A top-level ontology for smart document access

Abstract

The central idea of this paper is to apply the concept of a top-level ontology to make access and retrieval of resources, more specifically documents, in knowledge-intensive business processes smarter. Supporting the reuse of meta-data created elsewhere, the sharing of meta-data across applications and users as well as the application of meta-data to answer user enquiries about resources requires a mapping between diverse content-oriented meta-data standards. This can be done with the help of a top-level ontology. In this paper, we will analyze a number of widely used meta-data standards for personal information, communication, resource descriptions and specific standards for text, images, audio and video. These cover a wide range of electronic resources typically required by weakly structured, knowledge-intensive business processes which are then mapped to the presented top-level ontology.

Full text

A TOP-LEVEL ONTOLOGY
FOR SMART DOCUMENT ACCESS
RONALD MAIER
School of Management, Information Systems
Leopold-Franzens-University of Innsbruck, A-6020 Innsbruck, Austria
E-mail: ronald.maier@uibk.ac.at
JOHANNES SAMETINGER
Inst. f. Wirtschaftsinformatik – Software Engineering,
Johannes Kepler University Linz, A-4040 Linz, Austria
E-mail: johannes.sametinger@jku.at
The central idea of this paper is to apply the concept of a top-level ontology to make access and
retrieval of resources, more specifically documents, in knowledge-intensive business processes
smarter. Supporting the reuse of meta-data created elsewhere, the sharing of meta-data across
applications and users as well as the application of meta-data to answer user enquiries about
resources requires a mapping between diverse content-oriented meta-data standards. This can be
done with the help of a top-level ontology. In this paper, we will analyze a number of widely used
meta-data standards for personal information, communication, resource descriptions and specific
standards for text, images, audio and video. These cover a wide range of electronic resources
typically required by weakly structured, knowledge-intensive business processes which are then
mapped to the presented top-level ontology.
1. Introduction
Businesses and organizations can be characterized by increasingly knowledge-intensive
work, activities and business processes [Wolff 2005, 38]. Employees working in
knowledge-intensive business processes typically face a multitude of heterogeneous
applications with which they access a wealth of electronic resources scattered across a
number of systems. These resources are implicitly linked to each other, e.g., by having
the same authors, discussing the same topics, using the same formats, referring to the
same geographical location or, last but not least, being created, altered or applied in the
same processes or projects. Because these links are only implicit, users get no support in
retrieving all resources relevant for the tasks they are working on.
Meta-data have been proposed to establish these links and languages have been
standardized in order to annotate resources and share meta-data across applications and
contexts. Corresponding tools have been developed to support creation, storage and
retrieval of meta-data-based descriptions of resources [Maier/Peinl 2005]. However,
whereas languages and tools have been standardized, there are a large number of content-
oriented meta-data standards for diverse application areas and also annotating resources
and thus creating meta-data can become a nuisance to the user.
Section 2 describes the basic terms documents, meta-data and ontology. Section 3
reviews widely used meta-data standards for personal information, communication and
1

