Pci lecture 2 software it failures (1)

M Bronte-StewartM Bronte-Stewart 11
Professional Computing IssuesProfessional Computing Issues
COMP09017COMP09017
Unit 2: Software & IT FailureUnit 2: Software & IT Failure

Some flawed predictionsSome flawed predictions
'I think there is a world market for maybe five computers'I think there is a world market for maybe five computers''
– Thomas Watson, Chairman of IBM, 1943Thomas Watson, Chairman of IBM, 1943
'I have traveled the length and breadth of this country'I have traveled the length and breadth of this country
and talked with the best people, and I can assure you thatand talked with the best people, and I can assure you that
data processing is a fad that won't last out the yeardata processing is a fad that won't last out the year''
– Editor in charge of business books for Prentice Hall, 1957Editor in charge of business books for Prentice Hall, 1957
'But what... is it good for?''But what... is it good for?'
– Engineer at the Advanced Computing Systems division of IBM,Engineer at the Advanced Computing Systems division of IBM,
commenting on the microchip, 1968commenting on the microchip, 1968
'There is no reason why anyone would want a computer'There is no reason why anyone would want a computer
in the home'in the home'
– Ken Olson, Present, Chairman and founder of Digital EquipmentKen Olson, Present, Chairman and founder of Digital Equipment
Corporation, 1977Corporation, 1977
'640K should be enough for anybody'640K should be enough for anybody''
– Bill Gates, on MS DOS version 1 in 1981Bill Gates, on MS DOS version 1 in 1981

Missing the MarkMissing the Mark
Lotus NotesLotus Notes
– Developed in 1984Developed in 1984
– Offered to Microsoft in 1988 for $12Offered to Microsoft in 1988 for $12
millionmillion
(they valued it at $4m and declined)(they valued it at $4m and declined)
– By 1993 revenue was $100 millionBy 1993 revenue was $100 million
– Lotus purchased by IBM in 1995 forLotus purchased by IBM in 1995 for
£3.5billion£3.5billion

Software MattersSoftware Matters
““There are only two commodities that will count.There are only two commodities that will count.
One is oil and the other is software. And thereOne is oil and the other is software. And there
are alternatives to oil.”are alternatives to oil.”
Bruce BondBruce Bond

Social Need for ITSocial Need for IT
Humans want to increase their comfort,Humans want to increase their comfort,
knowledge and control.knowledge and control.
Hence;Hence;
– Agrarian Revolution (c. 8000 BC).Agrarian Revolution (c. 8000 BC).
– Industrial Revolution (c. 1700 AD).Industrial Revolution (c. 1700 AD).
– Information Revolution (c. 1980 AD?).Information Revolution (c. 1980 AD?).
(Alvin Toffler – The Third Wave)(Alvin Toffler – The Third Wave)

Where aren’t computers used today?Where aren’t computers used today?
CarsCars –– GM claim their cars have 100 million lines of codeGM claim their cars have 100 million lines of code
Air traffic controlAir traffic control
MedicineMedicine
Power plantsPower plants
SpacecraftSpacecraft
Police forcesPolice forces
Income TaxIncome Tax
SupermarketsSupermarkets
Credit card Companies……etcCredit card Companies……etc

But Can We Trust Computers?But Can We Trust Computers?
Computer applications are often very complex.Computer applications are often very complex.
Impossible to test some programs fullyImpossible to test some programs fully
– Eg Nuclear power station software!Eg Nuclear power station software!
Impossible to produce a program with no errors.Impossible to produce a program with no errors.
– Some are minor,Some are minor,
e.g.e.g. hyphenating words in a word processorhyphenating words in a word processor
– Some are major,Some are major,
cost millions of dollars orcost millions of dollars or
result in deaths.result in deaths.

What Can Go Wrong?What Can Go Wrong?
Almost anything:Almost anything:
– Billing errorsBilling errors
– Database accuracy problemsDatabase accuracy problems
– Consumer Hardware and Software ErrorsConsumer Hardware and Software Errors
– System FailuresSystem Failures
Child Benefits AgencyChild Benefits Agency
Inland Revenue systemsInland Revenue systems
Passport office systemPassport office system
Developers seem to find it difficult to get itDevelopers seem to find it difficult to get it
right or foresee unwanted outcomesright or foresee unwanted outcomes

Effectiveness of U.S. federal softwareEffectiveness of U.S. federal software
Although these statistics are old they give anAlthough these statistics are old they give an
indication of how much of the software budget wasindication of how much of the software budget was
wasted on poor quality specification andwasted on poor quality specification and
development.development.
Guess the % of the budget that was spent usefully:Guess the % of the budget that was spent usefully:
Delivered but never usedDelivered but never used 46%46%
Paid for but not deliveredPaid for but not delivered 29%29%
Abandoned or reworkedAbandoned or reworked 19%19%
Used after changesUsed after changes 3%3%
Used as deliveredUsed as delivered 2%2%

1010
Project FailureProject Failure
IT projects tend to go over budget and / or over timeIT projects tend to go over budget and / or over time
schedule and / or do not meet expectationsschedule and / or do not meet expectations
 Research by the BCS, Royal Academy of Engineering, OASIG,Research by the BCS, Royal Academy of Engineering, OASIG,
Oxford University and others suggest that only about 10% toOxford University and others suggest that only about 10% to
30% are successful30% are successful
 Organisations and governments spend well over $1 trillion on ITOrganisations and governments spend well over $1 trillion on IT
hardware, software and services.hardware, software and services. (Charette, IEEE)(Charette, IEEE)
 Conservative estimates put the cost of IT project failure at tensConservative estimates put the cost of IT project failure at tens
of billions of Euros across the EUof billions of Euros across the EU Jaques, 2004Jaques, 2004 (142 billion Euros in(142 billion Euros in
20042004 McManus & Wood-Harper, 2008McManus & Wood-Harper, 2008) and around $500 billion wasted on IT) and around $500 billion wasted on IT
purchases that fail to reach their objectives worldwidepurchases that fail to reach their objectives worldwide Feld &Feld &
Stoddard, 2004Stoddard, 2004
M B-SM B-S

