Computers and Business

David L. OlsonJames & H.K. Stuart Professor and

Chancellor’s ChairUniversity of Nebraska-Lincoln

Moore’s Law

• Intel: – Noted that the number of transistors possible on

integrated circuits doubled every so often• 2 years? 18 months?

• Seems to apply to about everything computer—technology related– Computing speed– Storage capacity– Transmission capability

Computer progress(see Moore’s Law)

• 1988 Intel 386D processor ran at 8.5 MIPS– First IBM PCs, Microsoft Windows

• 1992 Intel 486DX ran at about 54 MIPS– PCs supported Windows 3.1

• 1999 Intel Pentium II over 1,300 MIPS• 2008 Intel Core 2 Extreme 59,000 MIPS

Linear vs. Exponential

0

500

1000

1500

2000

2500

3000

3500

4000

4500

1 2 3 4 5 6 7 8 9

Period

Linear

Exponential

Ray KurzweilThe Singularity is Near: When Humans

Transcend Biology

NY: Penguin Books, 2005Singularity – future period when pace of

technological change will be so strong that human life will be irreversibly transformedWhen exponential crosses linear

Epochs• Epoch one: Physics and chemistry

– information in atomic structures• Epoch two: Biology

– information in DNA (DNA evolves)• Epoch three: Brains

– information in neural patterns (brains evolve)• Epoch four: Technology

– information in hardware and software designs (technology evolves)• Epoch five: Merger of technology and human intelligence

– integration in exponentially expanding base• Epoch six: The universe wakes up

– patterns of energy & matter saturated with intelligent processes & knowledge (vastly expanded human intelligence spreads through the universe)

A Theory of Technology Evolution:The Law of Accelerating Returns

• The nature of order: – paradigm shifts are major changes in methods and

intellectual processes to accomplish tasks

• Each paradigm shift follows S curve– Slow growth, rapid growth, leveling off

• Under accelerating returns, paradigm shifts occur faster

• 15 different lists with a lot of overlap– All show decreasing gaps in time

Achieving the Computational Capacity of the Human Brain

• We’ve had 5 paradigms of computing: – electromechanical calculators– relay-based computing– vacuum tubes– discrete transistors– integrated circuits

• Singularity representing profound & disruptive transformation in human capability in 2045– nonbiological intelligence created will be 1 billion times

more powerful than all human intelligence today

Achieving the Software of Human Intelligence:How to Reverse Engineer the Human Brain

BRAIN• Very slow• Massively parallel• Rewires itself• Details random• Emergent• Imperfect• Evolution• Patterns important

COMPUTERS• Fast (electricity)• Can be parallel• Humans wire• Deterministic• Programmed

Genetics, Nanotechnology, Robotics

• Designer baby boomers• Can we really live forever?• Somatic gene therapy– enabling us to change genes by infecting them with new

DNA, creating new genes• Kurzweil speaks badly of manipulating human life– but takes 250 pills per day

• Solving world hunger– we would not be creating the entire animal but rather

directly producing the desired animal parts or flesh.

Nanobots - 2004

• used for mission-critical software systems– control nuclear-power plants– 911– ICU– land airplanes– guide missiles

Artificial Intelligence

• Expert systems, Bayesian nets, Markov models, neural nets, genetic algorithms, recursive search

• Applications– military & intelligence, space exploration,

medicine, science & math, business finance & manufacturing, speech & language, entertainment & sports

Singularity our destiny

• We can live forever– evidently freezing an existing population that will

never die nor procreate • We are becoming Cyborgs• Transfer to nonbiological experience• Change– Warfare– Work

Intertwined Promise & Peril• Environmentalists

– world that has enough wealth and enough technological capability– should not pursue more

• The many benefits of progress can always be turned to evil purposes.

• The precautionary principle:– If the consequences of an action are unknown but judged by some scientists

to have even a small risk of being profoundly negative, it’s better to not carry out the action than risk negative consequences.

• Golden rice genetically modified to contain high levels of beta-carotene – precursor to vitamin A, needed to stop African children from going blind. – Greenpeace strongly opposes Golden rice as genetically modified.

Criticisms

• Malthusian– exponential trends don’t last forever• Computer limits not very limiting

• Software limitations– Computer instability, slow responsiveness• Intelligent algorithms are on the way

• Analog processing– Computers use digital medium• Idea is to combine digital & analog

Criticisms

• Ontology– Can a computer be conscious?

• What difference – human constructs

• Rich-poor divide– Technology makes rich richer

• History of technology shows evolution from inefficient/expensive to efficient/cheap (cell-phone)

• Holism– Biological is holistic, computers deterministic & modular

• With AI can use both

In Conclusion…

• We are building machines with– Powers far greater than the sum of their parts– By combining self-organizing design principles of

the natural world– With accelerating powers of human-initiated

technology• The Singularity will occur about 2040

Emile Zola (1901) Work

• “The machines did everything….What an elevating sight: an army of obedient mechanical laborers with never-tiring stamina…that were now the worker’s friends, instead of their competitors….They liberated instead of exploiting him. While he rested, they did his work.”– Aren’t machines (computers) wonderful?

