1. Seamless Knowledge with Topic Maps
A Standard Model for Metadata, Taxonomies, Ontologies, and Knowledge Management
Steve Pepper
Chief Strategy Officer, Ontopia Convenor, SC34/WG3 Editor, XML Topic Maps

2. Ontopia – The Topic Maps Company
Our mission:
To provide Topic Maps technology and services for information and knowledge management
Background:
Established April 2000 out of STEP Infotek
Headquarters in Oslo, Norway
Partners in 13 countries around the world
Recognized leaders of the Topic Maps community
Products:
Ontopia Knowledge Suite™
Consultancy, training, and application development through partners
On'topia, 1999.[f. Gr. ‘onto-’ (being) + Gr. ‘topos’ (place); see -IA.] I. An imaginary world in which knowledge is well organized. II. A company that provides tools to help you realize your own Ontopia…
Norway Partners:
Bouvet AS Lava Group

3. What is Topic Maps?
An International Standard for subject-based organization of information and knowledge management
More importantly…
What is Topic Maps used for?
(What are topic maps used for?)
Organizing large bodies of information
Capturing organizational memory
Representing complex rules and processes
Supporting concept-based eLearning
Managing distributed knowledge and information
Aggregating information and knowledge
etc…
Seamless Knowledge

4. Perspectives on Topic Maps
The information management perspective
A standard for subject-based organization of information in support of findability
The knowledge management perspective
A knowledge representation formalism optimized for use in information management – the world’s first standard for KM
The library science perspective
A way to collocate all knowledge about a subject – in particular its relationship to other subjects and to information resources 4

5. ISO 13250: Background and current status
Origins date back to early 1990’s
Davenport Group  GCARI / CAPH  ISO
First Edition
International Standard (ISO/IEC 13250:2000)
Model and syntax based on SGML
Second Edition
XML Topic Maps (XML version for use on the Web)
ISO 13250: 2003 (includes XTM)
Revised Edition
Multipart standard appearing in 2005
Includes data model, query language, constraint language

6. Topics Associations Occurrences Taxonomies Metadata Ontologies
The TAO of Topic Maps
Topics Associations Occurrences
Taxonomies Metadata Ontologies

7. The core Topic Maps model
Callas, Maria …………………… 42
Cavalleria Rusticana … 71,
Mascagni, Pietro
Cavalleria Rusticana . 71,
Pavarotti, Luciano ……………… 45
Puccini, Giacomo ………. 23, 26-31
Tosca ………………. 65,
Rustic Chivalry, see Cavalleria Rusticana
singers ………………………
baritone ………………………. 46
bass ………………………
soprano ……………… 41-42, 337
tenor ………………………
see also Callas, Pavarotti
Tosca ………………… 65,
The core concepts of Topic Maps are based on those of the back-of-book index
The same basic concepts have been extended and generalized for use with digital information
Envisage a 2-layer data model consisting of
a set of information resources (below), and
a “knowledge map” (above)
This is like the division of a book into content and index
knowledge layer
information layer
(index)
(content)

8. (1) The information layer
The lower layer contains the content
usually digital, but need not be
can be in any format or notation or location
can be text, graphics, video, audio, etc.
This is like the content of the book to which the back-of-book index belongs
information layer

9. (2) The knowledge layer
The upper layer consists of topics and associations
Topics represent the subjects that the information is about
Like the list of topics that forms a back-of-book index
Associations represent relationships between those subjects
Like “see also” relationships in a back-of-book index
composed by
composed by
Tosca
Puccini
born in
Madame Butterfly
Lucca
knowledge layer

10. Linking the layers through occurrences
The two layers are linked together
Occurrences are relationships with information resources that are pertinent to a given subject
The links (or locators) are like page numbers in a back-of-book index
composed by
composed by
Tosca
Puccini
born in
Madame Butterfly
Lucca
knowledge layer
information layer

