year in school Welcome Introductions Goals Course Roadmap Info 424 Project Assignments Grading Information Visualization Expectations Instructors Maureen Stone amp Polle Zellweger Questions TA Marilyn Ostergren InfoVis Lab tomorrow only MGH 030 Introductions Goals Maureen Stone Polle Zellweger Students will be able to StoneSoug Consulting MacZell Consulting o DEcribe the key design guidelins and techniques use for the visual display of information including their relationship to human percept on o DEign interactive visualizatons to support human activ ties using real data and a usercentered process 0 Explore and crtcally evaluate a w de range of visualizaton techniques and applicatons EIWlSJVHng lnlbi39nmLon Show Me the Numbers libs Show Me the Numbers Envisioning inronnau39on Designing Tables and Giaphslu Enlighten Edward Tune lean Stephen FEWZEIE4 Ta blea u Softwa re Maureen Stone amp Polle Zelleweger nfo425 UW iSchool Interactive Visualiza Ion Examples 9282007 0 Vlsuallzatlon Components Dean des gn T h q ram Mersnarm Course Roadmap PrOJect We 1 In earns of tudents desrgn and srrnmate an Overwew arunaanentat aaneeats adma mteractwe vrsuahzatron svstem based on rea data Weeks 274 we assrgnrnents Quarmtatwe wsuahzatan m depth m05 Y Ed a 5 Phase 1 Oct 1 73 Ehow Me the Number a Tameau 323 Se ct ana yze nd present yuur data Weeks Uses Few s prmmp es and Tameau 55213765522272233 phase 6 WEEK 97 project Desrgn and mate an rnteraetwe tam Wm Bramsmrm se ect re ne Destqn studres amp quest Speakers 39 WP Em Present 551 Week Bf chess Fma report due De ASSIgnments Grading Genera 45 Project re ated actwmes Ana yze and entdue vrsuahz tans 2530 WP Use Tameau ta exp ure re ne and wsuahze 39 15 V dua Hermann data xnem es feedbadlt atner prnjel s Exp nre an ampere vrsdanzat an systems 45 577 addmona assrgnrnents ummary asstgnmentOmnkm dterrn 1 memes pmect mated 2 Tameau data anatysrs mean a ex Dre re ne and wsuahze WEE Sysmm 392 Damn USE Ta P 39 Summary a srgn nt pruject data a Ana yze and prav de feedback an c assmates 10 3 55 Pamc Pat O prams Attend and part erpate m chess and Labs Send readmg respanses ta mfu424gm 5H earn anagram Maureen Stone amp Polle Zelleweger nfo425 UW iSChool 9282007 Class structure Expectations TueThu 533 1 fo At ecture mass starts 1 Lecture Um pram pay at 1 an as ready pmp use Two ectures 39 1 H D E Lenuveshdeswiibe 2 mm 2 D s21455 damn 2n pussdsasmsswsss um ynur a Frwdav ab At ab 1 am 2 Ciass starts pmmptiy at 1 an Handsm Dumputer actwmes Use the cnmputers fur the assgnment gwen 12 presen nuns Outswde of dass Work out i e of ciass I check Dias web 5 2 and ema reguiariy c yuurasswgnmentsmun We s m cancems eariy Cummun Date prumem Information Visualization Graphwca presentanon of mformanon Charts graphs magr s maps Mustrauuns OngmaHy handscraned stall N W cumpmersgenemed dynamm nteractwe Data gt Pictures gt Insight Usmg wswon to thmk 7 Bertm Lambevtl765 w W Dr John snow an Maureen Stone amp Polle Zelleweger nfo425 UW iSChool 9282007 Film Finder 1994 From Maunder 18741904 ThemeScape 1995 Spotfire MHMDHS Bf D Dcum ants quot531 ltmmuu Cuunesv m Lama Space Shuttle Challenger 1986 Tableau Software Hum Maureen Stone amp Polle Zelleweger lnfo425 UW iSchool 9282007 s nrv of eating Damgo n Fluid Jniun cam iiiiiii MAHAIAIAIIAIAIABAIAIAIIAI HEEL 39 Nu Ereion E Onmnzl unsnwsumzamn The goal of visualization is insight not picmres quotan M um His1EIN mOAR irmdarraue m m Launcmemmme Destined by mile Marthe nigger Lab Tomorrow Special location MGH 030 and MGH 044 Labs normally in MGH 30 Explore and analyze interactive InfoVis Tools Gapminder video Gapminder home httg gagminderorg www Video Lecture from 2006 TED Conference Maureen Stone amp Polle Zelleweger nfo425 UW iSchool 10112007 Today s lecture Channeling Few Projects Schedules amp grades Info 424 Intro to Perception Vis Critiques Projects amp Grading Introduction to Perception Show Me the Numbers ch 6 Eli nearme from some demos are small Few is Applied Tufte Graphical Excellence 0 Intersting data complex deas multivariate o CIEF precise concise prsentat on dataink rat o o Accurate commun cat on lie factor Learn to think like Few Graphical Integrity r ent value relationships accurately r SlZe matches data 7 Ave d area and volume encodings For the mml39mldterm 7 Adjust currency values for ln aton etc For part I of the project o Label carefully and clearly As a basic skill 0 Present data in context Few s Toolbox Practlce Practlce Practlce Easy to dscribe hard to apply Design a graph 0 Scenario 0 Data 0 What graph and why 0 Sketch or create Few s exercises 0 After chapter 5 mm WW Firm one 0 Answers are included nail a Hm Fix the graph In lab Friday Pilllrhw lrl Bad graph 0 Figure out message 0 Say why graph fails o RedEign Dr Jl in m pl mm ll gmmnl ll See Show Me the Numbers page 57 for details Maureen Stone StoneSoup Consulting lnfo425 UW iSChool 10112007 Course Roadmap Week 1 OvervleW amp fundamental concepts Readings Weeks 2 4 Quant tatlve vlsuallzatlol l in depth Show Me the Numbers ampTableau Weeks 58 FeW bakerhome test 7 a sign mostly tied to labs E vlsl l g Information 39 interactive visualization Weeks 9711 Project Project Design studies a guest speakers Grading PrOJect Point system 1000 points class partcipau39on 100 Projectrrelated activities 500 Assignments amp labs 320 Few bakerhome test 80 Design and simulate an interactive visualization system based on real data Models a realworld visualization design task Specifics Teams of 374 students Find your own data on a topc of interest Individual 680 Group 320 Userrcentered design iSchuul Guidelines Project Examples Project Examples From last year Other ideas Find the best place no go fishing Find the best set of flights for a trip balancing cost Find the best place m search for a job Compare the performance of baseball players Compare GDP for different countries across time Evaluate web site performance Monitor and troubleshoot multiple PCs at once Compare the performance of Fortune 500 oompanies Maureen Stone StoneSoup Consulting convenience and reliabil ty Find the best place to live in Seattle balancing cost commute time and safety Evaluate the impact of airline deregulation on airline orma nce Demonstrate the impact of global warming on the Puget Sound region Co over space and time climate plant and animal populatons crime rates airline safety etc Capsmne project data WWW la allt cum nfo425 UW iSchooI Project Phases PhaseI 3 weeks 160 points n Uses brainstorming userrcentered design Design phase and implementation phase s Goal Apply userrcentered design to visualizaton Final report due Wednesday Dec 12 10112007 Phase I P1 Select Teams Data and Topic P2 Individual Data Visualization P3 Group Data amp Task Visualization PF Individual Feedback to P3 Timeline Thursday 1011 to Thursday 1113 weeks Phase II P4 Project Design Presentation PF Individual Feedback P5 Inlab Usability Testing P6 Final Presentation PF Individual Feedback P7 Final Writeup Timeline Friday 112 to Wednesday 1212 56 weeks Phase I To create an interesting visualizaton you need interesting data related to an interesting task Goal Demonstrate you have a su fable topc wth nonr trivial goals plus data to support t Phase I P1 Select Teams Data and Topic Topic of interest Think about possible tasks Find