Martin Ford(2009) Create Space Independent Publishing Platform

• The Lights in the Tunnel: Automation, Accelerating Technology and the Economy of the Future – What if technology progresses to the point where

a substantial fraction of the jobs now performed by people are instead performed autonomously by machines or computers?

– People who rely for jobs for income the same as those who buy the products produced

Productivity Paradox

• Economy hasn’t had expected productivity gains from computers– Greenspan – no noticeable benefit– Brynjolffson – of course they benefit

Erik Brynjolfsson and Andrew McAfee 2011 Digital Frontier Press

Race Against The Machine: How the Digital Revolution is Accelerating Innovation,

Driving Productivity, and Irreversibly Transforming Employment and the Economy

• Computer progress advancing exponentially• AFFECT ON– Jobs– Skills– Wages– The Economy

US

• Great economic changes– Wages too high

• Outsourcing– Computer programming (service) to India– Manufacturing to China

• Technology– Robotics – no health benefits, no vacations, no complaints

• Computers– ERP systems replacing multiple legacy systems

» Layoff most human IT people– Business Analytics– BIG DATA

World Wide Web

• Innovations– eBay & Amazon Marketplace

• Over 600,000 earn living by new products for worldwide market

– Apple’s App Store, Google’s Android Marketplace• Easy to implement mobile applications, distribute them

– Threadless• Customers create & sell t-shirt designs

– Heartland Robotics• Robots-in-a-box• Small businesses can access inexpensively

Examples of Innovation

• Crowdsourcing– Amazon • Cloud

– Innocentive• Commercialized research & development

Amazon collaborative systems

Crowdsourcing: Mechanical TurkCloud: EC2

Amazon.com

• Many unprofitable years in the 1990s• Now highly successful• USER PARTICIPATION– Reader reviews, ratings– Personalized suggestion system• Once toyed with personalized pricing

• CLOUD– Can rent storage, software

Mechanical Turk• 200,000 participants

– Use free time to work for customers– Amazon the broker– Started in 2006

• EXAMPLE– Powerset search site wanted feedback – Through Mechanical Turk, got 100 people to rate results for 2 hours, at

$2/hour– Powerset then started Dolores Labs to assess accuracy and speed of

Mechanical Turk participants.• Cost $2,000, estimated $30,000 for full-time staffers

– Typical worker• Guitar teacher supplements income by 15 hours of work on Mechanical Turk at

$3/hour

Mechanical Turk HitsTask Requestor Expiration

dateReward Time

allottedHits available

Identify Arabic dialect in text An individual

31 Dec 2010

$0.05 per phrase

15 minutes

14196

PIO verification for US cities Nutella42 17 Dec 2010

$0.08 per city

30 minutes

1617

Find political bias University 9 Dec 2010 $0.05 60 minutes

1097

Translate English to Urdu An individual

23 Dec 2010

$0.70 24 hours 1002

Find URL for college club directory

techlist 7 Dec 2010 $0.05 15 minutes

1001

Judge quality of search results

Dolores Labs

13 Dec 2010

$0.08 60 minutes

996

Find product information for electronics

Dolores Labs

13 Dec 2010

$0.12 60 minutes

995

Search for terms Dolores Labs

13 Dec 2010

$0.55 60 minutes

962

Choose best search result Dolores Labs

13 Dec 2010

$0.10 60 minutes

951

Mechanical Turk Examples

• Select correct spelling for given search terms• Evaluate website suitability for a general

audience• Rate search results for given key words• Evaluate product similarity• Select appropriate category for given products• Categorize an article’s tone• Translate from one language to another

Informed Consent FormHeading Content

Purpose of research study:

To collect human annotations to improve automatic translation of Arabic into other languages.

Benefits: Can benefit society by improving how computer process human languages.

Risks: None

Voluntary participation:

You can discontinue at any time without penalty by clicking on “Return Hit”

We may end your participation if:

You don’t demonstrate adequate knowledge of the language, or are not following instructions, or if your answers significantly deviate from known translations.

Confidentiality: The only information about you retained will be a WorkerID serial number and your IP address.

Questions/concerns: You may e-mail questions to the principal investigator. How may contact Johns Hopkins University Institutional Review Board if you feel unfairly treated.

Clicking on the “Accept HIT” button

Indicates your understanding of this information. You have not waived any legal rights you otherwise would have as a participant in a research study.