IT project costsIT project costs
The Defence Information Infrastructure project to incorporateThe Defence Information Infrastructure project to incorporate
150,000 terminals for 300,000 users at over 2000 defence150,000 terminals for 300,000 users at over 2000 defence
sites is 18 months late and running more than £180m oversites is 18 months late and running more than £180m over
budget.budget.
The proposed cost of the controversial national identityThe proposed cost of the controversial national identity
system rose to over £5bn.system rose to over £5bn.
The NHS IT scheme was initially estimated to cost aroundThe NHS IT scheme was initially estimated to cost around
£2.3bn, later this figure climbed to £12.7bn.£2.3bn, later this figure climbed to £12.7bn.
Not surprisingly, a significant proportion of public sector ITNot surprisingly, a significant proportion of public sector IT
expenditure at risk of being wasted according to a House ofexpenditure at risk of being wasted according to a House of
Commons select committee.Commons select committee.
1212M B-SM B-S

““Prison IT system guilty of 'basic'Prison IT system guilty of 'basic'
project management failures” (2009)project management failures” (2009)
The £234m C-Nomis IT system for Prisons failedThe £234m C-Nomis IT system for Prisons failed
in almost every possible way .in almost every possible way .
The NAO concluded that the technical complexity had beenThe NAO concluded that the technical complexity had been
“significantly underestimated”.“significantly underestimated”. C-Nomis was treated as an IT projectC-Nomis was treated as an IT project
and not a business-change programme, project management wasand not a business-change programme, project management was
poor, and contracts with suppliers were weak.poor, and contracts with suppliers were weak.
Edward Leigh, chairman of the Public Accounts Committee, said of C-Edward Leigh, chairman of the Public Accounts Committee, said of C-
Nomis thatNomis that “kindergarten mistakes”“kindergarten mistakes” had been repeated:had been repeated:
““This Committee hears of troubled government projects all tooThis Committee hears of troubled government projects all too
frequently. But the litany of failings in this case are in a class of theirfrequently. But the litany of failings in this case are in a class of their
own. All of this mess could have been avoided if good practice inown. All of this mess could have been avoided if good practice in
project management had been followed.”project management had been followed.”
M B-SM B-S 1313

Notable Software FailuresNotable Software Failures
Radiotherapy Equipment (Therac-25)Radiotherapy Equipment (Therac-25)
– designed to operate in two different modesdesigned to operate in two different modes
low dosage and high dosage (only to smaller area)low dosage and high dosage (only to smaller area)
– high dosage given by mistake and patient diedhigh dosage given by mistake and patient died
– inquest showed that software system had replaced earlierinquest showed that software system had replaced earlier
“hard-wired” system“hard-wired” system
hard-wired system had built-in lock which prevented high-dosagehard-wired system had built-in lock which prevented high-dosage
being given without adequate screeningbeing given without adequate screening

The misguided torpedo assumptionThe misguided torpedo assumption
–– attempt to prevent the possibility of a torpedoattempt to prevent the possibility of a torpedo
returning and exploding against the ship it was firedreturning and exploding against the ship it was fired
fromfrom
safety system built so that it would self-destruct if itsafety system built so that it would self-destruct if it
turned 180 degreesturned 180 degrees
Unfortunately, during trials, a torpedo jammed in itsUnfortunately, during trials, a torpedo jammed in its
tube on board the ship, the test was abandoned andtube on board the ship, the test was abandoned and
the ship turned round to go homethe ship turned round to go home
…….. BANG!BANG!

The auto-land systemThe auto-land system
–– to help airplanes land safely two separate softwareto help airplanes land safely two separate software
programs were created.programs were created.
The first tracks the aircraft’s descent as it flies down aThe first tracks the aircraft’s descent as it flies down a
beam projected from the runway.beam projected from the runway.
– If the plane loses the beam, the engine power is increasedIf the plane loses the beam, the engine power is increased
and it flies round for another try.and it flies round for another try.
The second system detects when the plane is nearThe second system detects when the plane is near
the ground cuts the engine and raises the nose priorthe ground cuts the engine and raises the nose prior
to landing.to landing.
Beam
Runway

Auto-land systemAuto-land system
At the first live test all seemed to work well at theAt the first live test all seemed to work well at the
startstart
the plane flew down the beam, the detector sensedthe plane flew down the beam, the detector sensed
the ground, the nose was raised and the engine cutthe ground, the nose was raised and the engine cut
back,back,
but then the plane sank below the beam and the firstbut then the plane sank below the beam and the first
system cut in, lowered the nose and boosted thesystem cut in, lowered the nose and boosted the
engines …..engines …..
……. CRASH!. CRASH!

Auto-Land systemAuto-Land system
Those problems were correctedThose problems were corrected
But, new emergentBut, new emergent
problem…problem…
Unerring accuracy - now allUnerring accuracy - now all
aircraft landed on exactly theaircraft landed on exactly the
same spotsame spot
Runway started to break up!Runway started to break up!