dam in a form you can read into Tableau May need multiple dalasels to explore Create a webs te for the project P2 Individual Data Visualization User lask w at makes it effective Nonrtrivial useful for project goals Feeds into P3 Maureen Stone StoneSoup Consulting Phase I P3 Group Data amp Task Visualization Goal Summarize your data and tasks 5V7 visualizatons Users lasks and insights PF Individual Feedback Two omer projecls we assign Demonstrate your viz cr tquequot skills Help the other project nfo425 UW iSchooI Phase II Now that you have tasks and data design and Storyboard an interactive visualizat on Goal Apply a userrcentered design process to cream an effective interactive visualizaton with a clear purpose 10112007 Phase II P4 Project Design Presentation Goal Explore yourdesign space Brainstorm des39gns and scenaros Select two dis nctly different ones to refine Post on webs m PF Individual Feedback Same 2 projects as before Help projects refine their choices Now you39re ready to implement Phase II P5 Inlab Usability Testing Let your fellow students test your prototype Finertune or apply to P7 P6 Final Presentation Inrlecture demonstrat on of your system 35 detailed scenarios visuals for each step PF Individual Feedback Vis Crtque of P6 P7 Final Writeup Summary of process and selfranalysis Assignments General Analyze and crit que visualizatons Use Tableau to explore refine and visualize realrworld data Explore and compare visualizaton systems Summary assignment minirm dterm roject related Use Tableau to explore refine and visualize project data Analyze and prov de feedback on classmams projects 390 Maureen Stone StoneSoup Consulting 10112007 nfo425 UW iSchool szua System Menta Mude s Euccw mm m Physma Warm g m v m Manon 5 UCng m 3H Rem We shape smp sr m r m p Lxghts suvvaces Eye aphl ahyeds nevve Msua canex ow Sensalvon Emmwm 5mm mam mmquot Cone Response three va ues VIsual System UWPEW Encode spectra as CT Mme Lung m and shun LMS Surname mam Trmhrnmacy my LMS s seen Ram DEM 39 Dw erentspel re can mm 2 same m5 Sort of We a dwgwta cam and curves Uneven v dwsmhuted Curves Dam receptarsquot Three Cancentrated m m h a s LawrhqhtrecEDtar Perwphera wsmn Eyes vs Cameras Gnnd apt ts Smg e fucus We ba anue expusure FuH Wage capture ves Re atwe y pm upuus Cnnstan y scanmng saccades unstandy adju u g Cnnstan y adaptmg wh e bakery2 enta recunstrul mn D mage su expusure no r Wm mu 5 elmnan eBhnnness m Maureen Stone StoneSoup Consulting nfo425 UW iSChool 10112007 Visua percephon 3 um mm Camera work Squaw A s dalkel man B gm Vlsual perceptmn is not ust camera work Color is relative Squaxe A 2 mm Khan a ngw Simultaneous Contrast Bezold Effect Maureen Stone StoneSoup Consulting 6 nfo425 UW iSChool 10112007 Crispening Affects Scale Distribution Perdewed di erenue depends en backgrnund Frum Faircmid Color Appearance Modes Spreading W e pewsW5 me WHHEH anl Lhiiemmiy m Spatiai frequencv an yUU probabiy mum The aw DTP Why Armcdnug m rx uhccmvh m 39mnhndgc sme text was 9w 5 V x image cuiurs quot6 quot eusn l ililldi i In hm on Adjacent coiors biend 3 y nu Med n Hvulinl pmMem Tm N hm m the huumn mmd um nnl med 3 hum h mm hm mu word u u wluhc dmmmg mmv Mm mm Maw 07 Vmon am Wanda 813mm Unwm v Interference Interference RED PURPLE GREEN ORANGE BLUE GREEN PURPLE BLUE ORANGE RED 35 out the coior of the etters 35 out the coior of the etters Maureen Stone StoneSoup Consulting nfo425 UW iSChool 10112007 Why do we care Attention Exploit strengths avoid weaknessas m wsco beckman uwc ed refs demos 15mm WWW Optimize not inte ere Perceptioncognition alone is rarely enough Usercentered Person task attention Maureen Stone StoneSoup Consulting 8 nfo425 UW iSchool 10162007 Today39s lecture Questions on Expectations for the Test D gn tor mu tip e variabies Escaping Flatland Appropnate use of Discussion Escapmg Fiati nd 5 w a the Numbers ch 11 Envisiuning In Feedback on homework furrnatiun ch 1 Have m mum W5 Nextte ure mm mm vnurTu e bunk P1 Due Friday 130 Expectations for the Test Website with teams project data Present at Start of ab Handed out Fridav at Lab end ab anawzmg data Takerhome open book Do it on vour own Handrm to droprbox Tuesdav Website Link Project Document Questwons7 Test components Scenarms raphs a redesign mm the design aws Few Exercises Multivariate Muitwariate muitipie variabies Reiationai Reiationship between variabies Ft Maureen Stone StoneSoup Consulting nfo425 UW iSchooI How can you dlsplay addllional valiables Ham aw a a noun mm w w i mu m mu m A N I c mm mm mm a mm mm an a mo mo 3am 10162007 You can display many maphs within a single eyespun 15 ll 2 w ngC g E lugralsnmlcncc E I I I I Tufte Small multiples One common scale Multiple plots one graph us mm ansalu Mm 000 mm mm in am 5a 000 mm mm 2n mu m we solmare Especlallv tlme serles Multiple scales one graph Growth in Security Incidents m Maureen Stone StoneSoup Consulting 2D layers nfo425 UW iSchooI Not all good Vis is obvious Few 0 Data message audience 0 Make your message clear Tufte o Admires rchness complex ty 0 The illustrations repay careful studyquot 0 compex w tty rich with meaningquot Time series train schedule direc e amp number of stops crossings 9 Arrival amp departure ti ms time at stop number of trains 39 tion 39 39 A n Maureen Stone StoneSoup Consulting 10162007 Sunspot Observations mlu SI39NSPDT mm wmmrn uvu mmvvnllsmm RDTATIGNS luv um um 7 in lwl w m I39I Flow lines or ow maps Interactive Stacked Graph nfo425 UW iSchool 10162007 i ones v inenenie iMwsici igngnie eng 5 genemiemneni Effectively Visualizing MimiValued Flow Data Using Color and Texmre E smve Wmci v Liegyi Negative wnigw biue p g Sivangiv negative Reyngigs snegi siiess gieen nign sWii siiengin amnge gi magemai gepenging gn giieeiign 39 xug iFK 3rd dimension 253 we 531quot w 1quot1 23wu i iquot min Lvs g Cari dynamicaiiy fiiter CIELAB Space Not all 3D is bad ifdete is inherentiv 3D Map in a an snape skeieian m a 3D VDiumE ifdata is 2D Hei ht Wm Perspective maps sgnrgee pints w neigni 5m in Often used in Scienti c Visuaiizanon Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 ThemeSca pe 1995 Wm M Dozuments i i i i i smegma What is the purpose of a Vis Few Other Anew 3mm I Munitnnng I Aid in thinking m mg A WWW I nmmuncat an I Prnbiem SUMmg mm En an 025mm mam Entertainmen Decnratnn and WHD is the aumermw Escaping Flatland Tufte Enwsmnmg Informauan Discussion Questions What does escaping atiand mean7 What exampies did vou nd confusing7 What exampies did vou especiaHv hke7 Maureen Stone StoneSoup Consulting lnfo425 UW iSChool 10232007 Today s lecture Updated schedule Small Multiples MlcrOMacro amp small MUIt39ples Animation vs Small Multiples For each Discussion wth book Envisioning Information ch 2amp4 Digital examples mary I Sum Hope you have your books Updated Schedule Visual Information Seeking Mantra Thursday Exam and homework feed back Overview 100m ampfiter demilsronrdemand Friday Overview and maps in Tablea Overview zoom ampfilter detailsaonademand Full lecture on maps deferred until later in course Homework due no change Overview zoom amp filter detailsaonademand ASAP Send us your project links Thursday P2 Overview zoom amp filter detailsaonademand Individual Tableau visualizations Overview zoom amp filter detailsaonademand Tuesda 3 Group