OMG — It's all connected

My presentation from eyeo 2012.

Transcript

1. MORITZ STEFANER TRUTH&BEAUTY OPERATOR OMG— IT’S ALL CONNECTED eyeo 2012 | Minneapolis
2. MORITZ STEFANER TRUTH&BEAUTY OPERATOR 1999 – 2002 2002 – 2005 2005 – 2007 2007 – 2009 2004 – working for OECD, World Economic Forum, FIFA, Skype, ... Web designer BSc Cognitive Science MA Interface Design Research assistant FH Potsdam Freelance Truth & Beauty Operator
4. MAKE THE COMPLEX COMPREHENSIBLE
5. NETWORKS
6. EVERYTHING IS A NETWORK OMG!
7. Jacob Moreno, 1932 http://en.wikipedia.org/wiki/Jacob_L._Moreno
8. Mark Lombardi
9. Map of the internet http://www.opte.org/maps/
14. http://picturesfrommywindow.blogspot.com.es/2010/09/hair-balls.html
15. Organic Link Network Moritz Stefaner, 2002 FOCUS & CONTEXT SPATIAL COHERENCE
16. Relation Browser Moritz Stefaner, 2004 THE SMALL PICTURE
17. LOOK MA, NO LINES Elastic Tag Maps Moritz Stefaner, 2007
18. Global Risk Report 2012 Moritz Stefaner for WEF, 2011
19. maeve http://portal.mace-project.eu/maeve MACE project team, FH Potsdam
20. maeve http://portal.mace-project.eu/maeve/ FLUID, TANGIBLE INTERACTION USER CONTRUCTED VIEWS
21. Josh On , 2004 | http://theyrule.net
22. Prof. Boris Müller Jonas Loh Stephan Thiel Larissa Pschetz Steffen Fiedler
23. Till Nagel
24. Max Planck Research Networks http://max-planck-research-networks.net With Christopher Warnow / onformative
28. Quantitative analysis with Tableau name Year 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 Max-­Planck-­Institut  für  Physik  (Werner-­ Heisenberg-­Institut) Max-­Planck-­Institut  für  extraterrestrische Physik Max  Planck  Institut  für  Kernphysik Max  Planck  Institute  for  Solid  State Research Max-­Planck-­Institut  für  Radioastronomie Max-­Planck-­Institut  für  Astronomie Max  Planck  Gesellschaft  zur  Förderung  der Wissenschaften Max-­Planck-­Institut  für Sonnensystemforschung Max  Planck  Institute  of  Biochemistry Max-­Planck-­Institut  für  Astrophysik 0% 10% 20% 30% 40% %  of  Total  Weight 0% 5% 10% 15% 20% %  of  Total  Weight 0% 10% 20% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 5% 10% 15% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 2% 4% 6% 8% %  of  Total  Weight 0% 2% 4% 6% %  of  Total  Weight Sheet  1 name Max-­Planck-­Institut  für  Physik  (Werner-­Heisenberg-­Institut) Max-­Planck-­Institut  für  extraterrestrische  Physik Max  Planck  Institut  für  Kernphysik Max  Planck  Institute  for  Solid  State  Research Max-­Planck-­Institut  für  Radioastronomie Max-­Planck-­Institut  für  Astronomie Max  Planck  Gesellschaft  zur  Förderung  der  Wissenschaften Max-­Planck-­Institut  für  Sonnensystemforschung Max  Planck  Institute  of  Biochemistry Max-­Planck-­Institut  für  Astrophysik Max  Planck  Institute  for  Chemistry  (Otto  Hahn  Institute) Max  Planck  Institute  for  Molecular  Genetics Max  Planck  Institute  for  Polymer  Research Max-­Planck-­Institut  für  Plasmaphysik Max  Planck  Institute  for  Metals  Research Fritz  Haber  Institute  of  the  Max  Planck  Society Max  Planck  Institute  for  Biophysical  Chemistry  (Karl  Friedrich  Bonhoeffer  Institute) Max-­Planck-­Institut  für  Quantenoptik Max-­Planck-­Institute  for  the  Physics  of  Complex  Systems Max-­Planck-­Institut  für  Psychiatrie The  trend  of  %  of  Total  Weight  for  Year  broken  dow n  by  name.    Color  show s  details  about  name.    Details  are  show n  for  name.  The  view  is  fil-­ tered  on  name  and  Year.  The  name  filter  has  multiple  members  selected.  The  Year  filter  keeps  22  members.  