11. Summary of core concepts
Some pool of information or data
any type, any format, any location
information
Let’s look at some TAOs in the Omnigator…
A knowledge layer consisting of:
knowledge
Topics
a set of topics representing the key subjects of the domain in question
Puccini
Tosca
Lucca
Madame Butterfly
Associations
representing relationships between subjects
composed by
born in
Occurrences
links to information that is somehow relevant to a given subject
The TAO of Topic Maps

12. current topic
(multiple) types
multiple typed occurrences
multiple names
multiple typed associations

13. With this simple but flexible model you can
Represent subjects explicitly
Topics represent the “things” your users are interested in
Capture relationships between subjects
Associations provide user-friendly navigation paths to information
They also promote serendipitous knowledge discovery through browsing
Make information findable
Topics provide a “one-stop-shop” for everything that is known about a subject
Occurrences allow information about a common subject to be linked across multiple systems or databases
Represent taxonomies and thesauri
Associations may represent hierarchical relationships
Topic Maps permits multiple, interlinked hierarchies and faceted classification
Transcend simple hierarchies
Rich associative structures capture the complexity of knowledge and reflect the way people think
Manage knowledge
The topic map is the embodiment of “corporate memory”

14. Topic Maps and ontologies
The term “ontology” is used in many different ways:
“An ontology is the types and subtypes of concepts and relations that exist in some domain…” John Sowa: Knowledge Representation (Pacific Grove, 2000)
In Topic Maps the basic building blocks are
Topics: e.g. “Puccini”, “Lucca”, “Tosca”
Associations: e.g. “Puccini was born in Lucca”
Occurrences: e.g. “ is a biography of Puccini”
Each of these constructs can be typed
Topic types: “composer”, “city”, “opera”
Association types: “born in”, “composed by”
Occurrence types: “biography”, “street map”, “synopsis”
All such types are also topics (within the same topic map)
“Puccini” is a topic of type “composer” … and “composer” is also a topic
A topic map thus contains its own ontology!

15. Five cool things to do with a topic map
Querying
Constraining
Filtering
Visualizing
Merging

16. Querying topic maps Topic Maps is based on a formal data model
This means that topic maps can be queried, like databases
ISO Topic Maps Query Language (TMQL)
Companion to ISO 13250, currently being balloted in ISO
Allows more powerful use of taxonomies to retrieve information
Permits queries that would make Google boggle (see below)
TMQL is based on Ontopia’s query language tolog
Demo of querying in the Omnigator
Query example:
Give me all composers that composed operas that were based on plays that were written by Shakespeare

17. Constraining topic maps
ISO itself provides no way to constrain topic maps
Examples of constraints:
“All persons must be born somewhere”
“A person may have died somewhere”
Constraints are necessary in order to:
Permit semantic validation of content
Ensure consistency
Enable more intuitive user interfaces
Simplify application development
ISO Topic Maps Constraint Language (TMCL)
Companion to ISO 13250, currently being balloted in ISO
Will interoperate with OWL (Web Ontology Language)
Ontopia has developed OSL for its customers
Demo of OSL in the Omnigator

18. Filtering, scoping and personalizing topic maps
Multiple world views
Reality is ambiguous and knowledge has a subjective dimension
Scope allows the expression of multiple perspectives in a single topic map
Typical application: Combining related but divergent taxonomies
Contextual knowledge
Some knowledge is only valid in a certain context, and not valid otherwise
Scope enables the expression of contextual validity
Personalized knowledge
Different users have different knowledge requirements
Scope permits personalization based on personal references, skill levels, security clearance, etc.
Demo of scope-based filtering in the Omnigator

19. Visualizing topic maps
The network or graph structure of a topic map can be visualized for humans
This provides another “view” on information that can lead to new insights
Demo of visualization using Vizigator

