Framework supporting middleware executed on mobile devices in ...

mashpeemoveMobile - Wireless

Nov 24, 2013 (3 years and 8 months ago)

88 views

Chair for Communication Technology (ComTec), Faculty of Electrical Engineering / Computer Science

Service Creation

What’s the issue ??


Keynote @ WAWC’07



Dr. Olaf Drögehorn



Lappeenranta, August 16th, 2007

©
ComTec 2007

Dr. Olaf Droegehorn

2

Vision of ubiquitous computing

©
ComTec 2007

Dr. Olaf Droegehorn

3

Technologies

Markets

Users

Market players

Licenses and regulation

Business models

Billing and pricing

Competition

Operator strategies

Fight between standards

Standardisation bodies

Networks

Terminals

Software tools

Content management

Security

User needs

Service delivery

Types of services

Cost of services

Content

Ease of use

Personalisation

Privacy

Security

Drivers and barriers!

Technologies, markets and users

©
ComTec 2007

Dr. Olaf Droegehorn

4

Devices



User interface



Capabilities



Development platform



Network interface

Heterogeneous

networks



xDSL



Mobile networks



Bluetooth



IEEE 802.x



Digital TV or radio





Market players



Telecom operators



Other network operators



Broadcasters



Content providers



Content aggregators



Service providers



Payment providers

Content

management

Software

APIs

Terminals

Payment

Security

Quality
-

of
-
service

Networks

Personalization

Privacy

Digital

rights

User needs :



To communicate



To form communities



To be informed



To be entertained



To be efficient



To feel secure



To feel capable


and “in control”



To be educated

Services: Putting technology to work …



Services exist, because they fulfill a need or solve a problem for the end users



A good service must be designed for the target users and the context of use


-

and be user
-
friendly!



How can we make use of present and future networks, terminals and software


to address specific user needs?



Who are the stakeholders of a service?

Content server

(database)

Mobile phones

PDAs

PCs / Internet

Digital TV

Networks

Content

provider

Content

aggregator

Network

operator

Service

provider

(Mobile) end users

Regulator

Payment

provider

Equipment

producer

Services and networks
-

the big picture

©
ComTec 2007

Dr. Olaf Droegehorn

5

The Challenge for the Future of
Communication Service (Creation)

So, what shall we do, now

that we can do everything?

Bruce Mau,

Author of “S,M,L,XL”

©
ComTec 2007

Dr. Olaf Droegehorn

6

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

7

What is a Service ?


An application ?


A piece of software ?


An outcome of a process ?


=> There are many definitions (business,


economics, technologies, etc.)



=> Something, that fulfills a need


(for help, for fun, for communication)


=> Is UMTS fulfilling ANY need at all ?? ;)

©
ComTec 2007

Dr. Olaf Droegehorn

8

Things that can make a service fail


There is no need for the service


If nobody uses the service, the whole thing is a failure


The user interface doesn’t work


Bad navigation design, bad graphics


The service is too slow


Content / bandwidth ratio, application doesn’t respond


The service is unreliable


Application crashes! Blank screens, hangs, resets


Data is not up
-
to
-
date, e.g. yesterday’s weather


Transactions fail
-

what happened to my order?


The service is insecure


Personal data compromised, money disappears


The service isn’t used
-

boring concept

©
ComTec 2007

Dr. Olaf Droegehorn

9

Understanding a need


Example: Just want to get out ….


Looking for „EXIT“, or similar


Arriving here, you find:


©
ComTec 2007

Dr. Olaf Droegehorn

10

Service Creation challenges

Josef: There are no services, why ?


Why is Service Creation so difficult ?


Maybe we are expecting THE service

(like the killer
-
application)



We are NOT developing variations

(like product
-
designers do)



Most services are developed from scratch

(Users don‘t recognize elements/widgets,
therefore we try to enhance an already
available service)



We are not reusing existing
codes/technologies/widgets/platforms


©
ComTec 2007

Dr. Olaf Droegehorn

11

Service Creation, why do we need to talk
about it ?


