Human-Computer Interaction Björn Hartmann University of California, Berkeley EECS, Computer Science Division CS10, Spring 2012
My Background CS10 - UC Berkeley Spring 2012 2
1997-2002 UPenn- NOT in Computer Science (cc) SnapsterMax CS10 - UC Berkeley Spring 2012 3 http://www.flickr.com/photos/myworldinpixels/558060045
1999-2004 Jay Haze Club Milk, Tokyo, 2003 CS10 - UC Berkeley Spring 2012 4
2004-2009 CS10 - UC Berkeley Spring 2012 5
2010 CS10 - UC Berkeley Spring 2012 6
What is HCI? CS10 - UC Berkeley Spring 2012 7
Human-Computer Interaction (HCI) Human • End-user of program • Others (friends, collaborators, coworkers) Computer • Machine program runs on • Often split: clients & servers Interaction • User tells the computer what they want • Computer communicates results
User Interfaces (UIs) Part of application that allows • People to interact with computer • Computer to communicate results Can include hardware design • Buttons, sliders, other sensors HCI = design, prototyping, implementation & h t p://www.reactable.com evaluation of UIs
Design Computer Science Applied Psychology CS10 - UC Berkeley Spring 2012 10
Rich Gold, The Plenitude , MIT Press CS10 - UC Berkeley Spring 2012 11
Iterative Design Cycle Design Evaluate Prototype Getting it right the first time is hard! CS10 - UC Berkeley Spring 2012
Understanding Users Observe existing practices Create scenarios of actual use CS247, Stanford, 2006 Build models to gain insight into work processes http://www-personal.umich.edu/~chrisli/m2.html CS10 - UC Berkeley Spring 2012 13
Prototyping Interfaces Rapidly build a mockup of your UI Low-fidelity techniques: Paper prototyping Video prototypes Moggridge, Designing Interactions, p.704 Interactive prototypes: HTML, Javascript, Flash, ... http://www.balsamiq.com/products/mockups/examples#wiki CS10 - UC Berkeley Spring 2012 14
Storyboarding CS10 - UC Berkeley Spring 2012
Evaluation Formative Are we building the right thing? What should be different in the next iteration? Summative Does it work? Is it better than existing solutions? Can this teach us something about how people or the world work? CS10 - UC Berkeley Spring 2012
h t p://www.laurasmith.info/UsabilityTest.jpg Techniques Analytically, expert walkthroughs, laboratory studies,... CS10 - UC Berkeley Spring 2012 18
Why study user interfaces? “The results show that in today's applications, an average of 48% of the code is devoted to the user interface portion. The average time spent on the user interface portion is 45% during the design phase, 50% during the implementation phase, and 37% during the maintenance phase. ” – Myers & Rosson, CHI’92 CS10 - UC Berkeley Spring 2012 20
History CS10 - UC Berkeley Spring 2012 21
ENIAC, UPenn, 1946 CS10 - UC Berkeley Spring 2012 22
Doug Engelbart & Bill English, SRI, 1963 CS10 - UC Berkeley Spring 2012 24
(cc) Flickr user John Chuang http://www.flickr.com/photos/13184584@N08/1362760884/ CS10 - UC Berkeley Spring 2012 25
Ivan Sutherland, Sketchpad, MIT, 1963/64 CS10 - UC Berkeley Spring 2012 27
CS10 - UC Berkeley Spring 2012 28
digibarn.com CS10 - UC Berkeley Spring 2012 29
Earlier today... CS10 - UC Berkeley Spring 2012 30
GRiD Compass in Space (1985) CS10 - UC Berkeley Spring 2012 31
IBM Simon 1992 CS10 - UC Berkeley Spring 2012 32
http://www.dipity.com/bjoern/personal CS10 - UC Berkeley Spring 2012 33
What has changed? “For every ant in the world today, there are 100 transistors. ” - Gordon Moore, 2003 CS10 - UC Berkeley Spring 2012 34
CS10 - UC Berkeley Spring 2012 35
Ratio Era HCI Focus Computers:People Mainframe 1:many Human Factors Computing Personal Psychology, Cog. 1:1 Computing Science Ubiquitous Collaboration, Rich many:1 Computing Interaction CS10 - UC Berkeley Spring 2012 36