EC2

• Elastic Compute Cloud• Users can link over web– Utility (web service)– SOA useful, not necessary– Pay/use

• Has very large capacity– Competitor to Google, IBM, Microsoft– At least one case where Amazon servers went

down, causing grief to users

EC2 Operations

• Users bid on unused computing capacity– Pay by the hour– Amazon has a fluctuating spot price set by supply

& demand• If bidder exceeds spot price, job run and charged• If bid lower than spot price, job terminated until spot

price drops to bid price• Enables customers to cap costs

EC2 Products

• As of April 2010– 16 Amazon cloud computing products– Services (processing, disk storage, database)

• SimpleDB

– Simple Storage Service (S3)• Users can store and retrieve large amounts of data at any

time• Can make stored data private, public, or targeted• Authentication devices used for security• Billed on monthly pay-by-use basis

– In 2010 opened Singapore data center

InnoCentive

Web-based businessLink those with problems (businesses, NGOs,

public sector organizations)With those willing to work (over 160,000)

Description• Eli Lilly and Company start-up• Spun off to be independent company• CHALLENGES

– Ideation challenges - broad questions seeking new ideas. – Theoretical challenges - detailed solution requirements.

• Usually, intellectual property rights transfer from the solution provider to the client organization, although some clients prefer a non-exclusive perpetual license.

– Reduction to Practice (RTP) challenges - high level of detail• Require solvers to submit validated solutions in the form of original data or physical

samples. Clients are allowed to test proposed solutions. Intellectual property rights are always transferred to the client in RTP challenges.

– eRFP challenges - requests for proposals to the world. • When solvers submit proposed solutions, clients evaluate responses and select

solvers for further details. Terms of subsequent contracts are negotiates for scope of work, duration, etc.

Challenges by Discipline

• Business and Entrepreneurship• Chemistry• Computer Science and IT• Engineering and Design• Food and Agriculture• Life Sciences• Mathematics and Statistics• Physical Sciences• Requests for Partners/Suppliers

Examples5 Oct 2010

Title Tags Posted Award Active Solvers

Deadline Categories

Room temperature oxidation catalyst

Science 10/05/2010 $50,000 69 1/05/2011 Theoretical

Improving therapeutic targeting of Bevacizumab

Science 10/04/2010 $50,000 118 12/04/2010 Theoretical

Better cat litter Science 10/04/2010 $7,500 157 11/05/2010 IdeationTheoreticalRTPeRFP

End of life indicator system

Science-Engineering

10/04/2010 $10,000 193 11/04/2010 IdeationTheoreticalRTPeRFP

Purported ValueAdvertises for Seekers (clients) in categories:

• Corporations might benefit in keeping pace with the dynamic competitive landscape by tapping into the creative talents of those who register as Solvers. InnoCentive argues that since no firm can afford to hire all the best people, an army of hundreds of thousands of individuals are available for ideas. Companies are encouraged to have their employees participate in an open culture of innovation.

• Not for Profit organizations might benefit because they have limited staff and budget for research, and InnoCentive offers a platform that only requires payment when useful solutions are found.

• Governments might benefit because they also face limited resources due to lower tax bases and higher deficits.

• Solvers benefit by taking advantage of opportunities to work on problems sufficiently worthwhile for InnoCentive clients to pay.

TRENDS

• Networked global economy– Open– Wal-Mart; IBM; Nike

• Digitization– Moore’s Law

• computing power doubles every 18 months

– Metcalfe’s Law• Value of a network increases with square of number of users

– Coase’s Law• As transaction costs decrease, firm complexity diminishes• Firms tend to expand until marginal cost of internal transactions = external marginal cost

Friedman (The World is Flat)

• THREE CONVERGENCES– New players (through global access)• BRIC

– New playing field (Web economy)• Global warming• Green emphasis• Cultural conflicts

– Ability to develop new ways

Megatrends• Energy supply

– Peak Oil– Global warming

• Complexity– Unintended consequences

• Globalization– Japan; Asian Tigers; BRIC

• Digitization– Enterprise systems

• Paradox: More Integrated Systems ˃˃ Fewer Systems People

• DEREGULATION/PRIVATIZATION– Home mortgage crisis

• COMMODITIZATION OF PROCESSES– Software as a service

Convergence Revolution1st Wave 2nd Wave 3rd Wave

Industry Agriculture Industrial Knowledge Service

Economic Regional National Global

Competitive Advantage

Economy of Scale

Economy of Scope

Economy of Expertise

Economy of Convergence

Primary Focus

EXPLOITATION EXPLORATION

Convergence Components/Products

Functions Organizations Technology Industries Bio-artificial systems

Economies

Evolution & Organizational Innovation

LEVEL CONVERGENCE LEVEL PURPOSE1 Component/Product Product/Service innovation2 Functional Process innovation3 Organizational Value chain innovation4 Technology Technology value innovation5 Industry New industries