Avon County Council installed a new computerAvon County Council installed a new computer
program to pay staff wages.program to pay staff wages.
The spree started in a small way paying a caretaker £75 an hour.The spree started in a small way paying a caretaker £75 an hour.
Then it didn’t pay canteen workers at all for 7 weeks.Then it didn’t pay canteen workers at all for 7 weeks.
Next it paid a janitor £2,600 for a week’s work.Next it paid a janitor £2,600 for a week’s work.
A deputy headmistress received her year’s annual salary once aA deputy headmistress received her year’s annual salary once a
month.month.
Heads of department earned less than their assistants.Heads of department earned less than their assistants.
Some people had more tax deducted in a week than they earned allSome people had more tax deducted in a week than they earned all
year.year.
By February 280 council employees were out on strike.By February 280 council employees were out on strike.
1919M B-SM B-S
(Stephen Pile)

Denver International AirportDenver International Airport
– 16 months late16 months late
– $100 million over budget$100 million over budget
– Problems caused by new automatedProblems caused by new automated
baggage systembaggage system
Project poorly managedProject poorly managed
Significant changes to original spec.Significant changes to original spec.
Designers worked on their ownDesigners worked on their own
Components worked in isolation but notComponents worked in isolation but not
when integratedwhen integrated
No cohesion among project groupsNo cohesion among project groups
Testing poorly plannedTesting poorly planned

Denver International AirportDenver International Airport
Baggage system supposed to manage luggage from check-in to the correctBaggage system supposed to manage luggage from check-in to the correct
plane in under 10 minutesplane in under 10 minutes
4,000 bar coded carts scanned and tracked by lasers4,000 bar coded carts scanned and tracked by lasers
Computers used database of flights and gate numbers to get routingComputers used database of flights and gate numbers to get routing
information, control motors and switches and direct carts, but softwareinformation, control motors and switches and direct carts, but software
errors not fixederrors not fixed
Interference from politicians forced unrealistic delivery deadlines onInterference from politicians forced unrealistic delivery deadlines on
developersdevelopers
Scanners got dirty, became misaligned, did not detect all cartsScanners got dirty, became misaligned, did not detect all carts
Automated carts crashed into each other at track intersections, luggage sentAutomated carts crashed into each other at track intersections, luggage sent
to wrong places or fell off and blocked tracks – chaos! ( and bad publicity).to wrong places or fell off and blocked tracks – chaos! ( and bad publicity).

Ariane 5Ariane 5
European space rocket designed to sendEuropean space rocket designed to send
communication satellites into orbitcommunication satellites into orbit
Broke up less than 1 minute into its maiden flightBroke up less than 1 minute into its maiden flight
in 1996in 1996
Inquiry found that the on board computer hadInquiry found that the on board computer had
crashed due to software error, converting a 64-bitcrashed due to software error, converting a 64-bit
floating point value into a 16-bit stored integer,floating point value into a 16-bit stored integer,
exceeded maximum value but no exceptionexceeded maximum value but no exception
handling mechanism.handling mechanism.

2323
London Ambulance SystemLondon Ambulance System
New Software “Put Lives At Risk!”New Software “Put Lives At Risk!”
Main Findings of investigationMain Findings of investigation
– Inexperienced procurement teamInexperienced procurement team
– Staff mistrust and oppositionStaff mistrust and opposition
– Over-ambitious timetableOver-ambitious timetable
– Price put before qualityPrice put before quality
– Incomplete and untested softwareIncomplete and untested software
– Andersen Consulting report suppressedAndersen Consulting report suppressed
– Management failed to identify and solve problemsManagement failed to identify and solve problems
– Users “Users “did things wrongdid things wrong””
““...a faulty system implemented with undue haste, by a management...a faulty system implemented with undue haste, by a management
determined to impose its will...”determined to impose its will...”
M B-SM B-S

2424
Some Newspaper HeadlinesSome Newspaper Headlines
2010 New HMRC inland revenue tax system producing2010 New HMRC inland revenue tax system producing
wrong codeswrong codes “ we’ve heard it all before”.“ we’ve heard it all before”.
2010 Labour accused of wasting 26bn on failed IT projects2010 Labour accused of wasting 26bn on failed IT projects
“stupendous incompetence”.“stupendous incompetence”.
2009 Rural payments scheme put out to grass2009 Rural payments scheme put out to grass “a display of“a display of
scant regard for protecting public money”.scant regard for protecting public money”.
2009 C-nomis offender management system2009 C-nomis offender management system “a master class in“a master class in
sloppy project management”sloppy project management”
2008 Edinburgh Fringe Box-office system2008 Edinburgh Fringe Box-office system “weak”, “fundamentally“weak”, “fundamentally
flawed” “insufficient planning, lack of risk management, inadequateflawed” “insufficient planning, lack of risk management, inadequate
communications and no authorised business case”.communications and no authorised business case”.
2005 Strathclyde Police Computer System2005 Strathclyde Police Computer System “a complete disaster”“a complete disaster”
2003 Libra the IT system for magistrates courts2003 Libra the IT system for magistrates courts “One of the“One of the
worst projects I have ever seen”.worst projects I have ever seen”.
M B-SM B-S

IT Project Failure FiguresIT Project Failure Figures
Of 3,682 projects in 365 firms – 31% were cancelled, 53%Of 3,682 projects in 365 firms – 31% were cancelled, 53%
had cost overruns and poor functionality and only 12%had cost overruns and poor functionality and only 12%
were on-time and budget (Johnson)were on-time and budget (Johnson)
Only 7% of the 1000 firms Critical Research surveyedOnly 7% of the 1000 firms Critical Research surveyed
thought they were using IT effectively and 75% of thesethought they were using IT effectively and 75% of these
firms think that their IT systems are not providing a returnfirms think that their IT systems are not providing a return
on investmenton investment
31% of new IT projects are cancelled and over 50% of IT31% of new IT projects are cancelled and over 50% of IT
projects are over budget (PC Week)projects are over budget (PC Week)
But we don’t learn

““Hey Paddy, we’ve got dis here car ta pull out.”Hey Paddy, we’ve got dis here car ta pull out.”

““Shamus, we’re pullin’ it up now.”Shamus, we’re pullin’ it up now.”

““Here Paddy, what happens if da crane can’t hold it?”Here Paddy, what happens if da crane can’t hold it?”
““Shamus me lad, don’t you even tink about dat happening.”Shamus me lad, don’t you even tink about dat happening.”

