Copyright Management for the LUISA Semantic Learning Content Management System Roberto García Universitat de Lleida, Spain Tomas Pariente ATOS Origin SAE,

Copyright Management for the LUISA Semantic Learning Content Management System

Roberto García

Universitat de Lleida, Spain

Tomas Pariente

ATOS Origin SAE, Spain

www.luisa-project.eu

LUISA is an European Comission FP6 funded Specific Targeted Research Project

2

Table of Contents

• Introduction• Motivation• Related Work• Proposal• Copyright Model• License Modelling• Conclusions & Future Work

3

Introduction

• “Web-spread” of e-Learning solutions requires Interoperability

• Enabler: richer semantics than current metadata specifications for learning resources

• LUISA (EU-IST-FP6): reference semantic and service-oriented architecture for e-Learning

• Search, interchange and delivery of learning objects• Face interoperability at different levels using

Semantic Web technologies

4

Motivation

• At the content copyright level:eLearning DRM interoperability

• Main problem: lack of structured and formal ways to express the licensing terms of learning objects

• Sakai: predefined and simple copyright status sentences

• “Material is in public domain” or “I hold copyright”

• “Use copyright below” + text box

• LOM: set of attributes for stating learning object rights: • “Cost”, “Copyright and Other Restrictions” and

“Description”

5

Related Work

• Creative Commons:• Focus on open licensing schemes

• e.g. Open Courseware

• Extension for custom licensing schemes (CCPlus), but not based on formalised license building blocks

• DRM Standards: • ISO/IEC MPEG-21 Rights Expression Language (REL)• Open Digital Rights Language (ODRL)

6

Related Work

• DRM Standards issues [1]:• Don’t scale well to open environments like the Web

• More appropriate for closed domains

• Cause interoperability issues• One of the main DRM end-users complains

• Rights language limited expressivity • Difficult to accommodate copyright law

• Standardisation based on formal language grammar, no formal semantics

[1] Electronic Frontier Foundation Report by Doctorow, 2005

7

Proposal

• Use the Copyright Ontology [1] in order to model e-Learning copyright licenses

• More expressive than XML approaches• Facilitates interoperability, works at the semantic level• Based on a copyright law model, WIPO worldwide

recommendations [2]• Also based on Semantic Web technologies

• Copyright Ontology:provides the building blocks and restrictions to flexibly model LOs licensing terms

[1] Copyright Ontology, http://rhizomik.net/ontologies/copyrightonto[2] World Intellectual Property Organisation, http://www.wipo.int

8

Copyright Model

Copyright

EconomicRights

RelatedRights

MoralRights

DistributionRight

ReproductionRight

PublicPerformanceRight

FixationRight

CommunicationRight

AttributionRight

TransformationRight

IntegrityRight

DisclosureRight

WithdrawalRight

PermorfersRights

ProducersRights

BroadcastersRights

RentalRight

ImportationRight

SoundRecordRight

MotionPictureRight

BroadcastingRight

MakingAvailableRight

AdaptationRight

TranslationRight

Rights Model

9

Copyright Model

Fixation

Instance

Manifestation

Work

Performance

Objects Processes

Communication

AbstractionsVictor Hugo’s

Les Misérables

Creation Model

10

Copyright Model

Fixation

Instance

Manifestation

Work

Performance

Manifest

Perform

Improvise

Record

Copy

Copy

Objects Processes

CommunicationCommunicate

Abstractions

Derive

Distribute Retransmit

Action Model

11

Copyright Model

• Actions governed by rights, economic rights: • Reproduction Right: Copy• Distribution Right: Distribute

• More specifically: Sell, Rent and Lend

• Public Performance Right: Perform• Fixation Right: Record• Communication Right: Broadcast and Make Available.• Transformation Right: Derive

• Specialisations: Adapt and Translate

12

License Modelling

• Primitive actions – case roles action participants

initiator resource goal essence

Actionagent, effector

instrument result, recipient

patient, theme

Processagent, origin

matter result, recipient

patient, theme

Transferagent, origin

instrument, medium

experiencer, recipient

theme

Spatial origin path destination location

Temporal start duration completion pointInTime

Ambient reason manneraim, consequence

condition

13

License Modelling

• Combine these building blocks to model licenses• Both commercial and open access terms

• Main component: Action Pattern • Defined combining restrictions using logical operators

Copy

pointInTime [2008/01/01-2008/06/30]

agent = Subscribers

them

e =

lear

ning

Obj

ect

Non-Commercial

copy doi:10.1032/… actions

pointInTime ≥2008/07/01

licensed actions

Pattern ≡ Copy ⊓∀pointInTime.≥ 2008-01-01,

≤ 2008-06-30 ⊓∃agent.Subscribers ⊓∃theme.{learningObject}

14

License Modelling

• Additional actions and case roles: • Agree (Disagree): state what is permitted (prohibited)

• condition: links to another pattern that must satisfied• aim: links condition pattern to conditioned one

• consequence: links pattern to obliged one

:agreement a co:Agree;co:agent :owner; co:theme :Pattern.

Condition ≡ Transfer ⊓∃recipient.{owner} ⊓∃theme.{3EurosAmount} agent.Subscribers ∃ ⊓∃aim.Pattern (≤ 1 aim)⊓

:Patternco:condition :Condition.

15

License Modelling

• Description Logic reasoners • implement pattern matching

• If instance action classified in OWL Class for pattern

• Use SPARQL queries to check action permitted by agreed pattern and condition fulfilment:

ASK {?agreement rdf:type co:Agree; co:theme ?pattern.:copy rdf:type ?pattern; co:agent ?consumer; co:condition ?conditionPattern.?condition rdf:type ?conditionPattern; co:agent ?consumer; co:aim :copy.}

Co

py

pointInTime [2008/01/01-2008/06/30]

agent = Subscribers

them

e =

lear

ning

Obj

ect

Non-Commercial

copy doi:10.1032/… actions

pointInTime ≥2008/07/01

licensed actions

16

Conclusions & Future Work

17

Conclusions & Future Work

• Semantic DRM module for LUISA based on the Copyright Ontology

• Facilitate interoperability• Flexible enough to accommodate different licensing

schemes

• Future work: • Model LUISA partners licenses• Test the whole range of LUISA scenarios• Check scalability