User needs are not addressed



Software service creation lifecycles are by far
too long


Most services are developed from the scratch


Service composition is proprietary, at best



Ease the life of professional software
developers


Ease the life of providers/operators


Enable „non
-
professionals“ to create their
services


=> Each and every user should be able to create


services

©
ComTec 2007

Dr. Olaf Droegehorn

12

Why should EVERYBODY do that?


Looking to the web:


HTML was for specialists only


Nowadays everybody has a web
-
page/blog



New hype:


User generated content (for the web)


Flickr, LifeBlog, etc.



Next trend:


User customized services


Google Mash
-
Ups


©
ComTec 2007

Dr. Olaf Droegehorn

13

Why should we do that ?


SMS was never meant to be a
product/service


By using it, it formed a hype



Because people like to separate from
each other


=> The idea is to enable everybody to


improve/customize and generate


services

©
ComTec 2007

Dr. Olaf Droegehorn

14

Variations are needed
-

IKEA

(Users like to have the choice)

©
ComTec 2007

Dr. Olaf Droegehorn

15

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

16

„I wish to concentrate on improving my
services functionalities, and not
anything else...“

The service creation problem

for Service developers

©
ComTec 2007

Dr. Olaf Droegehorn

17

„There must be some simple ways to put
all the functionalities together as ready
-
to
-
sell services ...“

The service creation problem

for Service providers

©
ComTec 2007

Dr. Olaf Droegehorn

18

„Is it possible for me to customize service
with functionalities that i wish to have?“

The service creation problem

for End
-
users

©
ComTec 2007

Dr. Olaf Droegehorn

19

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

20

Service Creation Vision

A world of services that are…



Easy to create
,


Easy to share,


Easy to use,


…and still user
-
centric!

©
ComTec 2007

Dr. Olaf Droegehorn

21

Service Creation Vision (2)


Easy to create



Creation tools and publishing


Service taxonomies


Reuse of existing services and components


Semantic orchestration of components and loosely
coupled approach



Easy to share



Generalised client
-
server architecture



«

My server in my pocket

», «

My server at home»


Service deployment in just a few clicks


Semantic based publishing



Easy to use



Semantic Service discovery


Fine grain semantic
-
based search


Interoperability, composability of services

©
ComTec 2007

Dr. Olaf Droegehorn

22

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

23

The classical way to think about
services


Domain
-
Specific Software
Architectures (DSSA)



Build a specific „platform“ for a
specific domain

©
ComTec 2007

Dr. Olaf Droegehorn

24

Domain
-
Specific Software
Architectures (DSSA)


DSSA definition


„a context for patterns of problem elements, solution
elements, and situations that define mappings between
them“ (Software Engineering Institute, 1990)


Comprises a couple of crucial artifacts


scenarios


domain dictionary


context and entity/relationship diagrams


data flow, state transition, and object models


functional
-

and non functional requirements




Problem: DSSA is still driven by technology

©
ComTec 2007

Dr. Olaf Droegehorn

25

Service Categorization today with
regard to DSSA

Source: UMTS Forum, http://www.umts
-
forum.org

A typical DSSA just reflects these typical technical enablers

©
ComTec 2007

Dr. Olaf Droegehorn

26

Domain specific platforms


The issues are NOT new



The whole middleware hype was targeted
to this



Build once, use everywhere



Why not simply using THE platform ?

©
ComTec 2007

Dr. Olaf Droegehorn

27

Middleware supports applications

Application

Middleware

Device

Device


Middleware provides
homogeneous API to
applications


By providing services that
bridge between
homogeneous and
heterogeneous API



To shorten time
-
to
-
market
for applications


To abstract from
heterogeneity and allow
applications to be executed
on future mobile devices

©
ComTec 2007

Dr. Olaf Droegehorn

28

The original goals of

platforms & middleware


Masking heterogeneity


networks, end
-
systems, OSs, programming
languages



Providing a useful distributed
programming model


simplicity with generality





CORBA, Java RMI, etc. have been very
successful in this... business applications;
wrapping of legacy systems...

