norman sadeh 54m smart phones in q1 2010...2010/05/27 · norman sadeh, “m-commerce:...
TRANSCRIPT
ECom-IComp Experts Address – Hong Kong University – May 27,2010
Mobile Commerce: What Will the App Store of the Future Look Like?Store of the Future Look Like?
Norman SadehDirector, Mobile Commerce Lab.School of Computer ScienceSchool of Computer ScienceCarnegie Mellon University
Explosion in Mobile Device Ownership
By the end of 2010 – source: ITU Feb 2010
5B cell phone users 1B mobile broadband users
Smart phone sales 54M smart phones in Q1 2010
–50% YoY growth
B t l t d h t f But also towards a host of wearable devices & smart appliances
Copyright © Norman Sadeh, 2005-2010
Orders of Magnitude (US only)
•Over $40B/year in mobile data revenue2 % f ll l S (l 2009)
Copyright © Norman Sadeh, 2005-2010
•Approx. 25% of all wireless revenues in US (late 2009)
…Also in Hong Kong…
Q1 figures from SmartoneSmartone: Q1 2010
Copyright © Norman Sadeh, 2005-2010
Source: http://www.smartone.com.hk/about/investor/results/english/0910interim_present.pdf
i-Mode showed us the way…10+ years ago
i-Mode Subscription – Approx. $3/month
User Traffic Charge – Approx $0.003/packet
9% C i i i b i ti9% Commission on service subscription
91% of service Subscription
$1-3/month
Content Provider
Copyright © Norman Sadeh, 2005-2010
i-Mode Explained Compelling and transparent business model
contributed to a critical mass of contributed to a critical mass of compelling content Reasonable share of revenue
Addressing usability and trust Addressing usability and trust Directory of services
Quality content, better phones, etc.
Leveraged DoCoMo’s billing g ginfrastructure
Copyright © Norman Sadeh, 2005-2010
Growth of i-Mode Customer base
Source: NTTDoCoMo
Copyright © Norman Sadeh, 2005-2010
Apple Re-Discovered i-Mode’s Recipe
“More is not always better”
The i-Phone classic didn’t have 3G, GPS or a good cameraGPS or a good camera
….but what it offered was USABLE!
Copyright © Norman Sadeh, 2005-2010
Trust & Usability: Beyond the Access Device
By 2007, WAP and J2ME were widely available already butavailable already but… Typing a URL in a mobile browser remained
h lla challenge
Expecting users to download and pay p g p yfor apps was unrealistic
U bilit i i di h &Usability issues: Finding the app & downloading it
Trust issues: “I’m not giving my credit card details to this company I never heard of”
Copyright © Norman Sadeh, 2005-2010
The Advent of the App Store Model
Addresses key usability and trust issues: Directory & search functionality
Ratings Ratings
Payment
Content quality, including security
Copyright © Norman Sadeh, 2005-2010
The App Store Today
50M iPhones, 35M iPod Touch, 1M iPads
A d 200 000 t d Around 200,000 apps today
Over 4 billion downloads
Users spend an average of 30 minutes d i t ti ith a day interacting with apps
Source: Apple
…and a number of others have followed Android Market Symbian Ovi etc Android Market, Symbian Ovi, etc.
Copyright © Norman Sadeh, 2005-2010
Another Measure of Success
Copyright © Norman Sadeh, 2005-2010
Source: AdMob, Fall 2009
An Increasingly Rich Ecosystem
Thousands of location-based apps
Social networking apps
…this is just the beginning…
Copyright © Norman Sadeh, 2005-2010
Zipcar App: Mobile Device As Remote
Copyright © Norman Sadeh, 2005-2010
Source: App Store
Mobile Device Coupled with External Sensor
Copyright © Norman Sadeh, 2005-2010
Source:http://www.apple.com/ipod/nike/
Cell phones as Sensor Nets
TrafficWhere is everyone right now?Analyze consumer behaviorAnalyze consumer behavior
Copyright © Norman Sadeh, 2005-2010
Source: Sense Networks
So What Are the Things We Can’t gDo Yet?
A Valentine’s Day Scenario
Valentine’s Day ReminderCharacters: John & Mary
Buy Flowers
C J & y
Intentionall “m ndane”Book Performance
Intentionally “mundane”but not trivial if you want to:
Reserve Table at Restaurant•Access calendar info•Food preferences
Arrange for Transportation•Location•Home address
Send invitation & Set up Reminder•…not just John’s but also Mary’s
Copyright © Norman Sadeh, 2005-2010
N. Sadeh, “M-Commerce: Technologies, Services and Business Models”, Wiley 2002
What are the Missing Pieces?