““Hoy Paddy, oi didn’t even tink it, oh shoit!”Hoy Paddy, oi didn’t even tink it, oh shoit!”

““Shamus, moi boy, go get a bigga crane.”Shamus, moi boy, go get a bigga crane.”
““Alroight, Paddy.”Alroight, Paddy.”

““Ah Shamus, dis bigga crane is doin’ da job well.”Ah Shamus, dis bigga crane is doin’ da job well.”

““Dat it is Paddy, dat it is.”Dat it is Paddy, dat it is.”

““Now for da first crane Shamus, up it cooms.”Now for da first crane Shamus, up it cooms.”

““ What's happening Paddy, what's happening.”What's happening Paddy, what's happening.”

““Oh f*#k Shamus, we’ve got ta get an even bigga crane!’Oh f*#k Shamus, we’ve got ta get an even bigga crane!’

Lyytinen and RobeyLyytinen and Robey (2000)(2000)
““Organisations fail to learn from their experienceOrganisations fail to learn from their experience
in systems development because of limits ofin systems development because of limits of
organisational intelligence, disincentives fororganisational intelligence, disincentives for
learning, organisational designs andlearning, organisational designs and
educational barriers.educational barriers. Not only have manyNot only have many
organisations failed to learn they also haveorganisations failed to learn they also have
learned to fail.learned to fail. Over time they accept andOver time they accept and
expect poor performance while creatingexpect poor performance while creating
organisational myths that perpetuate short-termorganisational myths that perpetuate short-term
optimisation.”optimisation.”
4141M B-SM B-S

Wider effects of IT project failureWider effects of IT project failure
Perceptions of poor success rates and wastedPerceptions of poor success rates and wasted
resources affect decision makingresources affect decision making
The more IT projects are seen to go wrong the more:The more IT projects are seen to go wrong the more:
– the public become cynicalthe public become cynical
– staff learn to expect problems and delaysstaff learn to expect problems and delays
– Developers wonder if a lot of their work is likely to beDevelopers wonder if a lot of their work is likely to be
wasted effortwasted effort
– business people become nervous of technology changebusiness people become nervous of technology change
– those holding the purse strings may view IT as a worry andthose holding the purse strings may view IT as a worry and
a poor return on investmenta poor return on investment
M B-SM B-S 4242

ResearchResearch
So many computer projects fail and it is acceptedSo many computer projects fail and it is accepted
that billions are wasted every year, yet there isthat billions are wasted every year, yet there is
remarkable agreement on the main causes of ITremarkable agreement on the main causes of IT
project failureproject failure
If you would like to read more about this the findingsIf you would like to read more about this the findings
of a paper presented at the European Managementof a paper presented at the European Management
of Technology Conference 2009, has been includedof Technology Conference 2009, has been included
in the “Course Materials” folder of the Blackboardin the “Course Materials” folder of the Blackboard
PCI site (listed asPCI site (listed as M B-S Project Failure and Risks)M B-S Project Failure and Risks)..

4444
What are the main causes of failure?What are the main causes of failure?
6 major research studies reach remarkably6 major research studies reach remarkably
similar conclusions about the significantsimilar conclusions about the significant
causes that threaten project success:causes that threaten project success:
– OASIGOASIG
– Standish (CHAOS)Standish (CHAOS)
– Select CommitteeSelect Committee
– OGC / NAOOGC / NAO
– Schmidt, Lyytinen, Keil & Cule,Schmidt, Lyytinen, Keil & Cule,
– Fortune and WhiteFortune and White
M B-SM B-S

IT Project FailureIT Project Failure (OASIG study, 1996)(OASIG study, 1996)
Outcomes from IT investment :Outcomes from IT investment :
– 80% to 90% do not meet goals80% to 90% do not meet goals
– 80% delivered late and over budget80% delivered late and over budget
– 40% fail or are abandoned40% fail or are abandoned
– Under 40% address training and skills enoughUnder 40% address training and skills enough

IT Project FailureIT Project Failure (OASIG)(OASIG)
Main reasons why IT projects failMain reasons why IT projects fail
– Management agenda is too limitedManagement agenda is too limited
most IT investments are technology ledmost IT investments are technology led
main investment motive is to only cut costsmain investment motive is to only cut costs
– This narrow focus on technical capabilities and efficiency goals meansThis narrow focus on technical capabilities and efficiency goals means
that inadequate attention is given to the human and organisationalthat inadequate attention is given to the human and organisational
issues that can determine a project’s ultimate success.issues that can determine a project’s ultimate success.
– Users don’t influence development enoughUsers don’t influence development enough
– Senior managers don’t understand the links between technical andSenior managers don’t understand the links between technical and
organisational changeorganisational change
– Project management techniques and IT approaches are too technicalProject management techniques and IT approaches are too technical
– Companies fail to organise work or design jobs/roles properlyCompanies fail to organise work or design jobs/roles properly

Standish Group CHAOS SurveyStandish Group CHAOS Survey
Project Success FactorsProject Success Factors
1.1. User involvementUser involvement
2.2. Executive management supportExecutive management support
3.3. Clear and firm statement of requirementsClear and firm statement of requirements
4.4. Proper planning and formal methodologyProper planning and formal methodology
5.5. Realistic expectationsRealistic expectations
6.6. Minimised scope and smaller project milestonesMinimised scope and smaller project milestones
7.7. Competent, skilled staffCompetent, skilled staff
8.8. OwnershipOwnership
9.9. Clear vision and business objectivesClear vision and business objectives
10.10. Hard-working, focussed staffHard-working, focussed staff
11.11. Experienced project managersExperienced project managers
12.12. Reliable estimatesReliable estimates
4747M B-SM B-S