©
ComTec 2007

Dr. Olaf Droegehorn

29

The concept of middleware

is used more widely


Platforms were built for many different domains, like:



Applications


eCommerce, 7x24 back
-
end servers


eScience


real
-
time, embedded systems


mobile agent systems


peer to peer platforms


mobile computing applications


telecomms/ programmable networking



Underlying systems


PCs/ workstations


supercomputers


wireless PDAs


embedded devices


network processors


wireless, sensor, infrared etc. networks

©
ComTec 2007

Dr. Olaf Droegehorn

30

A victim of its own success?


CORBA tries to cope ...


add asynch., transactions, fault
-
tolerance, persistence, components,
load balancing, logging, real
-
time/ embedded/ lightweight CORBA,
...



complexity and unmanageability



prototypes arise to fill the gaps & to have smaller
solutions


asynchronous platforms (pub
-
sub, eventing, message queueing)
(MSMQ, JMS, JavaSpaces, ...)


Grid middleware (Legion, Globus, OGSA, ...)


web services (SOAP, WSDL, WSFL, ...)


mobility middleware (Rover, MOST, ...)


mobile agent systems (Tacoma, Aglets, ...)


peer
-
to
-
peer (JXTA, Jtella, ...)


real
-
time/ multimedia platforms (ReTINA, DIMMA, ...)


etc.



different assumptions, paradigms, models,
implementation environments, ...



incompatible platforms, no cooperation, no reusability!

©
ComTec 2007

Dr. Olaf Droegehorn

31

Result: Many platforms for

different disciplines

©
ComTec 2007

Dr. Olaf Droegehorn

32

Ending up in technology silos

Source: UMTS Forum, http://www.umts
-
forum.org

A typical DSSA just reflects these typical technical enablers

©
ComTec 2007

Dr. Olaf Droegehorn

33

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

34

A better way to think about services ?


Semantic
-
Driven Software
Architectures (SDSA)



Design a service without technology
in mind

©
ComTec 2007

Dr. Olaf Droegehorn

35

SDSA: Understand Services

beyond Technology

1.
No biased interfaces

_
keyboard/mouse, voice, gestures, touch
display

2.
No limiting form factors

_
mobile phone, notebook, desktop, PDA, smart
phone

3.
No usage constraints, i.e.

_
Unlimited power supply

_
Bandwidth abundance

_
Workflow convenience: no/short login, no
hardware break
-
down, full
-
fledged transparent
security, ...

4.
No prejudice about network connection (fixed vs.
wireless)


… but describe precisely the context, services are being used

©
ComTec 2007

Dr. Olaf Droegehorn

36

Domain
-
Specific vs. Semantics
-
Driven Software Architectures

DSSA

SDSA


can be an assemblage of
software components


focuses on describing the
problem domain


wants to be unambigious


makes contradictions a
valuable design principle


usually has a bias on device
form factors and interaction
paradigms


is open to device
implementation and interaction
design


just refers to an unambigious
terminology


makes usage of the
terminology a central feature of
service specification


attempts to generalize for the
purpose of a reference design


is specific to understand one
specific problem domain

©
ComTec 2007

Dr. Olaf Droegehorn

37

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

38

Activity Theory

Tools
&
artefacts
Telephone
Rules
Mutual listening
Common Language
Rules
Mutual listening
Common Language
Community
Group of peers
Division
of
Effort
Sender / Receiver: Mutual delivery
Receiving messages
Subject
Partners
in Dialog
Activity
Object
Dialog modulation
Outcome
Agreement
©
ComTec 2007