Percents  are  based  on  each  col-­ umn  of  the  table. name 1993 1998 2003 2008 Calculation1 Max-­Planck-­Institut  für  Physik  (Werner-­ Heisenberg-­Institut) Max-­Planck-­Institut  für  extraterrestrische Physik Max  Planck  Institut  für  Kernphysik Max  Planck  Institute  for  Solid  State Research Max-­Planck-­Institut  für  Radioastronomie Max-­Planck-­Institut  für  Astronomie Max  Planck  Gesellschaft  zur  Förderung  der Wissenschaften Max-­Planck-­Institut  für Sonnensystemforschung Max  Planck  Institute  of  Biochemistry Max-­Planck-­Institut  für  Astrophysik 0% 10% 20% 30% 40% %  of  Total  Weight 0% 5% 10% 15% 20% %  of  Total  Weight 0% 10% 20% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 5% 10% 15% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 5% 10% %  of  Total  Weight 0% 2% 4% 6% 8% %  of  Total  Weight 0% 2% 4% 6% %  of  Total  Weight Sheet  1 The  trend  of  %  of  Total  Weight  for  Calculation1  broken  dow n  by  name.    Details  are  show n  f or  name.  The  data  is  filtered  on  Year,  w hich  keeps  22  members.  The  view  is  filtered on  name,  w hich  has  multiple  members  selected.  Percents  are  based  on  each  column  of  the  table.
30. Data analysis with Gephi, custom tools
32. MPI für Physik (Werner-Heisenberg-Institut) MPI für extraterrestrische Physik MPI für Kernphysik MPI für Radioastronomie MPI für Astronomie MPI for Solid State Research MPI für Astrophysik MPI für Sonnensystemforschung Max Planck Gesellschaft zur Förderung der Wissens… MPI für Plasmaphysik MPI for Chemistry (Otto Hahn Institute) MPI of Biochemistry MPI for Metals Research MPI for Molecular Genetics MPI for Polymer Research Fritz Haber Institute of the Max Planck Society MPI for Biophysical Chemistry (Karl Friedrich Bon… MPI für Quantenoptik MPI for Gravitational Physics (Albert Einstein In… MPI for the Physics of Complex Systems MPI fur Kolloid Und Grenzflachenforschung Potsdam MPI für Psychiatrie MPI for Meteorology MPI für Biogeochemie Jena MPI of Microstructure Physics MPI for Evolutionary Anthropology MPI for Developmental Biology MPI for Gravitational Physics (Hannover) (Albert … Max-Planck Institut für Chemische Physik fester S… MPI für molekulare Physiologie MPI für Medizinische Forschung Heidelberg MPI für Züchtungsforschung MPI für marine Mikrobiologie MPI für Dynamik Und Selbstorganisation Göttingen MPI für Experimentelle Medizin MPI of Immunobiology MPI für Informatik MPI für Kohlenforschung MPI für Kognitions- und Neurowissenschaften MPI for Biological Cybernetics MPI of Molecular Cell Biology and Genetics MPI für Hirnforschung MPI for Infection Biology MPI fur Chemische Okologie MPI für Bioanorganische Chemie MPI of Molecular Plant Physiology MPI of Neurobiology MPI of Biophysics MPI für Eisenforschung MPI für Bildungsforschung MPI for Ornithology (Former: Max Planck Research … MPI für Terrestrische Mikrobiologie MPI for Mathematics in the Sciences MPI für Herz- und Lungenforschung Max-Plack-Institut für Neurologische Forschung Kö… MPI für molekulare Biomedizin MPI fur Mathematik MPI for Psycholinguistics MPI for Dynamics of Complex Technical Systems MPI für Limnologie Max-Plack-Institut für demografische Forschung MPI für Verhaltensphysiologie MPI für Ökonomik Friedrich Miescher Laboratory of the Max Planck S… Max-Planck-Arbeitsgruppen für strukturelle Moleku… MPI für Evolutionsbiologie International Max Planck Research School on Earth… MPI for Research On Collective Goods MPI fur Neurologische Forschung Koln MPI for the History of Science MPI für Gesellschaftsforschung MPI für Virusforschung International Max Planck Research School for Comp… Max Planck Inst. Evol. Anthropol