Select Committee on Public AccountsSelect Committee on Public Accounts
Report on Improving the delivery of Government ITReport on Improving the delivery of Government IT
projects (Study of 25 projects 1990 to 2000)projects (Study of 25 projects 1990 to 2000)
 ““Implementing IT systems has proved difficult”Implementing IT systems has proved difficult”
 ““Frequent cases of delay,Frequent cases of delay,
confusion and inconvenienceconfusion and inconvenience
for the citizen and poor valuefor the citizen and poor value
for money for the tax payer.”for money for the tax payer.”
4848M B-SM B-S

Select Committee Report conclusionsSelect Committee Report conclusions
1.1. Decisions about IT areDecisions about IT are Business not technicalBusiness not technical, they have profound effects on, they have profound effects on
customer service andcustomer service and must involve senior managementmust involve senior management..
2.2. End usersEnd users and their business needs must be identified before the projectand their business needs must be identified before the project
commences so thatcommences so that clear objectivesclear objectives are taken into account fully during design andare taken into account fully during design and
developmentdevelopment
3.3. Scale and complexityScale and complexity – how ambitious? Can it be undertaken in one go?– how ambitious? Can it be undertaken in one go?
4.4. SkilledSkilled Project ManagersProject Managers are essentialare essential
5.5. SoundSound methodologiesmethodologies and well conceivedand well conceived risk managementrisk management are called forare called for
6.6. Need for a high degree of professionalism in the definition, negotiation andNeed for a high degree of professionalism in the definition, negotiation and
management ofmanagement of IT contractsIT contracts
7.7. TrainingTraining can take up considerable resources but must address the needs of thecan take up considerable resources but must address the needs of the
users and of those maintaining the systems if the anticipated benefits are to beusers and of those maintaining the systems if the anticipated benefits are to be
realisedrealised
8.8. ContingencyContingency plansplans should be in placeshould be in place
9.9. Organisations shouldOrganisations should learn lessonslearn lessons from projects undertaken. A postfrom projects undertaken. A post
implementation review should establish the extent to which they secured theimplementation review should establish the extent to which they secured the
proposed business benefits, user expectations and technical requirements.proposed business benefits, user expectations and technical requirements.
4949M B-SM B-S

OGC and NAO Best PracticeOGC and NAO Best Practice 20052005
Common causes of project failureCommon causes of project failure
1. Lack of clear links between the project and the organisation's key strategic1. Lack of clear links between the project and the organisation's key strategic
priorities, including agreed measures of success.priorities, including agreed measures of success.
2. Lack of clear senior management and Ministerial ownership and leadership.2. Lack of clear senior management and Ministerial ownership and leadership.
3. Lack of effective engagement with stakeholders.3. Lack of effective engagement with stakeholders.
4. Lack of skills and proven approach to project management and risk4. Lack of skills and proven approach to project management and risk
management.management.
5. Too little attention to breaking development and implementation into5. Too little attention to breaking development and implementation into
manageable steps.manageable steps.
6. Evaluation of proposals driven by initial price rather than long-term value for6. Evaluation of proposals driven by initial price rather than long-term value for
money (especially securing delivery of business benefits).money (especially securing delivery of business benefits).
7. Lack of understanding of, and contact with the supply industry at senior levels7. Lack of understanding of, and contact with the supply industry at senior levels
in the organisation.in the organisation.
8. Lack of effective project team integration between clients, the supplier team8. Lack of effective project team integration between clients, the supplier team
and the supply chain.and the supply chain.
5050M B-SM B-S

Keil, Cule, Lyytinen & SchmidtKeil, Cule, Lyytinen & Schmidt (1998)(1998)
These researchers recruited 3 panels of experienced projectThese researchers recruited 3 panels of experienced project
managers in different places – Finland, Hong Kong & U.S.A. –managers in different places – Finland, Hong Kong & U.S.A. –
and asked them to identify and rank specific risk factors.and asked them to identify and rank specific risk factors.
– Lack of top management commitment to the projectLack of top management commitment to the project
– Failure to gain user commitmentFailure to gain user commitment
– Misunderstanding the requirementsMisunderstanding the requirements
– Lack of adequate user involvementLack of adequate user involvement
– Failure to manage end user expectationsFailure to manage end user expectations
– Changing scope / objectionsChanging scope / objections
– Lack of required knowledge / skills in project personnelLack of required knowledge / skills in project personnel
– Lack of frozen requirementsLack of frozen requirements
– Introduction of new technologyIntroduction of new technology
– Conflict between user departmentsConflict between user departments
5151M B-SM B-S

Fortune and White (2006)Fortune and White (2006)
Fortune and White reviewed 63 publications that focusFortune and White reviewed 63 publications that focus
on project Critical Success Factorson project Critical Success Factors
The top ten (in order of count of citations) of the 27 they quote are:The top ten (in order of count of citations) of the 27 they quote are:
– Support from senior managementSupport from senior management
– Clear realistic objectivesClear realistic objectives
– Strong / detailed plan kept up to dateStrong / detailed plan kept up to date
– Good communications / feed backGood communications / feed back
– User / client involvementUser / client involvement
– Skilled / suitably qualified / sufficient staff / teamSkilled / suitably qualified / sufficient staff / team
– Effective change managementEffective change management
– Competent project managerCompetent project manager
– Strong business case / sound basis for projectStrong business case / sound basis for project
– Sufficient / well allocated resourcesSufficient / well allocated resources
5252M B-SM B-S