visualization includes a map Overview zoom amp filter detailsaonademand Thursday Project feedback on P3 tOO soon 7 See Final Project document on website for more delail Questions 0verview Detail Techniques Tufte M croMacro Small mul ples Layering chapmr 3 Envisioning Information I fictive MicroMacro Readings Lenses Callaouts Coordinated views view warpin Dynam c layering and filtering Things m think about Paper vs displays ex amples of your own your project Maureen Stone StoneSoup Consulting nfo425 UW iSChool 10232007 Online examples Temporal Data Mava Lin 5 gsternatrc Landscages TeXtArc Home PhotoMesa desktop Dvnarmc caHsouts Googie stocks precoio exarngies Fisheve Calendar Grapn Editor Presentation Tooi Stir Tree M W Googie mags Perspective Wall Robertson 1993 was A Visualumion Syslem tor Explovmg and Anaiymg tame Dambmes MicroMacro summaw datar Cumulates rnta a larger Wnaa Maps at and perspectrve Data dot patterns Patterns in Words and numbers Stem and leafpints Vretnarn memnnal Pictures frorn pictures Scattered wnere39s Waldu Cuba canerent Mariey Objects trn39s Landscapes paper vs dispiavs7 Small Multiples Summary vrsuauy enromrng Compansons ofchanges ofme dr erences among objects me scope ofatemanves Srrnriar repeated instances TD snaw eases tram signais TD snaw vanatran snrng res TD snaw praaess cahgraphy angarnr eranes Smau Mummes Frozen anrmatron TD snaw pnsmnns pianets aness Muybr dge TD snaw praaess rnstructrans Abstractdata to detaried graphic Envisioning Information annas ta think abnut Paper vs displays vs annnatran wnat about real anrrnatron7 exam pies ynu ve seen ynur prDJEL L Maureen Stone StoneSoup Consulting nfo425 UW iSChool 10232007 After the Storm We We a New Animation vs Small Multiples Smallsmultwples often better Easier a see all steps Muybndge Dnesn t rely an mem Dry fur eempensens ntml Dfpaue 5 mm phcated Interactive much better A lmatlon pluses Mare deteu W255 preciser Better uveraH sense ermetmn Wm wa JulieMamsan Mireilleauvancaml Maureen Stone StoneSoup Consulting nfo424 UW iSchooI A brief overview of Flash Keyframe animation system 2D graphics editor Import images eps Adobe formats GUI design system Buttons menus Lists typerin boxes Activate wth Actonscript Programming system Actionscript Fully funct oning language like Iavascript Many predefined objects and classes 1272007 Keyfra me animation system Timeline of frames tage layers of graph cs on stage Autoplay frames nseoond Key frames and inmrpolation Properties Create graphics 2D graph cs editor Import images eps Adobe formats Create symbols and instances A MovieCIip is a sequence of frames GUI Design System UI Components Radio butmns etc Steps Instance component drag m stage Set its parameters Name the instance Write Actionscript to implement the actions Instance name is instance variable in script Programming System Actionscr39p Fully functioning language like Iavascript Predefined classes and objects rimary class is MovieCi 0nline Actionscript Language Reference Uses Supply interact on to designed content on stage Dynam cally generate graphics interaction Files Reads local file HTTP stream Wrms HTTP s am How to learn Flash Download the free trial from Adobe Getting started tutorials for the basics Use the help sequences and explore s I Flash for Dummies Colin Moock BSentiaActionScn39pt O Reilly Press 30 is latest version 20 is latest book Maureen Stone amp Polle Zelleweger nfo425 UW iSchool 10292007 Maps in Tableau Fridav 26 Oct 2007 PDHE ZEHWEgEr Basics Maps the earhest vrs tatrtude v axrs tongrtude x ems TWO notat ons DegreesmmuteSseconds 360 d rees 60 60 Decwmai 360 degrees mmuteS60 SecondS3600 Projections Earth rs 5pherrca gt rmpertect 2D repre ea Mercator prmectron 1569 preserves anges arreetran shape GaHspeters projectron 1973 epreservrng yhndncai prDJEI an Sentatwon n preserve angies areas mstanees drreetrans Thematic Mapping Vrsuahzatron wrth maps 2 majnr methuds Proportional symbol maps Chloropeth ma 5 D cuiur areas ta symbnhze data vames Scale Overvwew deta s mantrc zoormng maps at a greater Susie znnmed nut Shaw i255 detawi maps at a inwer Susie mum 2d m Shaw mare detawi amt mmprmrs 2a Ma Gaavie miv mm ievei7 337KB Polle Zellweger MacZell Consulting Map Visualization Sites Woridmapper Gapmmder More Worid data OECD urg INFO 424 Lecture 5 October 22 2007 Lecture 8 MicroMacro and Small Multiples In this lecture we continue our exploration of Tufte s design principles Our overarching goal is to explore the different ways of visualizing complex multivariate data In MicroMacro Tufte presents images that are created by combining many small pieces into a single whole Rather than removing detail for clarity this technique adds detail to create images that show clearly the whole as the sum of its parts The result provides both overview plus detail in the same image In Small Multiples visualization is tiled with small related images These can be variations on a common model or express a sequence over time By distilling each image into its essential form then repeating the variations become visible at a glance Again overview plus detail Animated sequences present images over time Considering their popularity they are surprisingly ineffective for conveying a precise sequence of visual instructions In most circumstances a sequence of still images like Tufte s small multiples are easier to interpret and use Goals for the lecture By the end of the class you will be able to 0 Describe the key characteristics of the MicroMacro technique and how it is applied 0 Describe the key characteristics of the Small Multiples technique and how it is applied 0 Appreciate the difficulty of using animation for visual instructions Reading Assignment Tufte Envisioning Information chapters 2amp3 Things to consider as you read 1 Consider how these techniques could be applied to your project 2 What are the quot 39 ofthese 39 39 J I 39 quotJ for digital applications and how might they be enhanced or improved Reflection questions The questions below are to help you think more broadly about what you ve read and its relationship to the class It is optional but strongly encouraged that you answer them and email your answers to info42 39 quot com to aid in J39 39 in class Email must be received by 7 am on the day ofthe class 1 Find an image or a link that illustrates each of these two techniques nfo425 UW iSchool 10162007 Perception amp Design w Me me Numbers ch 7amp9 Sho The Vlsua Dlspay onuanntanve 1n br manon ch 2 5n may the ml ma demas ale small Today s lecture Color brle Preattentlve Processlng Gestalt Prlnclples Bertln s semlotlcs Memorv Deslgl l for Commul llcatlol l Vls Crlthues Next Lecture Brlnq vaurTu e hm Phys22d Wand Msual Sydenl Menlal Madels FWS EE WD M WW5me Mama MDdE s ngm Cane Oppanem Pevceplual APPEBVBnue Energv 4 Resume Enmdlng 4 Maaels 4 Maaels nghlslsuvlacesl Eyelamle Redlwmlelshape spew Emma 5mm CW WW Weds quotElvehwsual 5mm dlslvlbullarl lhveevalues llgMness Space Canlen on ex Vuncllans chmrll SYoPl W5 Hue Adamallam F chw z Anew llgmness BECKEYDMHdl saluvallan SlE ewes CECAMDZ Munsell ch ExtemalVIalld elm mum mmquot m Dan39t mm memarlze Perceptual Color Spaces Cnlnr s name Angular scale nghtness brlghtl less lnear scale Bladlt D Wh 2 lrlterlslty Dr purlty W65 Radlal scale Munsell Atlas nudesy Eyelang