interaction design, user interfaces, and the future of ... · interaction design, user interfaces,...

26
1 Constantine & Lockwood, Ltd. Prof. Larry Constantine, Prof. Larry Constantine, IDSA IDSA , ACM Fellow , ACM Fellow Department of Mathematics and Engineering Department of Mathematics and Engineering University of Madeira University of Madeira Interaction Design, User Interfaces, Interaction Design, User Interfaces, and the Future of Model and the Future of Model - - Driven Driven Design and Development Design and Development

Upload: lyduong

Post on 08-Feb-2019

227 views

Category:

Documents


0 download

TRANSCRIPT

1

Constantine & Lockwood, Ltd.

Prof. Larry Constantine, Prof. Larry Constantine, IDSAIDSA, ACM Fellow, ACM FellowDepartment of Mathematics and EngineeringDepartment of Mathematics and Engineering

University of MadeiraUniversity of Madeira

Interaction Design, User Interfaces, Interaction Design, User Interfaces, and the Future of Modeland the Future of Model--Driven Driven

Design and DevelopmentDesign and Development

2

Constantine & Lockwood, Ltd.

3

Constantine & Lockwood, Ltd.

No Conspiracy, But No AccidentNo Conspiracy, But No Accident•• The place and role of users, usability, and The place and role of users, usability, and

user interfaces in software development.user interfaces in software development.•• The place and role of The place and role of

programmers in programmers in software development.software development.

4

Constantine & Lockwood, Ltd.

Do Users and the Context of Use Matter?Do Users and the Context of Use Matter?•• Inside the cockpit of Inside the cockpit of

the Airbus 320the Airbus 320……•• Video games Video games

anyone?anyone?

“Congratulations to our pilot, newly

promoted to Captain!”

“Congratulations to our pilot, newly

promoted to Captain!”

5

Constantine & Lockwood, Ltd.

Why Design for UseWhy Design for Use•• A product can be a success without being useful.A product can be a success without being useful.•• But usefulness and usability pay.But usefulness and usability pay.

Simpler documentation, Simpler documentation, lower support costs.lower support costs.Faster learning, lower training costs.Faster learning, lower training costs.Fewer errors, better data integrity.Fewer errors, better data integrity.Improved efficiency, dependability.Improved efficiency, dependability.Increased customer and endIncreased customer and end--user satisfaction.user satisfaction.

Design for use is an Design for use is an alternative to alternative to competing by competing by commodity pricing or commodity pricing or through feature bloat.through feature bloat.

6

Constantine & Lockwood, Ltd.

In the BeginningIn the BeginningDeus ex Deus ex machinamachina……

Technology at the center.Technology at the center. UserUser--friendly design.friendly design.UI and usability as afterUI and usability as after--thethe--fact decoration.fact decoration.

7

Constantine & Lockwood, Ltd.

Advances in ComputingAdvances in Computing•• IBM 1401IBM 1401•• AutocoderAutocoder•• ProgressProgress

–– machine languagemachine language–– assembly languageassembly language–– highhigh--level languageslevel languages–– 3GLs3GLs–– 4GLs4GLs–– application generatorsapplication generators–– executable modelsexecutable models

8

Constantine & Lockwood, Ltd.

1986. What has Changed, Ed?1986. What has Changed, Ed?•• ExceleratorExcelerator

(Index Technology)(Index Technology) Same old, same old. But now it’s object-oriented programming. And CASE.

Same old, same old. But now it’s object-oriented programming. And CASE.

Just pretty pictures. Good marketing. The real stuff is database and code generation!

Just pretty pictures. Good marketing. The real stuff is database and code generation!

9

Constantine & Lockwood, Ltd.

Why Models?Why Models?•• Building models is easier than building the real thing.Building models is easier than building the real thing.•• Models capture, carry, and organize understanding about a Models capture, carry, and organize understanding about a

problem or possible solutions.problem or possible solutions.•• Models permit exploration of problem and solution spaces.Models permit exploration of problem and solution spaces.•• Models can be validated against objective criteria.Models can be validated against objective criteria.•• Models can be tested and evaluated.Models can be tested and evaluated.MODELMODEL--DRIVEN PROCESSES:DRIVEN PROCESSES:•• provide an provide an ““audit trailaudit trail”” of assumptions, of evolving of assumptions, of evolving

understanding, and basis of solutions.understanding, and basis of solutions.•• facilitate tracing results back to requirements.facilitate tracing results back to requirements.•• enable reasoned derivation of successful solutions.enable reasoned derivation of successful solutions.

10

Constantine & Lockwood, Ltd.

Some Holes in the StorySome Holes in the Story

DATABASEDATABASEDESIGNDESIGN

INPUTINPUT--OUPUTOUPUTDESIGNDESIGN

11

Constantine & Lockwood, Ltd.

12

Constantine & Lockwood, Ltd.

Where are the Users?Where are the Users?