5353
Select Committee
OASIG NAO / OGC
Keil et al CHAOS (Top
Requirements:)
Fortune and White
(CSFs)
IT projects are driven by
business (not technical)
decisions
Many IT investments are
seen only as technology led
and aimed at cost cutting
Evaluation of proposals driven by
initial price rather than long term
value, especially securing delivery of
business benefits
Misunderstanding user
requirements
Clear business objectives
and Realistic
expectations
Strong business case /
sound basis for project
Insufficient involvement from
users
Users do not influence
development enough
Lack of effective engagement with
stakeholders
Lack of user involvement
or commitment User involvement User / client involvement
Clear objectives should be set
from the start
Need to set and review
strategic objectives for
change
Lack of clear link between the
project and the organisation’s key
strategic priorities, including agreed
measures of success
Unclear and changing
scope and objectives Clear and firm statement
of requirements
Clear realistic objectives
Lack of commitment from
senior management
Management agenda is
often too limited or narrow
Lack of clear senior management
ownership and leadership
Lack of top management
commitment
Executive management
support
Support from senior
management
Large projects may be
overambitious
Inadequate attention is
given to human and
organisational issues
Too little attention to breaking
development and implementation
into manageable steps
Number of organisational
units involved
Minimised scope and
smaller project
milestones
Project size, complexity,
number of people
involved, and duration
Skilled project managers are
essential to keep to time and
budget and appropriate
deliverables
Senior managers do not
understand the link between
technical and organisational
change
Lack of skills and proven approach
to project management and risk
management
Lack of required
knowledge and effective
project management
skills
Experienced project
managers
Competent project
manager and effective
change management
Success depends on good risk
analysis and sound
methodologies
Some project management
techniques and IT
approaches are too
technical
Lack of effective project
management
methodology
Proper planning and
formal methodology
Correct choice / past
experience of project
management
methodology / tools
Contingency plans should be
in place
Must work to detailed
implementation plans
Not managing change
properly
Reliable estimates
Strong / detailed plans
kept up to date
User and operator training
must be planned and designed
Failure to organise changes
in work and roles properly
Inadequate resources and skills to
deliver the total delivery portfolio
Inappropriate staffing and
ill defined responsibilities Competent, skilled and
focussed staff
Skilled / suitably qualified
and sufficient staff / team
There should be a post-
implementation review
Introduction of new
technology
Standard software
infrastructure
Planned close down /
review, acceptance of
possible failure
Need professionalism in the
definition, negotiation and
management of IT contracts
Lack of understanding of , and
contact with, the supply industry at
senior levels in the organisation
Ownership
Good communication /
feedback
M B-SM B-S

What can be done?What can be done?
Look before you leap benefit/risk assessmentLook before you leap benefit/risk assessment
Well specified project definition with rationaleWell specified project definition with rationale
More careful and accurate estimationMore careful and accurate estimation
Better Risk ManagementBetter Risk Management
Adherence to tried and tested methodologiesAdherence to tried and tested methodologies
Redefine “failure”Redefine “failure”

UK Government ApproachesUK Government Approaches
A set of Best Management Practice products offer flexible,A set of Best Management Practice products offer flexible,
practical and effective guidance, drawn from a range of thepractical and effective guidance, drawn from a range of the
most successful global business experiences.most successful global business experiences.
They have helped improve processes and operations forThey have helped improve processes and operations for
small businesses, public sector organizations and majorsmall businesses, public sector organizations and major
global enterprises.global enterprises.
They include: PRINCE2 (They include: PRINCE2 (Project managementProject management), ITIL (), ITIL (IT serviceIT service
managementmanagement), M_o_R (), M_o_R (Risk managementRisk management), MSP (), MSP (ManagingManaging
successful programmessuccessful programmes), P3O (), P3O (Project program and portfolio officesProject program and portfolio offices),),
MoP (MoP (Portfolio managementPortfolio management), MoV (), MoV (Management of ValueManagement of Value), P3M3), P3M3
((Portfolio, Programme, and Project Management Maturity ModelPortfolio, Programme, and Project Management Maturity Model))

PRINCE2PRINCE2
PRINCE2PRINCE2 ((PrProjectsojects inin aa CControlledontrolled EEnvironment) is a structured projectnvironment) is a structured project
management method based on experience drawn from thousands ofmanagement method based on experience drawn from thousands of
projects - and from the contributions of countless project sponsors, Projectprojects - and from the contributions of countless project sponsors, Project
Managers, project teams, academics, trainers and consultants.Managers, project teams, academics, trainers and consultants.
It is a widely recognised standard, which embodies established and
proven best practice in project management.
PRINCE2 is a non-proprietary method that can be applied to any projectPRINCE2 is a non-proprietary method that can be applied to any project
regardless of project scale, type, organisation, geography or culture.regardless of project scale, type, organisation, geography or culture.

ITILITIL (Information Technology(Information Technology
Infrastructure Library)Infrastructure Library)
ITIL = practices for IT service management (ITSM) that focuses onITIL = practices for IT service management (ITSM) that focuses on
aligning IT services with the needs of business.aligning IT services with the needs of business.
ITIL has five core publications, each of which covers an ITSM lifecycleITIL has five core publications, each of which covers an ITSM lifecycle
stage.stage.
– Service StrategyService Strategy,,
– Service DesignService Design,,
– Service TransitionService Transition,,
– Service OperationService Operation, and, and
– Continual Service ImprovementContinual Service Improvement..
ITILv3 (2011) underpins ISO/IEC 20000 (previously BS15000), theITILv3 (2011) underpins ISO/IEC 20000 (previously BS15000), the
International Service Management Standard for IT service management.International Service Management Standard for IT service management.
ITIL describes procedures, tasks and checklists that can be used by anITIL describes procedures, tasks and checklists that can be used by an
organization for establishing a minimum level of competency. It gives aorganization for establishing a minimum level of competency. It gives a
baseline to plan, implement, and measure. It is used to demonstratebaseline to plan, implement, and measure. It is used to demonstrate
compliance and to measure improvement.compliance and to measure improvement.

