EMMOs - Enhanced Multimedia Meta Objects for Re-purposing of Knowledge Assets

EMMOs–Enhanced Multimedia Meta Objects for Re-purposing of Knowledge Assets

Sunil Goyal,Siegfried Reich and Wernher Behrendt

Salzburg Research/SunTREC

Jakob Haringer Stra?e5/III

5020Salzburg,Austria

{sgoyal,sreich,wbehrendt}@salzburgresearch.at

Abstract.There exist a broad variety of concepts and tools for support-

ing knowledge workers during the various phases of their work.However,

structures created with one tool—e.g.a mindmap editor capturing a

discussion process—usually cannot be re-used by other tools or in other

contexts.The reasons are manifold,ranging from variations in syntactic

expressions to semantic heterogeneity.In this paper we argue that there

is a need for capturing and representing explicitly the structures not only

between multimedia objects but also between the various stages in a de-

velopment process.We outline a document model which we envisage to

store these relationships and we name that model“EMMO—Enhanced

Multimedia Meta Object”.EMMOs are enhanced by explicit knowledge

structures,and they are?rst-class,i.e.,they can be exchanged,traded

and re-used in di?erent contexts.

1Introduction

Knowledge workers require support for re-purposing of knowledge assets com-posed of multimedia documents.This paper addresses the domain of interactive electronic publishing.Our rationale is that the“document of the future”will be generated on the?y,based on knowledge assets,will contain any type of media, and will be available in appropriate form,for interactive use,to humans as well as machines,anywhere,when needed.

These visions date back to the early ideas of the hypertext pioneers,including Bush[2]or Engelbart[4].Today,there exist a large number of tools for the di?er-ent application domains[8]:browsers for the various breeds of navigational hy-pertext,spatial browsers for the domain of spatial hypertext[6],and editors and viewers for issue based information systems[3],for example QuestMap T M[9]. Exchange between these and re-purposing of existing relationships in di?erent contexts,however,is still an open issue.We argue that there is a need for an inclusive model that manages the relationships between the di?erent models.

Let us consider the following example.

Two years ago,Mr.Smith,CEO of Sporty-Cars,Inc.had put forward his vision–a Next Generation Sports Car–using MS PowerPoint T M slides for his upcoming board meeting.The market research done earlier had shown impressive

2S.Goyal,S.Reich and W.Behrendt

demand.Everyone at the board meeting cheered.The product design teams went through a rigorous argumentation process and captured it using tools like QuestMap T M.They discussed every single issue,prepared manuals,built CAD simulations and3D models.Everything looked impressive and the project was given a green light for prototype development.It was a great day for the company. Mr.Smith along with his colleagues came to see the prototype.Mr.Smith got into the new car and was suddenly plunged into darkness.The car seat had sunk down!Was he too heavy for the seat?

Mr.Smith wants to know where exactly did things go wrong?Which events led to what?But where should he look and analyze?In the MS PowerPoint T M slides,QuestMap T M,manuals,CAD simulations or3D models?

Current tools often lack explicit semantics,they pose interoperability prob-lems due to semantic heterogeneity and they are therefore unable to solve the above problem.

2EMMOs—Enhanced Multimedia Meta Objects EMMOs provide a concept of portable,stand-alone and living meta objects which assist users in traversing associations between artefacts.EMMOs are ob-jects that describe complex relationships between artefacts,can be loaded into browsers and modi?ed by users.They also keep the history of interactions be-tween the resources and their users and set individualised viewpoints over the information space.Therefore,EMMOs cover the development process and allow for knowledge assets to be re-used in di?erent contexts.Figure1gives a graph-ical representation of an EMMO.In particular,the Figure shows the di?erent interfaces for accessing and manipulating the internal structures.

An EMMO is composed of the following structure:

–A knowledge template holds domain intelligence pertaining to the universe of discourse.

–A set of transformation descriptions to allow the EMMO knowledge to be represented in various output formats.

–A links repository holds instantations of relationships between artefacts in digital libraries.The available types of relationships are speci?ed by the knowledge template(see above).

–A contents repository represents a persistent store which holds additional information–e.g.annotations pertaining to the instance.

–A trails repository keeps history of the usage of the EMMOs link repository.–A set of standard interfaces for authoring,an inspection environment,trad-ing interface,resource(self)description interface,module management and import/export interfaces.

The EMMO concept came originally from the domain of intertextual studies in literature and arts where scientists want to make explicit the di?erent relation-ships between texts in a way that approximates the process of cultural contex-tualization characteristic of interpretative processes(see also the project CUL-TOS—Cultural Units of Learning,Tools and Services https://www.wendangku.net/doc/4c5135808.html,/).

depicts_scene(j,k)

Chagall Breughel

Descriptive

Configuration

“... on each hand the flames

driven backward slope

their pointing spires, ...

then with expanded wings

he steers his flight

aloft, incumbent on the

dusky air

Milton, Paradise Lost,

Book 1, 222 - 226

Fall”

and deploy these portable

di?erent publishing scenarios.

within knowledge assets

reused in a di?erent context

viewpoints,fuzzy separations, etc.)