•• UML and the unified processUML and the unified process(9 books, (9 books, ©© 19971997--2003; 3175pp)2003; 3175pp)

•• Use casesUse cases(6 books, (6 books, ©© 19971997--2003; 1913pp)2003; 1913pp)

•• XP and agile developmentXP and agile development(6 books, (6 books, ©© 20002000--2002; 1868pp)2002; 1868pp)

In 21 classic, respected books In 21 classic, respected books ––

12.712.7 6767 (2.1%)(2.1%)

18.718.7 4747 (2.5%)(2.5%)

2.52.5 8 8 (0.4%)(0.4%)

INDEXINDEX TOTALTOTAL % of% ofENTRIES*ENTRIES* PAGESPAGES PAGESPAGES

* users, user interfaces, usability, GUI, end users,* users, user interfaces, usability, GUI, end users,……

3 books have only 1 mention;3 books have only 1 mention;3 have nothing at all to say!3 have nothing at all to say!

Even the gurus donEven the gurus don’’t get it!t get it!

13

Constantine & Lockwood, Ltd.

Models MatterModels Matter•• Models should be suited to the design task.Models should be suited to the design task.•• UML can be used to model user interface design UML can be used to model user interface design

and interface architecture, butand interface architecture, but……•• Some design modeling is better than none at all.Some design modeling is better than none at all.•• But big, clumsy, complicated models tend to lead toBut big, clumsy, complicated models tend to lead to……

14

Constantine & Lockwood, Ltd.

Model-Driven Everything

• model-driven user interfaces?• model-driven engineering• executable models• model-driven architecture• model-driven development• model-driven design• models in analysis and design

15

Constantine & Lockwood, Ltd.

ModelModel--Generated User InterfacesGenerated User Interfaces

•• Some processes can target alternate Some processes can target alternate deployment: Web, desktop, deployment: Web, desktop, smartphonesmartphone,,……

•• Some even attempt integrated Some even attempt integrated ““beautification.beautification.””

Pederiva et al., “The Beautification Process in Model-Driven Engineering of User Interfaces” Interact 2007

16

Constantine & Lockwood, Ltd. And the results?And the results?•• RealReal--world app generated by world app generated by OlivaNovaOlivaNova

PederivaPederiva et al., et al., ““The Beautification Process in ModelThe Beautification Process in Model--Driven Engineering of User InterfacesDriven Engineering of User Interfaces”” Interact 2007Interact 2007

Now “beautified.”Now “beautified.”

17

Constantine & Lockwood, Ltd.

Getting Better, Getting WorseGetting Better, Getting Worse•• To avoid To avoid ““analysis paralysis,analysis paralysis,”” agile development rejects agile development rejects

thoughtful, thorough design. No BDUF!thoughtful, thorough design. No BDUF!•• Users can be shortchanged in Users can be shortchanged in ““customercustomer--centriccentric”” focus focus

on features, delivered value, customer satisfaction.on features, delivered value, customer satisfaction.

“GUI-intensive projects are problematical for XP (and probably for many [agile] approaches).”

—Ron Jeffries

“GUI-intensive projects are problematical for XP (and probably for many [agile] approaches).”

—Ron Jeffries

“[Usability] is not a ‘weak point’[of the agile methods], it is an absence.”

—Alistair Cockburn

“[Usability] is not a ‘weak point’[of the agile methods], it is an absence.”

—Alistair Cockburn

•• But, things may be getting better But, things may be getting better --–– Jeff Patton, agileJeff Patton, [email protected]@yahoogroups.com–– Scott Ambler, Agile ModelingScott Ambler, Agile Modeling–– Cutter Agile Management Cutter Agile Management ““The Usability ChallengeThe Usability Challenge””

Big Design Up FrontBig Design Up Front

18

Constantine & Lockwood, Ltd.

Still, . . . Still, . . . •• UserUser--centered design is no guarantee either.centered design is no guarantee either.•• User experience design is often misguided because User experience design is often misguided because

user performance is what is truly important.user performance is what is truly important.•• Industrial design does no better than software Industrial design does no better than software

engineering.engineering.

“It probably won a prize!”—Donald Norman

“It probably won a prize!”—Donald Norman

“Focus upon humans detracts from support for the activities themselves.”

—Donald Norman

“Focus upon humans detracts from support for the activities themselves.”

—Donald Norman* www.jnd.org

19

Constantine & Lockwood, Ltd.

DRAW

If Only Things Were SimpleIf Only Things Were SimpleManagement fantasyManagement fantasy•• The new programmer The new programmer ––•• The new designer The new designer ––EveryoneEveryone’’s fantasys fantasy•• The new user The new user ––

CODE?

20

Constantine & Lockwood, Ltd.

Points of DeparturePoints of Departure•• Study process and UI problems (1991Study process and UI problems (1991--1992).1992).•• Use cases promising but early results disappointing. Use cases promising but early results disappointing.