Dr. Olaf Droegehorn

39

Understanding a service comprises more than workflow specification

Rules


Icons instead of text

Community


Assembly assistants


IKEA complaints hotline

Subject

Outcome

Object

Tools

Division of Effort


Assemble


Explain, not assist

From individual activities to

Activity Systems

©
ComTec 2007

Dr. Olaf Droegehorn

40

SMS acts as a
mediating artefact
of an Activity System

Object

Division of Effort

Subject

Rules

Community

Tools

Outcome


Register for service


Pay per text


Mobile providers


Possible senders/receivers


Store and forward text


Charge for service


Sender


Receiver


Create & send limited text


Receive text


Text delivered


Text received


T9 keyboard


Display


Activity Systems as a core idea for real
-
world services (here SMS)

©
ComTec 2007

Dr. Olaf Droegehorn

41

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion


©
ComTec 2007

Dr. Olaf Droegehorn

42

The Wireless World Research Forum


WWRF, founded in 1998

(from WSI & EU
-
FP5 Project)



Working Groups trying to draft future
research issues



Definition of a reference model of User
-
centered communication (I
-
centric)



WhitePapers on specific topics

©
ComTec 2007

Dr. Olaf Droegehorn

43

Terminals

Devices and Communication

End Systems

Service Platform

Generic Service Elements

for all layers

Service Semantic

Wired or wireless Networks

IP based

Communication

Subsystem

Business

Model

Networks

IP Transport Layer

Network Control & Management Layer

Service Support Layer

Service Execution Layer

Application Support Layer

Service


Bundling

Service

Control

Service


Discovery

Service


Creation

Environment


Monitoring

Service


Deployment

Conflict

Resolution

Ambient

Awareness

Personalization

Adaptation

User Model & Appl. Scenarios

Communication Space

(Contexts & Objects)


Reference Model for

I
-
Centric Communications

©
ComTec 2007

Dr. Olaf Droegehorn

44

WhitePaper on Service Creation


A WhitePaper to highlight open issues, existing
technologies & approaches


Technology driven service design

vs. Technology agnostic approaches


Existing Software Development Environments (IDEs)


Business models for Service Creation



Identified need for:


Harmonization of tools & semantics


Technology agnostic/independent way of serivce
design/description


Common semantics


Common understanding of technology enablers

(like OSA/Parlay, ParlayX, IMS, etc.)

©
ComTec 2007

Dr. Olaf Droegehorn

45

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

46

Semantic
Service

Discovery


Technology agnostic

Everyone

can define

own services

„Draw a service“

-

Simplified service creation process

One idea for service creation

©
ComTec 2007

Dr. Olaf Droegehorn

47

The S4ALL approach

Rule


„Draw a service“
-

Simplified service creation process

©
ComTec 2007

Dr. Olaf Droegehorn

48

Rule

Business Rules

Rule Evaluator

State Services

Action Services

©
ComTec 2007

Dr. Olaf Droegehorn

49

Service Building Blocks (SBB)


State SBB: Read status to enable
reacting on state changes. Also from
Context Providers, which enablers to
easily create context aware mobile
services



Action SBB: Trigger actions on services
at mobile device, in the home
environment, in the Internet



Rules connect State SBBs and Action
SBBs

©
ComTec 2007

Dr. Olaf Droegehorn

50

Time

Afternoon

Location

LasVegas

Start Presentation

&

Icomp_2007.ppt

Business Rule Example

©
ComTec 2007

Dr. Olaf Droegehorn

51

WI
-
Platform

Business Rule
Evaluator

Mood

Time

Game

Time

State ServiceBB

Mood

State ServiceBB

Game

Action ServiceBB

Rule

Evaluator

Business Rule Example (2)

©
ComTec 2007

Dr. Olaf Droegehorn

52


Contains following components






Rule

Evaluator

(Info.)

Collector

Service

Triggering

Rules

State info.
(from
service)


Output

Action Service(s)

Rule Reasoner

Rule

Repository

Rule Evaluation

©
ComTec 2007

Dr. Olaf Droegehorn

53

Selection of the Business

Rules approach


Model
-
Driven
-
Architecture (MDA) approach for service design:

Create a graphical model of the service



Business rule engine approach very common for connecting
bigger building blocks. Enables straightforward data flow
design.