36. Layouts
37. Interaction tests
38. DIVIDE & CONQUER FIND THE SMALL DATA INSIDE THE BIG DATA
39. We would be nowhere without free + open source software.
40. well-formed.eigenfactor Moritz Stefaner & Eigenfactor Team, 2009 | http://well-formed.eigenfactor.org
46. Hierarchical Edge Bundling Danny Holten: Hierarchical Edge Bundles: Visualization of Adjacency Relations in Hierarchical Data, IEEE Transactions on Visualization and Computer Graphics, Vol. 12, No. 5, September/ October 2006 Danny Holten and Jarke J. van Wijk: Force-Directed Edge Bundling for Graph Visualization, Eurographics/ IEEE-VGTC Symposium on Visualization 2009, Volume 28 (2009), Number 3
47. GROUP AND ORDER ROUTE THOSE EDGES
48. LOOK MA, NO LINES PHYSICAL METAPHOR
49. Müsli Ingredient Network Moritz Stefaner for mymuesli.com, 2012
52. HIGHLIGHT THE UNUSUAL
56. Force-directed Helps to identify overall gestalt, indentifying clusters usually a big mess Matrix Helps to see all links, identify link strengths ordering is crucial no perception of “gestalt” Radial layout Regular, aesthetic placement helps to identify individual nodes and links, ordering, grouping are crucial no relation between link length and strength CHOOSE THE RIGHT TRADE-OFF
57. Jacques Bertin: Domino Image credit: Nathalie Henry, Jean–Daniel Fekete
59. Jacques Bertin: Semiologie Graphique s
60. A User Study on Visualizing Directed Edges in Graphs Holten, van Wijk Figure 2. The six single-cue directed-edge representations used in the first user experiment. (a) “arrow”; (b) “light-to-dark”; (c) “dark-to-light”; (d) “green-to-red”; (e) “curved”; (f) “tapered”.
61. IFFERENCES o networks, and contrasts distinct differences. Unlike the hairballs, Nagel et al. 2012 - Sankey Arcs Krzywinski et al. 2011 - Hive Plots Wattenberg 2006 - Pivotgraph
62. Things to work with Layout Implicit vs explicit - do the axes have meaning, or is just the relative positioning important? Interaction and behavior Focus selection Start with blank slate, then place items Filtering, brushing, zooming, etc. Visual markers for nodes and edges Labels
63. Things to think about Does it really, really need to be a network? What is more important: Big picture or local neighborhood? Do we need really need lines? If so, how can we make them as unobtrusive as possible How can the user set focus points? How do we help with grouping and ordering?
64. THE HUMAN TOUCH
65. Global Risks 2012 Seventh Edition An Initiative of the Risk Response Network Insight Report
66. 2012: Global Risk Report Recurring liquidity crises Unmanageable inflation or deflation Failure of climate change adaptation Irremediable pollution Mismanaged urbanization Persistent extreme weather Critical fragile states Pervasive entrenched corruption Terrorism Backlash against globalization Food shortage crises Unmanaged migration Cyber attacks Massive digital misinformation Massive incident of data fraud or theft Chronic labour market imbalances Land and waterway use mismanagement Prolonged infra– structure neglect Extreme volatility in energy and agriculture prices Severe income disparity Unforeseen negative con- sequences of regulations Major systemic financial failure Unsustainable population growth Critical systems failure Global governance failure Rising greenhouse gas emissions Chronic fiscal imbalances Mismanagement of population aging Failure of diplomatic conflict resolution
67. 2011 Global Risk Report: Risk Interconnection Map
69. LABELING IS THE BLACK ART OF INFORMATION DESIGN
71. http://bl.ocks.org/1377729
75. http://moritzstefaner.github.com/gridexperiments
76. http://datastori.es podcast together with Enrico Bertini
81. emoto is a unique data art project that sets out to visualise the worldwide emotional response to the Olympic Games 2012. Created by Moritz Stefaner, Drew Hemment, Studio NAND. A FutureEverything project with MIT SENSEable City Lab for the Cultural Olympiad programme and London 2012 Festival. Supported by Datasift, co-sponsored by GE and funded by Arts Council England and WE PLAY/ Legacy Trust UK. info@emoto2012.org @emoto
82. emoto :) :( The games Social media response Sentiment analysis, classification Real time visualization Post-hoc analysis Projection Immersive experiences
83. THANKS! Q? A! http://moritz.stefaner.eu @moritz_stefaner