acbelh Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 Art amp Design Hue calm Wheel oupnstes eempxementteentrast men means 6 Maw mfferentcnlnrwheelsquot see W w nanaennt em tar examples chrarna saturat an In e srtv nrDunW Drstaneerrem erav Value hghtness Darktn heht ADDhEs tn an cnlnrs nntlust erav Wumus Wongr FrmUDS oonorDeslgn RGB39 PseudoPerceptual Models HLS HSV hsa NOT perceptually accurate on NOT pred et percewed hghtness Value er Lummance Color Names Basrc narnes Berhn amp Kev at n es Peyceplua Svmlar evalutn pnnanes Man mtrerentxangua s ge sarnewhateantraversrat Preatte ntive Effects A hrmted set or vtsual propemes processed preattentwew wrthout need for focusmg attentton Much steered for vrsuahzatron what can be percewed rrnrnedratew What propemes are good drscrrrnmetors whatcan rmslead vrewers Fvam Mam tense UCBevkeley Example Color Selectlon Vrewer ean raprmy and acmrately determme whether the target red errexe rs present er absent Drtrerenee detected m calm Fvam Mam nears UCBevkeley Example Shape Selection Vrewer can raprmy and acmrately determme whether the target red mrde rs present er absent Drtrerenee detected m farm mrvature Fvam Mam ream UCBevkeley Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 Preattentive Processing lt 200 e 250ms quah es es prenattentwe Eye muvem ents take at East 2DEIms yet certam prunessmg can be dune very qumkw mp ymg Dv eve prunessmg m paraH2 It a decrsron takes a xed amount or trme regard ess or tne number or drstrectors rt rs consrdered to be preattentwe me Mam nurse 0c Mew Demonstra Counttne 7 s mamaoammnaa 135793659 8276U55 nmepmpnmenane nmepmpemenane EmnS sanr s we numbey meme tne numbev nH s seen pvea emwew Contrast Creates Popout Popout vs Distinguishable Vrewer cannutreprmy and acmratew determme Wnetner tne target red errexe rs present Dr absent Wnen target nes Wu Dr mere features eeen nfwhmh are present m tne erstreeturs Vrewer must seeren sequermaHy me Mam rem Ucaevkeky Poprout TypmaHy see erstrnet va ues swmu taneuus y I p ta 9 under cuntrnHed Dundmnns 39 Dwstmgmshab e 2n eesny fur reesuneme srzed sumuh I l Mere rm 5 mntrnHed mntext UsuaHy need a Egend I I Hue and huhtness Lmhtness nnW Demonstration Conjunction of Features hug llwww ESE ntsu edulfacmtymea euppz o I I o o I I I I o I o cnns Hea Ev Nurtn Carnhna StatE Unwerstv Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 Preattentive Visual Properties Healey 97 length Triesman amp Gormican 1988 width Julesz 1985 size Triesman amp Gelade 1980 curvature Triesman amp Gormican 1988 Julesz 1985 Trick amp Pylyshyn 1994 terminators Julesz amp Bergen 1983 intersection Julesz amp Bergen 1983 closure Enns 1986 Triesman amp Souther 1985 colour hue Nagy amp Sanchez 1990 1992 D39Zmura 1991 Kawai et al 1995 Bauer et al 1996 Beck et al 1983 Triesman amp Gormican 1988 Depth of Field Sharp foreground blurred background preattentive focus Semantic Depth of Field Slide adapted from Robert Kosara http lkosaranetresearchindexhtml intensity flicker Julesz 1971 direction of motion Nakayama amp Silverman 1986 Driver amp McLeod 1992 binocular lustre Wolfe amp Franzel 1988 stereoscopic depth Nakayama amp Silverman 1986 3D depth cues Enns 1990 lighting direction Enns 1990 Em ergence Holistic perception of image 39 i 4 7 Slide adapted from Robert Kosara Use Grouping of WellChosen Shapes for Displaying Multivariate Data anquot as 19 u runmu s 41 1 I cuaa l b magnate m 1 V t SI V ant Mg 39 39 j i 5 I 39 I t e K ft a Iiif 39 39a393 3hn ggzhg a g g h 1 Ra 2quot j k I I I z J A v u From Marti Hearst UC Berkeley Gestalt Principles Idea forms or patterns transcend the stimuli used to create them Why do patterns emerge Under what circumstances Principles of pattern recognition Gestalt German for pattern or form configurat on Original proposed mechanisms turned out to be wrong Rules themselves are still useful Slide adapted from Tamara Munzner Gestalt Laws Relationship between objects Closure Proximity Closure Similarty r Contlnu ty Connectedness 1 l Symmetry 39 L 7 7 and more Proximity Slide adapted from Robert Kosara Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 Similarity 1 hxxxxxxx noncoo XXXXXXX onooooo one XXXXXXX 0 ounuunu XXXXXXX gm mum m hymn Wm Continuity smooth not abrupt change overrules proximity gm mum m hmiz menev Connected ness 3n overrule SiZE Shape am mum mnm menev Sym metw emphasizes relationships gm mum m hmiz menev Figure and Ground Munsiabie Hixsmn I Am 5 on NW 0 L r 0 Sim zdzvmdham Ruben my rm caniams m m 3 Unexpected Effects Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10162007 Influence on Visualization wnv We care xpimtstrengtns avuid Weaknesses Optimize nutinterfere Design criteria Effectiveness Expressiveness Nu faise messages Effectiveness Design criteri Faster to interpret More distinctions Fewer errors 5 00000 Ortms7 Design criteria Expressiveness aH tne data uniy tne data Shaw uniy vei i1 reiat Dnships Shaw Shaw Ordering M Bertin39s Graphical Vocabulaw Position tines m N Areas 13g g I Retinai vanabies Cninr EIII Grayscaie 33 Size 0 o Orientatnn Shape 039 Texture g viiii Berti s Expressiveness Rankings 212pm mm Spent mus Rankings Encoding quantitative data Most iccurale Pxxiiitin 393939 itiiiiin niiiiiii in Iv r liu mene 6 5 54mm Lusluccurale new Q neveimmeii im imam Sperm 2mm Maureen Stone StoneSoup Consulting Slide 34 jdm1 Retinal is an adjective Add quotvariablesquot quotinformationquot Color column is a different font than Gray column Gray gt Grayscale if it fits Jock Mackinlay 1102006 nfo425 UW iSchooI 1 01 62007 Rankings Data relationships Spatial Q o N Object Position Grayscale Extent S ZE c c 3 Color 0 arsnngusn Texture 0 snape O O Ouamta ve BEETS Good 0 Fair 0 Puur Space and Relationships Nominal Presence 3 Nominal Grd liLAliiiiil ordinal Grd liiAiliiilii o AltB Quantitative Scale Q EsA22 m m 2 an 4D 5D an m w an Spatial Coordinates r i rAr i r r i r i m Q 25 m m 2 3E 4D 5D an m w an Geogrepnlcel Scale A lame Qga2495rw u in m I 40 5m to m an an Composition of space lDaXlS 2am I4 3Dagtltls quotITS Z Sll lgieaxls I4 L L Doubieaxls g b g We IE l IE Mark compo tlori L Application Jock Mackinlay 1986 Automating the design of graph cal pre entat of relatonal informa on chris smite 2002 Polaris A Sysmm for Query Analysis and Visualization of Multidimensional Relat onal Databases ViZQL a language for query analysis and visualization Tableau Software Jumu aneau Examples From Jock39s Tnesis Automobile Relations Price cars a 12000 60000 Mileage cars a 10 40 Weight cars a 1500 5000 Repair cars a Great Good OK Poor Bad Nation cars a USA Germany France Cars Accord Audi BMW Maureen Stone StoneSoup Consulting nfo425 UW iSchool Key Idea 10162007 Visual feamres Mappmg quammes mto shapes does not Work 1 mm 5 But usmg extent Works Well How should data of venous types be encoded mto Mileage vs Price in n CnrpnmfwvW my Mum Ivy An Alternate View of Mileage and Prlce m W m m wequot mm M w m illm rm o 5 a e O Make Mileage Price mug ammumm Carma mm Does This Work an em Nw to Willow m Memory icomc memory momma Snapshnt Working memory Shurbterm leltEd 577 chunk5quot Diffeentfurms Longrterm memor Calm WaveJnfurmatmv Vixua zatmn Percelman mumquot Maureen Stone StoneSoup Consulting lnfo425 UW iSChool 10162007 Design for Communication Few and Tufte Organize