–News Publishing:generating types of news on the?y from archives.

–Education:support in the form of trail based systems,argumentation sup-port,community building tools etc.

–Engineering discipline:vast amount of technical documentation combined with questions such as“why did we design that?”.

–Executive Knowledge Management:support of knowledge management and decision support systems.

The idea of autonomous knowledge objects has been proposed in several vari-ations.Tsichritzis[11]envisioned mobile knowledge objects(KNOs)to be used in software engineering and o?ce automation.Memes[1]followed the approach that behaviours and ideas are copied from person to person by imitation and Tanaka[10]proposed a framework for building,managing and trading Memes. However,we argue that these approaches are not based on an underlying ontol-ogy,they are often domain speci?c and they do not address issues of semantic heterogeneity;therefore,they are unable to capture and support the whole de-velopment process.

EMMOs5 3Summary and Discussion

We have outlined the idea of multimedia meta objects which are enhanced by knowledge and which capture the relationships between knowledge assets so that they can be re-used in di?erent contexts such as news publishing,technical documentation,automotive engineering or software development etc.

Hence,we believe the EMMOs concept to be general and applicable to mul-tiple domains(which is what would make it a“C-level”concept following En-gelbart’s categorisation,see https://www.wendangku.net/doc/4c5135808.html,/).

Discourse,trails,perspectives,knowledge extraction,and others,are key con-cepts along with knowledge assets which provide assistance to knowledge work-ers during various phases of their work.We believe that there is a need for an infrastructure(C-level)to support the above needs of knowledge workers and applications can leverage this infrastructure and re-purpose it according to their own needs.More speci?cally,we believe that the following components are part of that infrastructure:

–a structure server supporting multiple interaction styles or hypermedia ap-plication domains[7]);

–an ontology system that manages relationships based on ontologies(which in turn capture a domain’s or community’s concepts);

–a process support system that allows users to keep track of the status of their work[5].This could well be an adapted work?ow management system;–a?le system which supports users’so that they can manipulate EMMOs with their existing tools;

–viewers and transformers allowing for di?erent representations of knowledge assets.

ACKNOWLEDGEMENTS

CULTOS is co-funded by the European Commission under the IST Programme (Information Society Technologies)under grant No.IST-2000-28134.

References

1.Susan Blackmore.The power of MEMES.Scienti?c American,280:52–61,October

2000.

2.Vannevar Bush.As we may think.The Atlantic Monthly,176(1):101–108,1945.

3.James H.Coombs Conklin and Michael L.Begeman.gIBIS:A hypertext tool for

exploratory policy discussion.ACM Transactions on O?ce Information Systems, 6(4):303–331,1988.

4.Douglas https://www.wendangku.net/doc/4c5135808.html,puter-Supported Cooperative Work:A Book of Read-

ings.Edited by Irene Greif,chapter Toward High-Performance Knowledge Workers.

Morgan Kaufmann,1988.

5.J¨o rg M.Haake and Weigang Wang.Flexible support for business processes:Ex-

tending cooperative hypermedia with process https://www.wendangku.net/doc/4c5135808.html,rmation and Software Technology,41(6):355–366,1999.

6S.Goyal,S.Reich and W.Behrendt

6.Catherine C.Marshall and Frank M.Shipman.Spatial hypertext and the practice

of information triage.In Proceedings of the’97ACM Conference on Hypertext, April6-11,1997,Southampton,UK,pages124–133,1997.

7.Peter J.N¨u rnberg,Kaj Gr?nb?k,Dirk Bucka-Lassen,Claus A.Pedersen,and Olav

Reinert.A component-based open hypermedia approach to integrating structure services.New Review of Hypermedia and Multimedia,pages179–205,1999.

8.Peter J.N¨u rnberg,John J.Leggett,and Erich R.Schneider.As we should have

thought.In Proceedings of the’97ACM Conference on Hypertext,April6-11, 1997,Southampton,UK,pages96–101,1997.

9.Simon Buckingham Shum.Negotiating the construction of organisational memo-

ries.In U.M.Borgho?and R.Pareschi(Ed.),Information Technology for Knowl-edge Management,pages55–78,1998.

10.Yuzuru Tanaka and Jun Fujima.Meme media and topica architectures for edit-

ing,distributing and managing intellectual assets.In2000Kyoto International Conference on Digital Libraries:Research and Practice.November13-16,Kyoto, Japan,pages208–216,November2000.

11.Dionysios C.Tsichritzis.O?ce Automation,Concepts and Tools,chapter15Ob-

jectworld,pages379–398.Springer,1985.