So What do we look out for ?So What do we look out for ?
Main obstacles include:Main obstacles include:
– no business caseno business case
– unrealistic expectationsunrealistic expectations
– requirements creeprequirements creep
– problems of dealing with changeproblems of dealing with change
– organisational inertiaorganisational inertia
– difficulty in anticipating eventsdifficulty in anticipating events
– inadequate leadership & managementinadequate leadership & management
– insufficient testinginsufficient testing
– boss / financial pressureboss / financial pressure

So What do we look out for ?So What do we look out for ?
Hardware/Software vendors sell “solutions” (Hardware/Software vendors sell “solutions” (Techno-hype)Techno-hype)
– Incompatibilities and difficulty building, modifying andIncompatibilities and difficulty building, modifying and
integrating information systemsintegrating information systems
Overconfidence and carelessnessOverconfidence and carelessness
– Including failure to plan and inadequate understanding of risksIncluding failure to plan and inadequate understanding of risks

The big pictureThe big picture
Real business problems involve more than justReal business problems involve more than just
changing computer hardware and softwarechanging computer hardware and software
– Changes to the organisation and the way it worksChanges to the organisation and the way it works
– Changes to people, new roles, retraining, motivationChanges to people, new roles, retraining, motivation
– Changes to information – manipulation, analysis,Changes to information – manipulation, analysis,
storage and flowstorage and flow
– Changes to structure and cultureChanges to structure and culture
– Changes to processesChanges to processes

Difficulties Building IT-Based SystemsDifficulties Building IT-Based Systems
Complexity (real-time, multitasking, global)Complexity (real-time, multitasking, global)
– (The bigger they are the harder they fall)(The bigger they are the harder they fall)
– Too many stakeholdersToo many stakeholders
– Small, self-contained, well understood systems have bestSmall, self-contained, well understood systems have best
chance of successchance of success
Rushed and speedy implementationRushed and speedy implementation
Not planned or managed properly,Not planned or managed properly,
Potentially influenced by commercial or political factorsPotentially influenced by commercial or political factors
Lack of incentives or laws to do a better jobLack of incentives or laws to do a better job
Integrating new system with legacy technologiesIntegrating new system with legacy technologies

Problems of Dealing with ChangeProblems of Dealing with Change
Lack of proper trainingLack of proper training
Unforeseen effectsUnforeseen effects
Incompatibilities and legacy systemsIncompatibilities and legacy systems
Technical progress can seem to raise moreTechnical progress can seem to raise more
problems than it solvesproblems than it solves
New efficiencies mean fewer staffNew efficiencies mean fewer staff
employed?employed?
Organisational inertia – resistance toOrganisational inertia – resistance to
change,change,

Difficulty in Anticipating EventsDifficulty in Anticipating Events
Ignorance as to how any particular innovationIgnorance as to how any particular innovation
will develop or be adapted over time.will develop or be adapted over time.
e.g. World Wide Web (only “invented” in 1992) earlye.g. World Wide Web (only “invented” in 1992) early
adopters did not realise how far reaching andadopters did not realise how far reaching and
pervasive it would become.pervasive it would become.

Computer RelianceComputer Reliance
Could modern life survive without IT?Could modern life survive without IT?
Is “dependence” on computers differentIs “dependence” on computers different
from dependence on electricity?from dependence on electricity?
Computers are now virtually everywhereComputers are now virtually everywhere
– is thisis this goodgood oror badbad oror neutralneutral??

Control & DesignControl & Design
Does the computer control us, or doDoes the computer control us, or do
we control the computer?we control the computer?
Four possible control scenariosFour possible control scenarios
– computercomputer controlscontrols computercomputer
– computercomputer controlscontrols humanhuman
– humanhuman controlscontrols computercomputer
– humanhuman controlscontrols humanhuman

1) Computer Controls Computer1) Computer Controls Computer
Computer Controls ComputerComputer Controls Computer
Key Points:Key Points:
– Often called aOften called a Finite State Automaton,Finite State Automaton,
– Malfunctions, but should be noMalfunctions, but should be no
breakdown of controlbreakdown of control
– control is clearly defined and does notcontrol is clearly defined and does not
alteralter
– e.g.e.g. automatic pilotautomatic pilot

2) Computer Controls Human2) Computer Controls Human
Computer Controls HumanComputer Controls Human
– Breakdown of control mayBreakdown of control may
lead to anger, fear,lead to anger, fear,
demoralisation,demoralisation,
– e.g. traffic lights, automatede.g. traffic lights, automated
cinema seat bookingcinema seat booking

3) Human Controls Computer3) Human Controls Computer
Human Controls ComputerHuman Controls Computer
– Breakdown of control may leadBreakdown of control may lead
to feelings of frustration,to feelings of frustration,
helplessness, inadequacyhelplessness, inadequacy
– e.g. almost anything “Office”e.g. almost anything “Office”
related, fax machinerelated, fax machine

4) Human Controls Human4) Human Controls Human
Human Controls HumanHuman Controls Human
– breakdowns all the timebreakdowns all the time
– e.g. “I don’t know what they’re talking about”e.g. “I don’t know what they’re talking about”
– ““the boss does not have a clue what is needed”the boss does not have a clue what is needed”
– ““men are from Mars, women are from Venus”men are from Mars, women are from Venus”
So why does control breakdown when humansSo why does control breakdown when humans
are involved?are involved?
It has a lot to do with INTERPRETATIONIt has a lot to do with INTERPRETATION

Impact of InterpretationImpact of Interpretation
1 Computer1 Computer controlscontrols ComputerComputer
– no interpretation requiredno interpretation required
2 Computer2 Computer controlscontrols HumanHuman
– human interpretation prohibitedhuman interpretation prohibited
3 Human3 Human controlscontrols ComputerComputer
– some human interpretation requiredsome human interpretation required
4 Human4 Human controlscontrols HumanHuman
– All is interpretationAll is interpretation

InterpretationInterpretation
CASE tools and IT Methodologies forcing moreCASE tools and IT Methodologies forcing more
mechanistic and structured behaviour have beenmechanistic and structured behaviour have been
created to try to reduce human effect on systemscreated to try to reduce human effect on systems
Likewise introducing computer-to-computer interactionLikewise introducing computer-to-computer interaction
into a human-to-human situation will lose allinto a human-to-human situation will lose all
interpretation, e.g. Stock market crashinterpretation, e.g. Stock market crash