your data Few General graph design Group prioritize sequence I Accuramlv map quanttv Define your message COM S SEQCSa eS Show only the data Avoid 30 display Daraink ratio 39 focus your me sage Tufte Graphical Integrit Highlight what39s important esent value rela onships accurat ely e Repr sentation ofnum e s precisely matches data Aquot i gy and contrast d d volume encodings e A Just currency value to n aton or other correlated P t39ze Y0 message changes such as populat on changes Layout for readability Label carefully and clearly I Present data in Conmxt Data Gathering Vis Critiques htt collrseswashin tonedu inf0424 Damsourceshtm Look at the images discuss what type of graph is it Why is tgood How many dimensions I DamsInk ratio How could it be improved Relationship to grade on39t worry I Grade reflecis effort I Context of second week Ascritgues Overview Next class Tufte Envisioning Information Escaping Flatland Few ch 11 Design Solutions for Multiple Variables Project Questions Hand out Few test Bring your Tufte book Maureen Stone StoneSoup Consulting INFO 424 UW iSChool 1112007 Interactive Visualization uesdav 30 Oct 2007 PDHE Zeiiweger Today39s Lecture Goais ofinteractive infovis Tecnnidues snaWing bath DVErViEW and detaii snaWing detaiiseanoernand nere Exarnpies Dvnamic Queries rnrnan task nding a best rnatdn Static Information Visuallzation Goai How to achieve it Supporting context Static informatlon Vlsualization Goai ciear concise View otdesired messages erectiveness raster easier nare aecurate ta interpret Expressiveness snaw aii tne data and da nat rnisiead How to acnieve it Data encadings caiar snape size arientat an recnn ques rnicrarnacra srnaii rnuitipies Supporting context Human perceptian papaut cestaitprinapies User guaistasks Expinre mmmun cate User messages carnparisan ranking partetaewnaie dEViatiDn mrreiatnn frequency distribut an tirne series Interactive information visualization Goai interactive svstem tnat supports user goais caad static ViEWS iinked tagetner weii AHDW user ta tccus an gaais ratner tnan cantrais e Stavmg in meriaw usina ViSiDn ta thinkquot How to acnieve it Presentatiun cnaasing Wnat ta present i teract an tecnnidues Supporting context Human aciiities rnatar skiHs perceptian Engnitive skiHs s 2x are nd best rnatcn User aperatians tas s averwew zaarn titer detaiise anedernand reiate history extract snnederrnan pacermnn nmatyrms Cnunnlnn Structure i Cnrml39y Human inievadiun cunimis anr soils 1 Find or coiiect raw data eari amp structure data 3 Seiect appropriate data encodings 4 sent subset of data 5 Handie user interaction Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1112007 Information Visualization Model Revealed wan wavemagWus mm 1 percerve mm rmeracz m lasks Response Time ammatun vrsua1cuntmmtyshders system respunse cumErsatan break 10 sec cugnmve respunse magnumm M tan Interaction it39s what drstmgmshes mfovrs from statrc v1sua1 representatrorrs on paper Ana vsrs rs a process often rteratwe wrth branches and srde Journevs How do vou de ne mteractwe 7 smnmmmr m 5mm Visual Information Seeking Mantra overvrew mum amp ner deta srnnrdemand overvrew mum amp ner deta srnnrdemand Recal g Shneiderman s Tasks Overview see overall patterns trends oom see a smaller subset of the data Filter see a subset based on values Details on demand see VEILIE of objecls Relate see relationships compare values History keep track of actions amp insighls Extract mark amp capture data Simple Interactive Example Even srmpre rmeremmn can be qmte puwerM Stacked msmgrem m mm m smzmnmm m s Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1112007 Interactive Temporal Map m dummy a tad Mmeme m smmam m s Techniques for overview amp details everwew deta s Focus context Semantrc zoommg Overview Details Separate vrews No drstortrorr Shows overvrew detaM srrnu1tarreousw r Informatron s fragmented across mu hp e vrews EXamp e Goog e Maps Focus Context Smg e vrew Show mto m context Contextua1 mto 1 dose to toca1 pomt a Drstortrorr mav make some parts hard to nterpret a Drstortrorr mav obscure structure m data EXamp e Teb eLensXerox PARC lnxwght77 41s 7 Mr Semantic Zooming Show moredxfferent mfo as vou zoom m or drxH down EXamp e Pad presentatron tom Perhm NYU 39 m m m uMEDM H mm Speci Ic Interaction Techniques Se1ectrorr MDuseuverhuvertumnp 5 rt Change representatrorr Hrghhght corrrrectrorrs Brushmg u hnkmg FHtermg Panmng amp zoommg Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1112007 Popup tooltips Mouse Selection Hovermg mouse cursor brmgs up deta s of tem Chckmg on an tem se1 tsx and atmbutes oft e data pomt are shown Se ected tem Atmbutes sun mm m m mm snum m m Sum Rearrange Sorting in Tab1eLens Can sort data Wxth respect to a pamcu ar u att bute m Tab e Lens co umns atmbutes eftand ngm sun mm m m mm snum m m Sum my Mama m manmymmm Brushing amp Linking nghhghts that re ate datapomts m mumpxe mews Changing Representation Egtltamp1e EZChooser MERL Selecting mffa39entrepresentanun 39nm uphuns aziman Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1112007 ZoomingPanning Manv mfows svstems prowde zoommg and panmng capabmnes on dwsp1av Puregeumetrmzuum E V A 7 300912 MED 39 Semantmzuum 7 Perhn presematun mm I mwem Mmu cmLx Dynamic Queries WeHrknoWn and verv usefu mfows techmque Let39s exp ore more deta s Wen Shnewderman e Amberg 94 smmm m 5m Database Queries Querv 1anguage Select houseraddress From at yrea ydb Where pnce gt 00000 and pnce lt 400000 and 2 n bedrooms gt 4 smmam m an Data base Queries P1uses7 Mmuses7 smmm m 5m Typical Quew Response 124 hth found 743 Oak St is beauum 2 523 Pme Ave 7 0 Mt found Problems Musueam 1anguage onw Show exactmatches Don t know magmtude of resmts No he pfu context 1 shown Reformu atmg to a new querv can be s1ow smmmaam m sum Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1112007 Dynamic Query Rationale Speomng a querv bnngs mmedwate dwsp1av of resu ts Responswe mteractwon lt lsec WKH data concurrent presentamon of sohmon m through tne data promote exp oratwon make 1 a much more We expe ence mnesnanng vs batch mammal m an Dynamic Query Features szua1 representanon of wond ofactxon mc udmg both the objects and acnons lmmedwate and contmuous dwsp1av ofresu ts Engagedum M sen Imperfection idea at heart of Dvnarmc Qaenes There anen Sim my sn t ane perfect respnnse a a quer Want a understand a set nftradenffs and onaase same best cnmpmmwse Vuu may 125m mare abnut yuur pramem as yuu Expmre smnmdvm m 55m Dynamic HomeFinder Dvnarmc que es deeo N hamson Ug waw1aw1 1993 Att ute Explorer Dvnarmc quenes Brushmg amp hnkmg deeo snows exp oratwon for house purchase Coxoneneoded querv sensmthv mformatwon Spence amp Tweets lmpeHa CoHego 1998 Dynamic query apps on the Web WW bluemle camdiamand seam ma WWW vaieglannawcellgnans smimnmm m sad Polle Zellweger MacZell Consulting IN F0 424 UW iSChool DQ Critique Strengths Work is fasmr Promote reversing u Very natural interacu39 shows the data ndo explorat on on Weaknesses Operations are fundamentally conjunctive Can you formulate an arb trary boolean express on i A 2 A v A5 A A6 v 7 But may riotbe frequent Controls take space 5 at manta iiriii inii sum 1112007 Summary Goals of interactive infovis Techniques showing both overview and detail 7 separate VleWS overviewdetall e unified