KeysKeys

•• Essential use cases and basic process (1993).Essential use cases and basic process (1993).•• Widely used and practiced: Siemens, McKesson, Nortel, Widely used and practiced: Siemens, McKesson, Nortel,

Navy, Lockheed, Gartner, DaimlerChrysler, IBM,Navy, Lockheed, Gartner, DaimlerChrysler, IBM,……•• Full spectrum of problems: eFull spectrum of problems: e--commerce, automation, commerce, automation,

healthcare, finance, banking, insurance, education, healthcare, finance, banking, insurance, education, embedded systems, CAD/CAM, office products, embedded systems, CAD/CAM, office products, ……

•• Integrated with varied processes: OOSE (Integrated with varied processes: OOSE (AhlqvistAhlqvist, , 1996), RUP/USDP, XP, agile modeling,1996), RUP/USDP, XP, agile modeling,……

simplification not elaborationsimplification not elaborationabstraction not realismabstraction not realismintentions not interactionintentions not interaction

21

Constantine & Lockwood, Ltd.

Unified Modeling?Unified Modeling?

•• For data modeling you want For data modeling you want appropriate models by skilled appropriate models by skilled data modelers.data modelers.

•• For process modeling you For process modeling you want appropriate models by want appropriate models by skilled process modelers.skilled process modelers.

•• For interaction modeling you For interaction modeling you want appropriate models by want appropriate models by skilled interaction designers.skilled interaction designers.

* A. Chanda, E. Allen, Rice University

•• Different models and different skills are needed Different models and different skills are needed for different parts of the modeling process!for different parts of the modeling process!

22

Constantine & Lockwood, Ltd.

Much is AskedMuch is Asked•• Focus shifted from users and user experience to Focus shifted from users and user experience to

human activity and user performance.human activity and user performance.•• ModelModel--driven inquiry grounded in systematic models driven inquiry grounded in systematic models

of human activity.of human activity.•• ModelModel--based process for user interface and based process for user interface and

interaction design.interaction design.•• Models tailored to the needs of Models tailored to the needs of

interaction design and designers.interaction design and designers.•• Models based in powerfulModels based in powerful

abstractions and separation abstractions and separation of concerns.of concerns.

•• Tools supporting flexible, seamless problem solving Tools supporting flexible, seamless problem solving with all models interconnected.with all models interconnected.

UsageUsage--centered Designcentered Design

Canonical Abstract PrototypesCanonical Abstract Prototypes

Human Activity ModelingHuman Activity Modeling“With dynamic extensions!”

“With dynamic extensions!”

* Nóbrega, University of Madeira

23

Constantine & Lockwood, Ltd.

Human Activity ModelingHuman Activity Modeling

•• Based in activity theory, models human activity:Based in activity theory, models human activity:““loosely ordered collection of flexibly adaptive actions loosely ordered collection of flexibly adaptive actions having distinct but disparate goals contributing to a having distinct but disparate goals contributing to a shared or common purpose.shared or common purpose.””

•• Simple notation, small number Simple notation, small number of concepts systematizing of concepts systematizing activity theory, linked with UML.activity theory, linked with UML.

•• Captures and organizes aspects of activity Captures and organizes aspects of activity context most salient and relevant to interaction context most salient and relevant to interaction design in support of real user requirements.design in support of real user requirements.

“Business process modeling done right!”

- Nuno Nunes

“Business process modeling done right!”

- Nuno Nunes

24

Constantine & Lockwood, Ltd.

Anatomy of an AbstractionAnatomy of an Abstraction

Supports separation of concerns.Supports separation of concerns.And they are And they are ““executableexecutable””!*!*

abstract UI components abstract UI components with standardized userwith standardized user--

centered semanticscentered semantics

dynamic relationships dynamic relationships among UI componentsamong UI components

content, content, function, function,

layout and layout and organizationorganization

CAPsCAPs and and dCAPsdCAPs: formalized : formalized abstract wireframe schematics.abstract wireframe schematics.

* Nóbrega, University of Madeira

25

Constantine & Lockwood, Ltd.

Prescription for ProgressPrescription for Progress•• Models to capture, hold, and organize Models to capture, hold, and organize

understanding of problems and solutions.understanding of problems and solutions.•• Focus on user performance, human activity.Focus on user performance, human activity.•• Tools to support, not take over human activity.Tools to support, not take over human activity.•• Fix and finish UML (notations suited for human Fix and finish UML (notations suited for human

activity and user interaction modeling).activity and user interaction modeling).•• Use people for what people are best at: Use people for what people are best at:

creative exploration, analysis, invention, creative exploration, analysis, invention, and design!and design!

OROR…… “Hello. Are you listening OMG?”

“Hello. Are you listening OMG?”

26

Constantine & Lockwood, Ltd.

Thank you.Thank you.

[email protected]@UMa.ptwww.LabUSE.orgwww.LabUSE.org