Higher degree of context awareness Dialogues with users Abilit to disco e & inte act ith Ability to discover & interact with
different services Not just services identified ahead of
time Security & Privacy New Revenue Sharing Models
Copyright © Norman Sadeh, 2005-2010
In Practice…
The explosion in location-based services can be attributed to the availability of can be attributed to the availability of location APIs Apple SDK Android Symbian etc Apple SDK, Android, Symbian, etc.
Many apps are developed in 6-8 weeks Developers don’t have the resources to
develop an App for our Valentine’s Day p pp yscenario unless … new layers of interoperability are …new layers of interoperability are
introduced…
Copyright © Norman Sadeh, 2005-2010
MyCampus
Campus as “everyday life microcosm”Campus as everyday life microcosmEnhance campus life through context-aware
services accessible from mobile devices
A growing collection of context-aware A growing collection of context-aware agents/applications that: Users can acquire/subscribe to/invokeCustomize themselves through automated Customize themselves through automated
identification and access of relevant contextual resources
Copyright © Norman Sadeh, 2005-2010
contextual resources
What are Semantic Web Technologies?
Collection of languages (W3C’s RDF/RDFS, OWL, etc.) to: Define ontologies Collection of concepts and relationships
bet een these concepts in a gi en domainbetween these concepts in a given domain e.g. Space, Calendar Activities, Access
Devices User Preferences etcDevices, User Preferences, etc. Facts and rules defined in relation to these
ontologies Shared ontologies enable disparate
applications to communicate with one thanother.
Content becomes machine understandable
Copyright © Norman Sadeh, 2005-2010
Personal Resource Ontology: An Example
PersonalResource
IS-A
Onto
Activity Information
Resource
LocationInformation
ResourceList of Friends
ologyCampus Location Microsoft Outlook
INSTANCE
Sk h k Wi l
Campus LocationTracking
Microsoft OutlookCalendar FactGoogle
Skyhook WirelessLocation Tracking
A t ti id tifi ti f il bl f t t l i f ti
tsCalendar
Copyright © Norman Sadeh, 2005-2010
Automatic identification of available sources of contextual information
Benefits
E t f t t l Easy to re-use sources of contextual information across different applications
Can adapt to resources available for different users at different points in time
Can discover and interact with new Can discover and interact with new services
Copyright © Norman Sadeh, 2005-2010
What This Might Look Like
Copyright © Norman Sadeh, 2005-2010
Each entity has its own set of policies & policy evaluation agents
How Do You Interact With This?
…Let’s just look at the privacy side
Android Expects You to Agree Upfront
…and never change your mind…
Copyright © Norman Sadeh, 2005-2010
Apple Expects You …
to answer the same …to answer the same question 1000 times…
…with multi-tasking in OS4, they’ll show you which app they ll show you which app accessed your location in the past 24 hours……even if it’s too late…
Copyright © Norman Sadeh, 2005-2010
We Need Something better
Neither approach scales
N ith h i bl Neither approach is usable
Copyright © Norman Sadeh, 2005-2010
In practice…
People have context-dependent preferences…
Green: ShareRed: Don’tRed: Don t
Copyright © Norman Sadeh, 2005-2010
www.locaccino.org
Copyright © Norman Sadeh, 2005-2010
Feedback Through Audit Logs
Copyright © Norman Sadeh, 2005-2010
Evaluating the Usefulness of Auditing
Examining Users’ Privacy Rules at the end of the study
Copyright © Norman Sadeh, 2005-2010
Dialogues & User-Oriented Learning NewTechnology
Policy My colleagues can see my locationon weekdays between 8am and 5pm
JanePolicyVisualization
Creation
Jane and Eric are late for our BobPolicy
Policy Enforcement
Jane is in Oakland but I can’t access Eric’s location
meeting. Show me where they are!