Customer value innovation6 Bio-Artificial Molecular economy

New Products/Processes

• Component/Product convergence– iPad, Smartphone

• Functional convergence– BPR– Matrix organization– Temporary supply chain linkages

E-Globalization

• Develop Corporate Global Mindset• Build Global Competitive Advantage– AGGREGATION

• Seek economies of scale & scope• McDonalds, IBM, Microsoft

– ADAPTATION• Maximize local relevance

– Mini-IBMs by country, Google China

– ARBITRAGE• Exploit differences in national, regional markets

– US Headquarters-Chinese factories-Indian call centers-US retail– Organizational convergence

Technology Convergence

• BIOTECHNOLOGY– Drugs – convergence of pharmaceuticals, agriculture, chemical– Neural technology – convergence of biology, computer

technology– MRI/PET scanning – convergence of engineering, biology

• NEURAL SCIENCE• NANOTECHNOLOGY• Evidence:

– Cell phones – convergence of internet access, music, photography, telecommunications

– Microelectromechanical systems (MEMS)

Industry Convergence

• Two or more previously distinct industries become direct competitors or cooperators

• Apple vs. music distributes• Walt Disney – travel convention

– FUNCTIONAL• Products of 2 industries linked

– PCs & TV

– COMPLEMENTARY• Multiple objects of value from 1 location

– Travel & lodging (air tickets & hotel rooms, rental cars)– Commercial & Investment banking (post-Glass-Steagall)

Industry Convergence Examples

• News– Newspapers – disappearing• Network TV swamped by Cable Networks

– CNN, ESPN, directTV

– Telecom• AT&T breakup

– EDUTAINMENT• Education & entertainment

– INFOTAINMENT• Information & entertainment

Biological & Artificial System Convergence

• Application of microtechnology to business• MINIATURIZAATION = computer chips, health care, gene splicing• VISUALIZATION – X-ray, material chemistry• MANIPULATION – virtual reality & tactical feedback• EVALUATION – partition test medium (genome project)

– Biotechnology• Genetic modification – US/European debate

– Nanotechnology• The science of the very small

• SINGULARITY– Ray Kurzweil (2005)– Computers surpass human knowledge

Innovation Through Open Systems

• Web 2.0– Openness, participation, collaboration

• Service oriented architecture– Strategy to turn applications & information

sources from different organizations into SERVICES accessible via Web• IBM: On Demand Business• Hewlett-Packard: Adaptive Enterprise• Dell: Blade computing

SOA & New Products

• iPod• Banking – Salesforce– Automate forecasting processes in CRM– Provide open but secure system

• Bank of the West– Enabled access to software to better identify

seasonal traffic• Wal-Mart– RFID linkage

Open Source Development

Red Hat [2009]: Can save by:1. Enabling use of commodity hardware rather

than proprietary machines2. Avoids maintenance contracts3. Greater functionality, reliability, performance4. Faster learning curve, available support tools5. Avoid vendor lock-in6. Reduce need for security consultants & tools

CONFENIS 2010 Natal

Open Source ERP Products

• Compiere• OpenMFG• Open for Business Project• Tiny ERP• Web ERP• Open Office• OpenBravo• OpenProSourceforge.net listed over 1,000 ERP projects May 2009OPTION FOR SMEs

CONFENIS 2010 Natal

Strategic Innovation

• Applying new convergence developments to business opportunities

• TYPES OF INNOVATION– Improve products or services

• Speed, customization, aesthetics

– Improve value chain processes• Lower cost, improved quality

– Continuous improvement• Participation by customers, producers

– Strategic• New ways of doing old tasks• SOUTHWEST AIRLINES• IKEA

Convergence Evolution• Component/Product Convergence

– Clock radios, cell phones, Amazon’s Kindle• Functional Convergence

– BPR – Dell’s value chain, McDonalds’ food delivery• Organizational Convergence

– Supply chains – Wal-Mart, Nike, Dell• Technology Convergence

– Nanotechnology, biotechnology• Industry Convergence

– Combining industries – Apple iTunes, Health Tourism• Open-source Convergence

– Web 2.0, open source ERP• Biological & Artificial Systems Convergence

– Miniaturization – gene splicing, molecular biology, MRI

CONCLUSION

• The world is flatter• Widening economic & digital divides• Convergenomics brings new business models

& opportunities• Strategic innovation required to survive

Large-Scale Business Software

• Enterprise Resource Planning• Customer Relationship Management

Artificial Intelligence

• “Next Killer App”• “Tool for Managers, not for workers”• Robotics technology in military– P.W. Singer, 2009

• Advance pattern recognition

Job Automation

• Globalizing labor and capital– Not globalizing consumption

• Shift away from lowest labor to lowest energy costs