View focuscoritegtltt r semantc ZOOmll l showing detailsaonrdemand e mouseoverhover queries vs select on queries 7 brushing amp lll lkll lg Examples Dynamic Queries Common task finding a best match L nks Interactive Stacked histograms us Digital Hismry w Google Maps MW TableLens limo LWww lrlxlghtcomLDl oductssdkstlL Pad Presentaton Tool hQ Mmrl nyu edulwgerlln experlmens zoomPresentauon him EZCho r http bl lsa merl corn EOEOLmyezchooseerydamseB sQ dlrectory Home Finder video hg gzwww cs umd edughcilggubsgvideoazf shim Cell Phone Finder htp WWW myrateplari comgcellphonesl Blue Nile Diamond Search h vaw bluenile comldiamond search asg traclltds Polle Zellweger MacZell Consulting INFO 424 Lecture 2 October 2 2007 Outline and questions Graphical Excellence The goal of this lecture is to lay the foundations for de ning excellence in information visualization and to introduce the students to Tufte and Few Goals for the lecture Re ne the de nition of visualization Describe the different purposes visualization supports Describe what contributes to excellence in visualization Describe how visualization can be misleading Introduce Tufte s principles and how they are applied Introduce Few s approach and his relationship to Tufte Reading Assignment Few Show Me the Numbers Chapter 1 Introduction Tufte The Visual Display of Quantitative Information Ch 1 Graphical Excellence httn39 com 9 W2 hinctnn 1quot inf A 39 quot TufteVi lialDi nlav chlndD Watch the Gapminder Video if you did not see it in Thursday s lecture httpwww canminder u v39 7006debunkin m thaboutthethirdworld htmD Optional Reading Tufte The Visual Display of Quantitative Information Ch2 Graphical Integrity httn39 cmir 9 W2 hinutnn flquot inf 424readin sTufteVi llalDi nlav chlDdD This describes in detail the practices that can lead to misleading visualizations of quantitative data which will be covered in the lecture The reading is optional here to avoid student overload it will be required later in the course Things to consider as you read 1 What are Tufte s principles of graphical excellence and how are they illustrated in his examples Can you nd all the variables encoded in each 2 Few poses his own questions as you read asking you to pause and answer yourself before you go on Doing this will greatly enhance your understanding Reflection questions These questions are to help you think more broadly about what you ve read and its relationship to the class It is optional but strongly encouraged that you answer them and email your answers to info42 A quot com to aid in J39 39 in class Email must be received by 7 am on the day ofthe class 1 Visualization can be used for many purposes Few and Tufte each provide a list of key purposes and your experience with the lab assignment may have suggested a few more Make your own list and describe which are most interesting to you nfo425 UW iSchooI 10252007 Today s lecture Exam Feedback Project update grading etc Exam questions 1 The Condo manager Part 1 One unit is costing more than the average 0 Full credit for deviation 0 3 points for categorcal comparison or ranking Part 2 The cost is correlated with the sales 0 Full credit for timeseries marked wth sales 0 Full credit for correlat on of total costs with sales 0 2 points Turnover rateyear vs average rateyear 1 point if you forgot to adjust for inflation Throughout full credit on b if it supported a Exam questions 5 The dorm manager Part 1 MealsStudentMonth is increasing 0 Full cred t for e 2 numbers iabie of averages e lime series comparison of meaisstudentmonth across at least 2 years 1 point if you forgot to normalize by the number of students 0 It was expected that you combine the three meals Exam questions 5 The dorm manager Part 2 Ratio of type of foods is constant 0 Full credit for e partrofrwhole for the camgories across at least two years bars or stacked bars 7 time series of servingsstudent for each category 0 1 point if you forgot to normalize by the number of students 0 1 point if you made the viewer compare clusters of categories 0 2 oints Confused about the quest on but answered a plausible question Exam Question 2 Pie Chart 0 Difficult to estimate area and angle 0 Perspective makes it worse 0 Other key problems separation wasted space Exam Question 3 3D Graphs 0 Diff cult to estimate height 0 Diffcult to compare I o Other Occlusion missing 39 39 data useless grid etc Maureen Stone StoneSoup Consulting nfo425 UW iSchool 10252007 p1 Select Tearns Data and TopTc p2 1ndmdua1 Data stuahzatTon p3 Group Data amp Task stuahzatTon PF 1ndmdua1 Feedback to D3 Twmehne Thursdav 1011 to Thursdav 1111 3 week Project Grading lSCHOO GquehDes a o ExcEplmna Wuvk duesn l get much oener 3 7 Slmnuwuvk Vangum bulcuumbewm vuved 3 3 Competent Wuvk Meetsveounernentsvuuv bul sn l espean slmnuuvweak 3 o Acceptame om snowssorne aws mommy 39 39 39 2 7 Mnnnauvpassno ManvWeaknessesuvde menmes 2 52 o De menl Duesn l meel rnnrnaT expectatons awed w enuvs ounconssenoes 2 o1 o UnaccEplame Mwsundevsluud tne naluve o1 tne Wuvk vequwed m snows very nme undevslandmu Enuvs and neonss1enoes lmuuuhuul be uw ncummele I TulaHv madequale Wuvk Wuvkwas turned n bul 1D wassubslamallvmcummelEmvmwssedlhepumlenlwew o o Nulhandedm Phase I Schedule Project Feedback on P315 canceled TDD early m project sequent There WlH be PF fur P4 and PE p4 Project deswgn presentanons group deswgns p2 Data vwsuahzatwons mdwwdual Tableau Duetuday an o Web hnk and nardoopy to us 3 Task and Data group Tableau DueTuesdayat oo Webhnk Goals p2 Data wsuanzaoons mdwwdual Tableau Vnu have wable data Everyune nas engaged Wlth wt P3 Task and Data group Tableau Why are you shnwmg MS data2 eres ed2 wnat are tne messagestask 57 HDW does data supportth task2 Draft Prnject name and desmptmn Grading Schedule Feedback on P2 from Polle at lab tomorrow Feedback on P2P3 from both of us at lab 112 wntten grades soon after Maureen Stone StoneSoup Consulting nfo425 UW iSChool 10252007 Project Links Server stats Lane L Jason w and Chas w h tt Listudents WESHH i tori edu uxx1424i424 Htm Government Spending Ryan Devin Daniel T Mon ca h ttg students Washington edud A241 Basketball Yih Sun Sang Timothy Nak I httg Students Washington eduzsemik831info424 Jobs Heather Quentin Benji DanieI K htto astooents washinoton eoozbensonntozeziz Deviant behavior shawn Barry John 39 htt in mi com12nstu Maureen Stone StoneSoup Consulting INFO 424 Lecture 5 October 11 2007 Lecture 6 Perception and Design In Lecture 5 we illustrated the complex highly contextual nature of vision Because we spend a long time discussing the project we didn t get into any of the speci cs proposed in the previous prep In this lecture we will describe the basic visual principles used in visualization and relate them to design principles used in visualization The Few reading does this application for the business graphics that are his focus He also takes this opportunity to discuss Tufte s data ink ratio which was introduced in VDQI The Visual Display of Quantitative Information chapter 1 Lecture 2 We include the second VDQI chapter here which talks about Graphical Integrity including Tufte s Lie Factor Goals for the lecture By the end of the class you will be able to 0 Explain preattentive processing its strengths and its limits Describe the Gestalt