Research Directions CS10 - UC Berkeley Spring 2012 37
(c) Dan O’Sullivan CS10 - UC Berkeley Spring 2012 38
Some saponas / Chris Harrison thing CS10 - UC Berkeley Spring 2012 39
CS10 - UC Berkeley Spring 2012 40
CS10 - UC Berkeley Spring 2012 41
CS10 - UC Berkeley Spring 2012 42
I'wish'I'knew'you I'like'your'picture You'are'cool I'was'paid'to'link'to'you I'want'your'reflected'glory Everybody'else'links'to'you I’d'vote'for'you Can'I'date'you? Are you my friend? We'met'at'a'conference'and'it'seemed'like'the'thing'to'do. yes no I'kind'of'like'you I'like'you I'really'like'you I'feel'socially'obligated'to'link'to'you I'know'you I'beat'you'on'Xbox'Live Hi,'Mom I'have'fake'alter'egos
Zipf / Power Law Distribution Source: Ed Chi, PARC
HCI Research at Berkeley CS10 - UC Berkeley Spring 2012 46
Crowds & Collective Intelligence Design & Touch & Development Physical Tools Computing
Crowd Work Platforms Crowd Applications (Shepherd CSCW12, (BribeCaster) Crowds & Turkomatic CSCW12) Expertise Sharing Tools Collective (HelpMeOut CHI10, Intelligence Local Crowd Work HyperSource CHI11) (Umati CHI12) Empirical Analyses (Stack Overflow CHI11) Design & Touch & Improved Tutorials Development Physical (ShowMeHow UIST11, Tools Computing d.mix CHI10) Smarter Code Editors Prototyping & Multitouch Applications (Stacksplorer UIST11) Manufacturing & Toolkits Tools (Eden CHI11, (d.tools UIST06, Pictionaire CSCW10, Exemplar CHI07, Proton CHI12) Fabbit)
Crowds & Collective Intelligence Design & Touch & Development Physical Tools Computing Multitouch Applications & Toolkits (Eden, Pictionaire, Proton)
Supporting Co-Located Design 50
Question: How might we help design team members manage information more fluidly across digital and physical boundaries ? 51
52
53
54
Table Infrastructure CS10 - UC Berkeley Spring 2012 55
The FourBySix Table CS10 - UC Berkeley Spring 2012 56
Object Manipulation XY World Z Translation Rotation Z Translation Uniform Scale One-Dimensional Arcball Scale Rotation Throw-and-Catch Local Z Rotation 57
58
Crowds & Collective Intelligence Local Crowd Work (Umati) Design & Touch & Development Physical Tools Computing
CS10 - UC Berkeley Spring 2012 60
CS10 - UC Berkeley Spring 2012 61
Typical tasks image labeling comment spam detection business listing verification human OCR CS10 - UC Berkeley Spring 2012 62
Can we crowdsource expert tasks by motivating groups of local experts? CS10 - UC Berkeley Spring 2012 63
CS10 - UC Berkeley Spring 2012 64
How about peer grading? CS10 - UC Berkeley Spring 2012 65
SnackShop 4600 CS10 - UC Berkeley Spring 2012 66
CS10 - UC Berkeley Spring 2012 67
CS10 - UC Berkeley Spring 2012 68
CS10 - UC Berkeley Spring 2012 69
CS10 - UC Berkeley Spring 2012 70
CS10 - UC Berkeley Spring 2012 71
1 Week 300+ Unique Users 7500+ Answers Graded $200 of Candy CS10 - UC Berkeley Spring 2012 72
How Good are the Answers? Gold Standard: 11 Expert Graders (TAs, CS PhD students) Sanity Check: Mechanical Turk (w/ and w/o qualification) CS10 - UC Berkeley Spring 2012 73
Worse than chance (20%) CS10 - UC Berkeley Spring 2012 74
11 Umatiers are better than 1 expert! More Turkers = more noise CS10 - UC Berkeley Spring 2012 75
www.cs.berkeley.edu/~bjoern CS10 - UC Berkeley Spring 2012 76
Human-Computer Interaction Redux ...is concerned with the design, implementation, and evaluation of user interfaces. Mobile, social & ubiquitous computing: now is an incredibly exciting time to work in HCI! Want more? CS160 User Interface Design Course Thread in Human-Centered Design http://coursethreads.berkeley.edu CS10 - UC Berkeley Spring 2012 77
bjoern@eecs.berkeley.edu www.cs.berkeley.edu/~bjoern
Recommend
More recommend