InterpretationInterpretation
Wherever “interpretation” needs toWherever “interpretation” needs to
be reduced (or even removed) thebe reduced (or even removed) the
introduction of computers can leadintroduction of computers can lead
to better controlto better control
Wherever “interpretation” isWherever “interpretation” is
required it is necessary to have arequired it is necessary to have a
human in controlhuman in control

How does this affect our approachHow does this affect our approach
to Software Developmentto Software Development
A balancing actA balancing act
– customer wants -v- error freecustomer wants -v- error free
– Results in trying to make peopleResults in trying to make people
like machines (machine centredlike machines (machine centred
design) or machines like peopledesign) or machines like people
(human-centred design)(human-centred design)

Software DevelopmentSoftware Development

Designing People and ComputersDesigning People and Computers
Division of labour between Computers and PeopleDivision of labour between Computers and People
People are good at:People are good at:
– Understanding, social negotiation, imagination, seeing theUnderstanding, social negotiation, imagination, seeing the
whole situationwhole situation
Computers are good at:Computers are good at:
– Repetitive tasks involving speed, consistency andRepetitive tasks involving speed, consistency and
enduranceendurance
– Execution of unambiguous instructionsExecution of unambiguous instructions

Human-Centred DesignHuman-Centred Design vv Computer-Centred DesignComputer-Centred Design
Human-CentredHuman-Centred
– Technology aids theTechnology aids the
individualindividual
– To make life and workTo make life and work
easier and moreeasier and more
satisfyingsatisfying
– User friendlyUser friendly
Machine-CentredMachine-Centred
– SimplificationSimplification
– AutomationAutomation
– People adjust to thePeople adjust to the
machinemachine
– User hostile?User hostile?

The ExtremesThe Extremes
Anthropomorphisation - ascribing humanAnthropomorphisation - ascribing human
attributes to an animal or objectattributes to an animal or object
– e.g.“The computer made a mistake”e.g.“The computer made a mistake”
– ““come on you stupid machine!”come on you stupid machine!”
Reverse Anthropomorphisation - usingReverse Anthropomorphisation - using
computer functions and attributes tocomputer functions and attributes to
describe peopledescribe people
– e.g. “I am interrupt driven!”e.g. “I am interrupt driven!”

ConclusionsConclusions
Computer technology can beComputer technology can be
a powerful and valuable toola powerful and valuable tool
but it is not “Magic”but it is not “Magic”
Technology “failure”Technology “failure”
Don’t blame the stove for aDon’t blame the stove for a
badly cooked meal!badly cooked meal!

Computer Technology is a ToolComputer Technology is a Tool
Positive benefitsPositive benefits
– Reduces repetitiveReduces repetitive
mental or physical labourmental or physical labour
– Enable increasedEnable increased
productivity or safetyproductivity or safety
– Helps us to produceHelps us to produce
better products and learnbetter products and learn
new skillsnew skills
Negative costsNegative costs
– Can make us slaves andCan make us slaves and
dehumanise usdehumanise us
– Can reduce productivityCan reduce productivity
and safetyand safety
– Can waste money, timeCan waste money, time
and effortand effort

How Do We Balance ThisHow Do We Balance This
Making trade offs:Making trade offs:
– Perfection is not possible, Computer systems arePerfection is not possible, Computer systems are
inherently flawed and changing fastinherently flawed and changing fast
How close to perfection do we expect them to be?How close to perfection do we expect them to be?
What is an acceptable error rate?What is an acceptable error rate?
Should we spend more on testing before release?Should we spend more on testing before release?
Or wait till users find problems?Or wait till users find problems?
– How do we achieve an ideal?How do we achieve an ideal?
Should computer professionals be held liableShould computer professionals be held liable
for damages caused?for damages caused?

Making trade offsMaking trade offs
How do we weigh up the expense of change forHow do we weigh up the expense of change for
a negligible gain in performance?a negligible gain in performance?
Or assess the costs of improvement versus theOr assess the costs of improvement versus the
costs of failure?costs of failure?
– In terms of liability, customer dissatisfaction, ….In terms of liability, customer dissatisfaction, ….
Can one waste money on too much safety inCan one waste money on too much safety in
safety-critical computer systems?safety-critical computer systems?

Professional & EthicalProfessional & Ethical
ConsiderationsConsiderations
Flaws can cause serious disruption/danger, butFlaws can cause serious disruption/danger, but
complexity of modern computer systems makescomplexity of modern computer systems makes
errors, oversights, etc. a near certaintyerrors, oversights, etc. a near certainty
Risks of using computers should be comparedRisks of using computers should be compared
withwith
– i) risks of other methods andi) risks of other methods and
– ii) with benefits obtainedii) with benefits obtained

Does not mean that computer errors andDoes not mean that computer errors and
failures should be excused or ignoredfailures should be excused or ignored
Does not mean that carelessness orDoes not mean that carelessness or
negligence should be toleratednegligence should be tolerated
Does not mean that accidents should beDoes not mean that accidents should be
excusedexcused
– as part of learning processas part of learning process
– or because, on balance, contribution ofor because, on balance, contribution of
computers is overwhelmingly positivecomputers is overwhelmingly positive

Should serve to remindShould serve to remind
computer professionals ofcomputer professionals of
importance of doing jobimportance of doing job
responsiblyresponsibly
– They must assess risksThey must assess risks
carefully and honestlycarefully and honestly
– must include safety protectionsmust include safety protections
– Eg plans for backup, shutdownEg plans for backup, shutdown
and recoveryand recovery

Pci lecture 2 software it failures (1)

More Related Content

Similar to Pci lecture 2 software it failures (1) (20)

Recently uploaded (20)

Pci lecture 2 software it failures (1)