20. Merging topic maps Topic Maps can be merged automatically
You can always and in any situation take any two arbitrary topic maps and merge them to a single topic map
This cannot be done with databases or XML documents
The merge capability enables many advanced applications
Information integration across repositories
Sharing and reusing taxonomies
Automated content aggregation
Distributed knowledge management
The concept that makes merging possible is subject identity
Topic Maps has a robust mechanism for using URIs as identifiers

21. Principles of merging in Topic Maps
In Topic Maps, every topic represents some subject
The collocation objective requires exactly one topic per subject
When two topic maps are merged, topics that represent the same subject should be merged to a single topic
When two topics are merged, the resulting topic has the union of the characteristics of the two original topics
name
occurrence
association role
A second topic (in another topic map) “about” the same subject T
name
occurrence
association role T
...and the resulting topic has the union of the original characteristics
name
occurrence
association role T
Merge the two topics together...
Demo of merging in the Omnigator…

22. How Topic Maps improves access to information
Intuitive navigational interfaces for humans
The topic/association layer mirrors the way people think
Powerful semantic queries for applications
A formal underlying data structure
Customized views based on individual requirements
Personalization based on scope
Information aggregation across systems and organizations
Topic Maps can be merged automatically…

23. Applications of Topic Maps
Taxonomy Management
Metadata Management
Semantic Portals…
Information Integration
eLearning
Business Process Modelling
Product Configuration
Business Rules Management
IT Asset Management
Asset Management (Manufacturing)

24. Taxonomy management
Addresses the problem of managing unstructured content
Organization by subject is seen as the solution – because that’s how users search
More and more companies are looking into and developing taxonomies
A taxonomy is a simple form of topic map
Topic Maps provides subject-based organization de-luxe
Using Topic Maps offers many benefits:
Standards-based means vendor independence and data longevity
Associative model allows for evolution beyond simple hierarchies
The taxonomy can also be used as a thesaurus, a glossary or an index
Identity model permits merging and reuse
The Dutch Tax and Customs Administration (Belastingdienst) uses the OKS as the basis of a taxonomy management system
This capability can also be added to Content Management Systems

25. Metadata management
On behalf of the Norwegian Government Administration Services Lava Group is building a metadata server
Metadata for government publications will be managed using the OKS
Will be used in the central public information portal (ODIN)
(System currently under development)
The system provides
Authoring system used by the editors
Vocabulary Editor for adjusting the metadata vocabulary used
Metadata Export to various systems
Web services based on the metadata
Unique identifiers for documents
Unparallelled future flexibility
Indexes
Engine
ODIN
Meta- data …
FAST
Search engine
ODIN
Metadata server (OKS)
Logistics
Exported subjects
ASCII-export
MUP
Lovdata

26. Semantic portals
Topic Maps as Information Architecture for web delivery applications
Web sites, portals, corporate intranets, etc.
Site structure is defined as a topic map
Each page represents a topic (subject-centric)
User-friendly navigation paths defined by associations
Topics used to classify content
High potential for portal connectivity using TMRAP
Permits evolution towards Knowledge Management solutions
The OKS has been used to create portals, e.g.
Kulturnett.no (Norwegian public sector portal to cultural information):
Apollon (University of Oslo research magazine):

27. Dynamic Content Aggregation
An Application of Seamless Knowledge
Automatic Portal Integration
Topic Maps Remote Access Protocol

28. Semantic portals Think of Topic Maps as an Information Architecture
Topic Maps is an ideal model for portals and other forms of web-based information delivery
The basic concept is to have the topic map drive the portal
Not just a navigational layer on top of something else
The very structure of the portal is a topic map
All content is organized around topics (“subject-centric organization”)
Each page represents a topic (we call this a “Topic Page”)
Topics act as points of collocation
They provide a “one-stop shop” for everything that is known about a particular subject
Navigating the portal == Navigating the topic map
Associations provide very intuitive navigation (“As we may think”)

29. A topic page the current topic (multiple) types multiple names
multiple typed occurrences
multiple typed associations