principles of visual perception and their application to visualization Understand how the Bertin semiotics visual language for encoding quantity is applied to visualization not covered in the reading but will be discussed in lecture Describe basic models for color and contrast detail in later lectures Relate these principles to design techniques for visualization During practicum we will 0 View and discuss some of the visualizations handed in for the Assignment 2 Vis Critique 0 Demonstrate some techniques for collecting data from the web Reading Assignment Few Show Me the Numbers Chapters 7 amp 9 Tufte The Visual Display of Quantitative Information chapter 2 Ch2 Graphical Integrity httn39 mm 9 W2 hinmnn r1quot in 424readin squteVi unlDi nlav chlpdf Things to consider as you read 1 Chapter 7 is Few s application of Tufte s principles of Excellence both of which are based on perceptual principles Try to make the connections as you read 2 Chapter 9 addresses many of the same issues as VDQI 2 Again think about the connections as you read Reflection questions The questions below are to help you think more broadly about what you ve read and its relationship to the class It is optional but strongly encouraged that you answer them and email your answers to info42 A quot com to aid in J39 39 in class Email must be received by 7 am on the day ofthe class 1 Few makes I hope a persuasive case for avoiding 3D in static graphs Would the problem be as severe if you could interactively manipulate the graph 1 1292007 INFO 424 UW iSChool Document corpora I StarTree fur NASA ats Document searcn resu ts THEBarS pupuutpnsm Woodma Thursday 29 NDV 2mm PDHE ZEHWEQEr Document contents T2gtlttAn2WDrd frequenmes m 1 F md Datuments Text amp Documents waym 39 quot 39 WWW MWJWM h quotK v gummangm Jag vnmutwwwnrx luau 39 39 mama K 3 y r39lfl u wmnl mm EInmmnginn on n awewrh mum39m gritw Polle Zellweger MacZell Consulting INFO 424 UW iSchool 1 1292007 Hypertext documents Problems with annotations 0 Paper limited number 6 size distant from primary impacts primary presentation 9 Hypertext removes annotation from primary context Goal annotation in context Lightweight contextual access fluid shift of attention from primary to annotation and back 9 easy comparison of material reduce reader disruption Demo fluid documents Talk overview o Hypertext Graphical techniques FlUid links FlUid links stud Open uid hypermedia 39 Other applications Fluid spreadsheets Fluid fiction Fluid reading primer 39 Conclusions Polle Zellweger MacZell Consulting INFO 424 UW iSChool 11292007 Graphical Techniques Demo graphical Techniques 39 rth tlyj53 39 Ell v d Interl ne 2 31quot 5 d 52quot ng ifh39ri sif j 9 39 DEEmIlion o n mind and ile they have Lruines of the Loire Valley of France El Tram Wine grape onn AlsaceLorraine northern ta1 5 of S uternes an winekafthe Loi E whites of Alsace good whites of C Sauternes are made with the semillin 339 Margin callout molded faglrrn tal153aun39ignan Blane is used i I C 39 ernes and Barsac the Mwhites of 77 J es are made with the senillian F after h rotted and rrnjlded Fluid documents approach Four steps 0 visual cue to annotation 0 reader indicates interest lightweight l g a r Ull 9 primary and annotation negotiate 1 I y a y for space and salience r s coo 6 animated transition Problem with hypertext Fluid links KING MAKES IT THREE By Doug O39Harra and Craig Medred 39 Machemquot Choos39 9 AKquotWEdde Mmls mg osses lightweight contextual i nto Pounded by erce coastal Winds Jeff King of Denali Park Whelhequot 0 f ll W irrwirg zzfsgszgfi tziiziiiziiif xza near anchor without obscuring source Ii giIldVi tOijIlggays 11110qu dhi d WEI rlauegshiliiia grgiilnd lealizgaIgr thatS cut fisibililfyatlo alrrsrosltgs 5 U p m I 5 O n I I nothing He later said the Weather Was the Worst he39d VV1lTIe edmi lditamdrace h r get a preview of the 39 destination while still in the source context manage when to follow a link Zellweger Chang Mackinlay Fluid Links for informed and incremental link transitions Hypertext 98 Use animation smooth experience of viewing glosses Enable augmented features improved gloss content enhanced hypertext navigation Polle Zellweger MacZeII Consulting 3 INFO 424 UW iSchool 11292007 Demo fluid links Fluid links summary LighTweighT conTexTual animaTed access To addiTional info Increase engagemenT wi Th source help reader choose links reduce disrupTion of following links Blur boundary beTween source amp desT compuTed glosses mulTiway links nesTed glosses Fluid Links user sTudy Research quesTions Is TexT animaTion disrupTive To eye behavior or reading How do differenT Techniques affecT The way readers l 7 V l use glosses MeThod Answer quesTions in Two hyperTexT corpora 6 condiTions InI39ne Margin Overay animaTed oTnoTe Popup No Gloss noT animaTed Use eyeTracker To focus on glass manipulaTions ll Pr 1 lt Ma 1mm Chang 7 h aimsmt CHIZDOG STudy condiTions Experimental seTup amp measures Also subjective questionnaire subjective discussion Polle Zellweger MacZell Consulting 4 INFO 424 UW iSChool 1 1292007 Eye movement visualization 0 Custom eyegtm Visual7aan m 3 n o m m m m 0 Helped LIS code Thousands of glass I I I E El Cl events quickly EDD Ell 2 3 4 5 E 7 agta timei rid Results effects of animation o Animated glosses drew the users39 eyes 4 o llllargins hard to read i me o Distant glosses i quotlatng without animation quot l footnotes were 5 sometimes missed t Glasses that opened mp he anchor iaanasssen iaarasmissss e used significantly more quickly than distant ones More results o Mouseover gloss activation is full lightweight almost 13 of detected gloss ev dwell wnlle followltlg lltlllt readlngrrelaled mouse motions nts were inadvertent t User preferences were higth varied and strong although glosses general yvalue suggests needfor care in animated designs t Browsing without glosses is like surfing blindquot i39quot e il l illl in ii as i Fluid Open Hypermedia o PrabemrAddi ersonal information third party material and sharing it if desired Fluid Open Hypermedia helps users add glosses that readers can open and close as V desired add glosses that can contain lltlks and glosses Hanan zettwsgei Gmsz kalay oanmammm instigate exienatrgenagingwnssisnasns WWszz 0 Streamline 6t reLls o Exploit visual percepti 655 provides rich annotato Push downquot go Improvements Augment existing Web pages directly no changes to work practice Can begin annotations from IE rightclick Context menu Can reuse appearance definitions r control of salience of anch n gloss s s animation permits reading while gloss is opening Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1 1292007 Controlling salience 0 z l0W canrra of salience of ammo farans 5mm r rue lyr1el msw use CSS in Specify mchur glass appearance cascade mhents easily 39um context if desired and allows linkannotation anchors to compose Pushdown animation 0 A ow readers 70 View glasses 1 comexr idea071555 antrow WIT1 millMal willslim animated openingdosing clarifies page changes glasses typically hidden reader can interactively upen push ownquot technique gradually reveals the glass below the anchor while he fullnwmglmes are pushed aewmhepsge in mm V V Problem with spreadsheets You see table amp numbers but not formulas Example editing the layout of a table