B b’
PolicyEnforcingEngines
access Eric s location Bob’sPhone
Why couldn’t Bob see where I was?Eric
Policy Auditing &
Bob is a colleague. So far only your friends can see where you are
Explanation
Refinement What if my colleagues could see my location too?Eric
In the past you denied access to ll S
Dialog
Learningfrom the
Copyright © Norman Sadeh, 2005-2010
Time your colleague Steve
OK, make it just my superiors
from the past
Likely to See a New Privacy Layer
UsersApps developers
Open de elope s APIOpen developers API
User-Controllable Privacy Platform
Policy enforcement engine,i l diti
y
Location middleware provide
Platform providers
incl. auditing
Platform providersData providers (maps, P.O.I)
Copyright © Norman Sadeh, 2005-2010
…This New Privacy Layer
Will have to: S t ich conte t sensiti e Support rich, context-sensitive
privacy preferences
Be capable of supporting auditing and dialogue with users
Be scalable
E t ll t d t id f …Eventually extend to a wide range of personal and contextual attributes
Copyright © Norman Sadeh, 2005-2010
Concluding Remarks The App Store model addresses fundamental
trust and usability issues Yet, for new context-aware and pervasive
computing applications to emerge, the model will need to evolvewill need to evolve
…Introducing new layers of interoperability: For developers For developers Contextual Attributes, Privacy & Security Support for smart dialogues (incl some machine Support for smart dialogues (incl. some machine
learning to incrementally build/refine user models) For other partners The cell phone is only one piece of the emerging
ecosystem – appliances & services d i h t f h i d l
Copyright © Norman Sadeh, 2005-2010
… and richer set of revenue sharing models
Expect….
Many more applications that: T k d t f i i l ich Take advantage of an increasingly rich
ecosystem
Increasingly provide for more intelligent interactions with users …Eventually turning into personal
assistants that help users with pincreasingly complex tasks
This transformation will take place over This transformation will take place over many years…
Copyright © Norman Sadeh, 2005-2010
Copyright © Norman Sadeh, 2005-2010
Relevant Websites
www.mcom.cs.cmu.edu/ l i www.locaccino.com http://mcom.cs.cmu.edu/user-controllable-
i d i /security-and-privacy/ www.zipano.com
Copyright © Norman Sadeh, 2005-2010
Relevant Publications Norman Sadeh, “M-Commerce: Technologies, Services and Business
Models”, Wiley 2002 Norman Sadeh, Jason Hong, Lorrie Cranor, Ian Fette, Patrick Kelley, Madhu
Prabaker, and Jinghai Rao. Understanding and Capturing People’s Privacy Prabaker, and Jinghai Rao. Understanding and Capturing People s Privacy Policies in a Mobile Social Networking Application Journal of Personal and Ubiquitous Computing 2009.
J. Tsai, P.G.Kelley, L.F.Cranor and N.M. Sadeh, “Location Sharing Technologies: Privacy Risks and Controls”, to appear in “I/S: A Journal of Law and Policy for the I f ti S i t ” 2010Information Society”, 2010.
Ramprasad Ravichandran, Michael Benisch, Patrick Gage Kelley, and Norman M. Sadeh. Capturing Social Networking Privacy Preferences: Can Default Policies Help Alleviate Tradeoffs between Expressiveness and User Burden? PETS ’09Burden? PETS 09.
Janice Tsai, Patrick Kelley, Paul Hankes Drielsma, Lorrie Cranor, Jason Hong, and Norman Sadeh.Who’s Viewed You? The Impact of Feedback in a Mobile-location System.CHI ’09.CHI 09.
Patrick Kelley, Paul Hankes Drielsma, Norman Sadeh, Lorrie Cranor. User Controllable Learning of Security and Privacy Policies. AISec 2008.
Michael Benisch, Patrick Gage Kelley, Norman Sadeh, Lorrie Faith Cranor, Capturing Location-Privacy Preferences: Quantifying Accuracy and User-Capturing Location-Privacy Preferences: Quantifying Accuracy and User-Burden Tradeoffs. CMU-ISR Tech Report 10-105, March 2010
Norman Sadeh, Fabien Gandon and Oh Buyng Kwon. Ambient Intelligence: The MyCampus Experience School of Computer Science, Carnegie Mellon University, Technical Report CMU-ISRI-05-123, July 2005.
Copyright © Norman Sadeh, 2005-2010
University, Technical Report CMU ISRI 05 123, July 2005.
Acknowledgements
Research funded by the US National Science Foundation the US Army Science Foundation, the US Army Research Office, CMU CyLab, Microsoft, Google Nokia FranceTelecom and ICTIGoogle, Nokia, FranceTelecom, and ICTI
The User-Controllable Privacy Platform on top of which Locaccinoon top of which Locaccinois built is now commercialized by Zipano T h l iTechnologies.
Copyright © Norman Sadeh, 2005-2010