30. current topic
occurrences
associations

31. the current topic
multiple names
multiple occurrences
multiple associations

32. The rise and rise of semantic portals
In Norway, this concept has been put into practice on a scale that is verging on the industrial, especially among government agencies
At present there are over a dozen, with more on the way
Some semantic portals in Norway:
In production
(Ministry of Education)
(Research Council of Norway)
(Consumers Association)
(Norwegian Defence)
(Ministry of Agriculture)
(Ministry of Justice)
(Ministry of Culture)
Under development
(Norwegian Conservative Party)
Skatteetaten (Tax Office)
Statsministerens kontor (Office of the Prime Minister)
Statistisk Sentralbyrå (Central Bureau of Statistics)
IFE/Halden (Nuclear Reactor Project)
etc.

33. Towards seamless knowledge
As the number of portals multiplies, the amount of overlap increases…
Take these three portals as an example:
forskning.no (Research Council web site aimed at young adults)
forbrukerportalen.no (Public site of the Norwegian Consumer Association)
matportalen.no (Biosecurity portal of the Department of Agriculture)

34. Genetically modified food at forskning.no

35. Genetically modified food at Forbukerrådet

36. Genetically modified foodstuffs at Matportalen

37. Three semantic portals – One common subject
 one “virtual portal”
with seamless navigation in all directions

38. Achieving seamless knowledge
Very little is required for these portals to achieve a simple but effective form of Seamless Knowledge
They have already achieved subject-centric organization of their content
Without this, Seamless Knowledge is beyond reach
From a technical perspective, only two additional pieces are required to complete the puzzle:
#1 An identity mechanism
To make it possible to know when their subjects are the same
Published subjects solve this problem
A flexible and robust mechanism for using URIs as global identifiers
See
#2 An exchange protocol
To enable information to be requested and exchanged automatically
Ontopia has developed Topic Maps Remote Access Protocol

39. Topic Maps Remote Access Protocol (TMRAP)
Hi! Do you know the subject “genetically modified food”?*
* The actual question was: Is the subject known in your system?
Portal A: forskning.no
Sure! My Topic Page is at 
Portal B: Matportalen
This scenario (called VISIT) is supported by TMRAP

40. The Omnigator Rap demo (Part 1: VISIT)
Two Omnigators are running on this machine
Different browsers (Opera and Internet Explorer)
Different skins (Ontopia National Colours and Vive Québec)
Different names pepper poivre
Different TMs (Italian Opera and Various Geographical TMs)

41. VISIT: Some considerations
The functionality is deceptively simple, yet potential very powerful
From the user’s point of view the VISIT links might have been hand-coded (there is no visible difference)
The cool thing is that they are generated entirely automatically
This is dynamic content aggregation in practice!!
NO MAINTENANCE of cross-site links is required
Solves the publisher’s cross-site link management problem with technology
And we can go a step further with relatively little effort
Rich data based on Topic Maps can be merged …
… so we can exchange not only links,
but also whole chunks of rich data
We call those chunks topic maplets
This is how it works…

42. TMRAP “GET” scenario using topic maplets
Hi! What do you know about “genetically modified food”?*
* The actual question was: What information do have about in your system?
Portal A: forskning.no
Oh, this and that. Here you are. Be my guest!
Portal B: Matportalen
This scenario (called GET) provides another level of content aggregation

43. GET: Some considerations
The functionality is even more powerful…
The “seamlessness” factor is much greater
(In fact we have “dumbed it down” in this demo in order to show what is actually going on: The GET functionality could be activated automatically)
Application areas are slightly different:
Useful when seamlessness is more important and branding issues less important
E.g., within a corporate or government environment

44. “Now! …. That should clear up a few things around here!”
Conclusion
“Now! …. That should clear up a few things around here!”
Subject-based classification provides a solution to the findability problem
Topic Maps are the international standard of choice for doing this
Topic Maps can represent
taxonomies
thesauri
indexes
metadata
ontologies
…all in a single, intuitive model
Any questions?