Solution annotation in context Cell relationships Formula text 6 Annotations Cell relationships m m Cells formula cue 300 400 1 3 1 ulwnt Tll DWQDWE fll 500 39 1000 500 500 l 2500 l SllJlng calls Polle Zellweger MacZell Consulting INFO 424 UW iSChool 1 1292007 Formula text Interline compression if N500 SUMA2C3 Annotations gtltgtltgtlt quotW x xx 2000i 7 00 gtltgtltgtltgtltgtlt gtltgtltgtltgtltgtlt gtltgtltgtlt gtltgtggtltgtltgtlt gtltgtlt Demo fluid spreadsheets Fluid spreadsheets summary 0 Cell relationships lightweight interaction static d animated filled regions and table are visually distinct 9 Formula text use interline compression of free space 9 Annotation placed fluidly Polle Zellweger MacZell Consulting The Fluid Reader 0 Probem Making sense of hypertext narratives 0 The Fluid Reader supports finegrained hypertexts uses interactive animation to adjust conten moose mzuswmm Rem m Wuhanmuss gamma INFO 424 UW iSchool 1 1292007 XFR museum exhibit 9 San Jose c useum of Innovation 6 experiments in the future of Reading Fluid storytelling 0 Explores new forms of narrative when text becomes highly dynamic t Harry the Ape by Rich acid with the creatures that live in his fur and the creatures that live in their fur t Alternative sentence endin s more info explanations asidesjolltes lies 0 Resonance between content a medium nesting of creatures nesting of alternatives Demo fluid reader UI changes for museum setting 0 Touch screen no mouseover 0 Alternative endings vs annotations choice painin sentence es V A old ending disappears while new ending appears text sweeps dawn to guide reader to new ending color change a left margin show nesting Underlying tree structure initial View Polle Zellweger MacZell Consulting INFO 424 UW iSchool 1 1292007 Path B open first red triangle Path C open first triangle in each alternative Early authoring MS Word outline mode 0 Better for some paths Early authoring than for others MS Word outline mode Authoring requirements 0 Basic editing functions add remove content at any point 0 Ensure that ideas flow sensibly and sentence mechanics are properly observed across multiple finegrained nodes 0 View entire narrative and its structure 9 Compare Two or more paths in detail Polle Zellweger MacZell Consulting INFO 424 UW iSChool 11292007 The Fluid Wr rlwar Infemcmd Tmtable 66mpa r i ng two39parhs Editing fhe freemsz Edmhg Hm free l nble 3 5 me W AJ 1 39 V f Polle Zellweger MacZell Consulting 10 INFO 424 UW iSchool 1 1292007 Fluid Reader Wri rer O The Fluid Reader provldes continuouslyeuisible context supports fineegrained hyperlexls uses interactive animation to adJLlSl cometit O The Fluid Writer unaligned lreelablevlsuallzallon supports reading along multiple paths comparing focusing editing O Fluid Narratives myslery slory writing in progress reducing Cognlllvz load may per it more Emoymz l more attention to detail 77 for reader and author use allernallves for character plot development Problem wi rh reading o Humans are wired for spoken language o Reading and writing are artificial systems that must be taught Readers decode symbols into sounds o Learning to read is difficult 20 ol39Amerlcari adults are funcllonally lllllerale The basic decoding process o Visqu segmentation separate written word into groups of cat letters such that each group shows asingle sound C a t o Sound assignment choose a sound for each group Sound blending 32 Hence sounds l0 form S okzt l word qErigllsh le rlryrlghl p kat o Check and possibly try again that makes sertse in context k a t Why decoding English is hard o 43 sounds approx gt 26 letters letters lh show soutid th unrelated to t or h O Manymany mappings between sounds 6i leHers ow c t cu ow l oos p ul ch oes l o Other complications some letters cati am at a distancequot vuwelte plnrpn culrcl huprhp silent letters nuw tarrnigan dum aet r inu Polle Zellweger MacZell Consulting IN F0 424 UW iSchool 11292007 The PhonoGraphixTM approach 0 Developed recenle by McGuinnesses based on reading remediaTion experience analysis of English for spelling paTTerns frequency 0 STarT wiTh sounds Teach corresponding leTTers 0 Use every leTTer in The word le TorighT 0 Show The resulTs of visual segmenTaTion A cowboy39s job is to take care ofcallle A cowboy39s biggest helper is his horse Together a cowboy and his horse can look after hundreds of cows McGumnessCampG me u The Fluid Reading Primer 0 ExTend and improve upon PhonoC IraphixTM use animaTed Typography from Fluid DocumenTs 0 AnimaTe The decoding process on demand visual segmenTaTion sound assignmenT wiTh audio sound blending wiTh audio incremenTal help To promoTe aTTemst by reader 0 Careful animaTion design address common faulTy reading sTraTegies 0 Coded form is even closer To ordinary TexT no bold characTers or underlines To disTracT readers 3 Then snm wmte awoke anu w seven ime uwarves b Then Shaw Whlte awoke anu saw Seven Mile uwarves 7 Tn s 39 whne anke and saw seve lime awames W seven lillle d Then 5 DH Whlte awake and saw seven lime dwzwes 53 SEWquot We UWEWESV d Then 1 u ow While awoke and saw seven Iltlle dwzrves c Then 5 now Whlte awoke an a waives 2 Then 5 now wmte awake and saw seven little owarves STep 1 Visual segmenTaTion STep 2 Sound assignmenT iTh audio Then s n on While awake and saw savequot quotme waves Then Snowwhlieawoke and saw seven mile dwarves STe 3 Sound blendin P wifh audio 9 STep 2 for a dIsconTIguous e Demo fluid primer Fluid primer infrasTrucTure 0 STories wriTTen in plain TexT 0 DicTionary of words m h ough 1 01 BM V am 7 includes inf ecTed forms run runs ran running 0 Audio files of all 43 English sounds 0 Audio files of words Polle Zellweger MacZeII Consulting INFO 424 UW iSChool 1 1292007 Summary fluid hypertext 0 Fluid docunents ow annotations in context via lightweight animation of graphical characteristics t Fluid links provide hypertext link information 0 Fluid links study encouraging observations individual differences t Open fluid hypermedia r eadercreated annos on existing Web pages Summary fluid applications 0 Fluid spreadsheets expose underlying structure 0 Fluid fiction eader explore effects on reading stories writer show multiple alternatives together narratives explore effects on authoring 0 Fluid reading prim liow internal structure of words use audio Implementations t Fluid links 6i Fluid spreadsheets Java ka 1 XML marxup t Fluid fiction JavaJD l 2 oulllnzrbaszd authoring format o Fluid reading primer Java Jazz zoomable UI toolle a JDK 13 o Open fluid hypermedia nternet Explorer plugrln 0 corn 00M HTML Cascading Style sheets More information Fluid Links for informed and incremental link transitions HypertextQE The impact of Fluid Documents on reading and browsing an observational study CHI 2000 Fluid annotations in an open world HypertexlDl Fluid visualization of spreadsheet structures visual Languagesou The Fluid Reading Primer EbeMEDIA 2001 Reading and writing fluid hypertext narratives Hypertext02 Collaborators gtlterox PARC BaerQlChang Ken Flshktn Taken Igarasni 0 ufTu Bregurleemeyer5tanfurd 0 Susan Harkness pegli le R on Gold Anne MangenVulda UnvaullegeNurway Paula Newman university oanrhus N elsOlufBuuvn gJeno Brenbclzk urten Brelanerg Polle Zellweger MacZell Consulting
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'