Make it easy for the service developers, not giving the power of
a programming language



Especially for Web service connection (state and action service
building blocks provide Web service interfaces) this approach
fits perfectly



The ease and power of the approach depends on the provided
service building blocks. They have to enforce the „Service
-
oriented architecture“ approach

©
ComTec 2007

Dr. Olaf Droegehorn

54

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion

©
ComTec 2007

Dr. Olaf Droegehorn

55

Semantic Service Delivery Platform
























SDF

Semantic
Descriptions


SemSDP


IDE Extension

Service Creation
Workbench


deploys

publishes

SCE

Semantic
Descriptions


SEE

Semantic
Service

Discovery

Business Rules
Execution

Service
Building
Block
Execution
Engine

Triggers

Service Delivery Platform


©
ComTec 2007

Dr. Olaf Droegehorn

56

Service Creation Workbench


Scenario shown: Play video on user’s mobile device when he/she passes a certain location

©
ComTec 2007

Dr. Olaf Droegehorn

57

Rules, Facts, and …

State

SBB

Converter

Action

SBB

Atom

©
ComTec 2007

Dr. Olaf Droegehorn

58

The S4All Environment

Service Creation Workbench

Service Discovery Wizard

Rule Editing

Deployment GUI

SEE Interface

Rule editor GUI

Repository Search

Syntax/Semantic Discovery

State Service

Building Blocks

Business Rules

Action Service

Building Blocks

State/Context

Sources

Actions

Service Execution Environment

Execution Environment

Deployment Interface

Subscription Management

User Subscriptions

Business Rule Engine

CRUD Services

state request

evaluation

triggering

discover

discover

CRUD

Repository

Building Blocks

Rules

©
ComTec 2007

Dr. Olaf Droegehorn

59

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion


©
ComTec 2007

Dr. Olaf Droegehorn

60

Evaluation (1)


The SCW makes creation of new services very intuitive

=> enabling the combination of state information from
user’s environment (“context”) with action triggering



Our exemplary scenarios show that using the S4ALL
tools significantly improves the process for developers to
create mobile services



Service building blocks can be reused



Service logic is easy to grasp through its graphical
representation



Service logic can easily be updated



Deployment requires “one
-
click”


©
ComTec 2007

Dr. Olaf Droegehorn

61

Evaluation (2)


The tools need a suitable SEE



Rule Evaluation is a common task and is very well
supported



The approach can be used by operators as well as End
-
Users



A larger test
-
case will be done in a European Project (IST
-
SPICE)



A set buidling blocks will be provided soon, for a specific
SEE (WI
-
P) as open source

©
ComTec 2007

Dr. Olaf Droegehorn

62

Outline


The problem of service (creation)


What might people want (depends on the role)


The vision of Service Creation


The old way of service creation


A new way of service creation


Activity Theory


WWRF WhitePaper on Service Creation


An example how to do user
-
driven Service
Creation


The Service Creation Workbench


Findings from the example


Conclusion


©
ComTec 2007

Dr. Olaf Droegehorn

63

Conclusion (1)


Services are not really there


Service design/creation is technology driven


Many solutions don‘t serve a need


Develop some solution and try to sell it as a service
(MMS)


A lot of people were looking for THE service


Software
-
Development tools are out there, but
how to build a service with them?


No harmonization in tools and semantics


No service design, but technology push


©
ComTec 2007

Dr. Olaf Droegehorn

64

Conclusion (2)


An easier process for rapid service creation is
needed


Easy service composition / customization is
needed (to build variations)


Tools need to be harmonized to use existing
enablers (let‘s use IMS meaningfully)


Business opportunities need to be exploited


=> Find more in the WWRF
-
WhitePaper

Chair for Communication Technology (ComTec), Faculty of Electrical Engineering / Computer Science

Thank you !


droegehorn@uni
-
kassel.de