45. What now? Read The TAO of Topic Maps* Download the Omnigator*
Learn LTM* and create your own first topic map
Consider doing a thesis on a Topic Maps-related subject
Attend Ontopia’s training class*
June 6th: Full day introduction to Topic Maps
June 7th: Ontology design for Topic Mappers
(other days cost money)
Make sure you register!
All details at

46. Similarities Differences Interoperability
RDF and Topic Maps
Similarities
Differences
Interoperability

47. “Two households, both alike in dignity…”
During the late 1990s the W3C and ISO developed two semantic technologies in parallel
Two communities, largely unaware of each other
Tackling the same fundamental problems
Findability
Semantic interoperability
The results were RDF and Topic Maps

48. How the two families stack up
OWL
TMCL
TMQL
SPRQL
RDF Schema
QUERY
Topic Maps
ORG SYNTAX MODEL CONSTRAINTS
RDF
ORG SYNTAX MODEL REASONING
XML
LTM
HyTM
XTM
RDF/XML
RDF/A N3
ISO Seamless Knowledge
W3C Semantic Web

49. Similarities that cry out for unification
Striking similarities
Both “extend” XML into the realm of semantics
Both allow assertions to be made about subjects in the outside world
Both define abstract, associative (graph-based) models
Both are intensely concerned with “identity”
Both allow some measure of inferencing or reasoning
Both have XML-based interchange syntaxes
Both have constraint languages and query languages
This lead to calls for unification
“Free-for-all” between the two Erics at Extreme Markup 2000
Michael Sperberg-McQueen suggested locking the RDF people and the Topic Maps people in a room together until they had harmonized the two…

50. But there are important differences too…
Different roots
Topic Maps has its roots in traditional finding aids (indexes, thesauri, etc.)
RDF has its roots in document metadata and formal logic
Different levels of semantics…
RDF is more low level; Topic Maps has more higher-level semantics
Different models
Identity, scope, association roles, n-ary relationships, variant names, …
Different goals
RDF: An artificially intelligent web for software agents
Topic Maps: Findability and knowledge integration for humans
So unification never happened
But the perception of rivalry is a cause for confusion

51. It’s time to move beyond bigotry
Let’s look for the synergies instead!
Both families have user communities
Neither standard will go away anytime soon
Common interest in the success of semantic technologies
Semantics are hard enough to explain to the market as it is
A standards war will indeed lead to a Plague o’ Both Our Houses…
RDF and Topic Maps are different
Different strengths, different weaknesses
Let’s recognize this
And let’s go for interoperability
That’s the goal of RDFTM…

52. RDFTM RDF/Topic Maps Interoperability Task Force
A task force within the Semantic Web Best Practices and Deployment Working Group
Chartered to deliver two documents:
Survey of Existing Interoperability Proposals (WG Note)
Guidlines for RDF/Topic Maps Interoperability (WG Note or Recommendation)
First draft of Survey recently delivered to WG
First draft of Guidelines for Extreme Markup 2005

53. RDF or Topic Maps? Some rules of thumb
The basic premise:
RDF is more low-level; oriented towards machines
Topic Maps is more high-level; oriented towards humans
OWL is a step beyond; oriented towards artificial intelligence
Do you simply want to encode document metadata?
RDF is an ideal model for assigning properties to documents (e.g. Dublin Core, PRISM) – you probably won’t need OWL
Do you want to achieve subject-based classification of content?
Topic Maps provides the best combination of flexibility and user-friendliness
Do you want both metadata and subject-based classification?
Go straight for Topic Maps – or, if you already use RDF for metadata, “view” it as a topic map in conjunction with a TM-based taxonomy or subject classification
Do you want to enable applications based on software agents?
Use RDF/OWL on a foundation of Topic Maps-based knowledge organization