2
resource descriptions. Section 4 shows how they can be mapped to a top-level ontology
and discusses how this helps to solve KM challenges. Section 5 briefly compares our
approach with related work before Section 6 finally concludes the paper.
2. Document Descriptions
In the following, the three concepts documents, meta-data and ontologies are briefly
reviewed from the perspective of ICT environments that provide support for knowledge-
intensive work, also called enterprise knowledge infrastructures [Maier et al. 2005].
2.1. Documents
Organizations create, store, transfer and use more and more data within their boundaries,
but they also exchange data with suppliers, customers and partners. The forms and types
of data have been extended largely from structured data found in (relational) data base
systems to semi-structured data typically found in file systems, document or content
management systems. Compared to structured data, semi-structured data has not been
managed equally well in most organizations. A large number of terms have been coined
in order to show the various types of semi-structured data, e.g., content, (digital) asset,
document, (knowledge, learning or media) object. Documents represent a certain type of
container for content, e.g., including formats or meta-data. In the following, we will use
the term document as we are interested in the container metaphor. Thus, documents can
be exchanged including meta-data so that they can be embedded in a new context at the
receiver’s end and accessed smartly with the help of the meta-data.
Documents can be searched for with the help of full text search tools included in the
operating system or installed on top, e.g., Desktop search engines, search engines for file
servers. Meta-data based search, however, is not possible in these cases. When passive
documents are uploaded into document management systems (DMS), meta-data search is
possible, but in most cases restricted to manually attributed meta-data. Also, when the
document is sent to a different environment, it looses all meta-data as these are stored in
a DMS data base separated from the documents. This challenge is taken on by the
concept of active documents. According to this approach, documents include meta-data
and functionality and can be classified into passive, adaptive, active and proactive
documents, depending on the amount of functionality that is realized with the help of
active components [Schimkat 2003, 54, Carr et al. 2004].
2.2. Meta-data
Meta-data is data about data. The term refers to any data used to aid identification,
description and location of resources. Examples of meta-data are title, author, date or
location used to describe a book or person stored in catalogues, television guides,
taxonomies, or tables of contents. Meta-data is an increasingly ubiquitous term that is
understood in different ways by different professional communities that design, create,

3
describe, preserve, and use information systems and resources. Until the mid-1990s,
meta-data was a term most prevalently used in the domain of geospatial data and data
management and systems design and maintenance in general. For these communities,
"meta-data" referred to a suite of industry or disciplinary standards as well as additional
internal and external documentation and other data necessary for the identification,
representation, interoperability, technical management, performance, and use of data
contained in an information system [Swetland, Baca 2000]. Many meta-data formats
exist, see Section 3, some of these are simple in their descriptions; others are complex
and rich.
2.3. Ontologies
Knowledge modeling aims at a formal description of (documented) organizational
knowledge that can be processed by computers. Ontologies are formal models of an
application domain that help to exchange and share knowledge with the help of ICT
systems. “An ontology is a formal, explicit specification of a shared conceptualization”
[Gruber 1993, 199]. Ontologies have to be formal which requires that they should be
machine-readable. However, there are different degrees of formality of ontologies, from
a thesaurus like WordNet to ontologies capturing formal theories for common-sense
knowledge like Cyc. Explicit specification means that concepts and relationships as well
as constraints on the use of concepts are defined openly and not left to the interpretation
of the ontology’s users. Shared refers to the requirement that the conceptualizations
made in an ontology have to be agreed upon. Finally, conceptualization is an abstract
model, a representation of a domain or phenomenon which investigates concepts of that
domain or phenomenon relevant to the ontology’s users.
Ontologies are not static, but evolve over time. Ontologies model objects in
domains, relationships among those objects, properties, functions and processes
involving the objects and constraints on and rules about the objects [Daconta et al. 2003,
190]. Thus, ontologies support clear-cut, concise, semantically rich and unambiguous
communication between persons aided by an ICT environment. In the Semantic Web
stack [Berners-Lee et al. 2001], the Web Ontology Language (OWL) helps to
semantically map descriptions of resources serialized with the help of the Resource
Description Format (RDF). OWL provides a standardized markup language that can be
used to describe the top-level ontology that maps the meta-data standards described in
Section 3.
3. Meta-data Standards
There are several high-level meta-data standards. One high-level schema is the Meta-data
Encoding and Transmission Standard (METS) developed by the Digital Library
Federation. METS proposes to define meta-data elements for complex digital objects. It
offers a flexible mechanism for encoding descriptive, administrative, and structural meta-
data for a digital library object, and for expressing the complex links between these

4
various forms of meta-data. It is therefore a useful standard for exchanging digital library
objects between repositories. In addition, METS provides the ability to associate a digital
object with behaviours or services.
There are many initiatives for domain-specific content-oriented standardization.
Knowledge work comprises active participation in processes of knowledge conversion,
i.e. according to Nonaka and Takeuchi [1995] processes of externalization, socialization,
internalization and combination. From the perspective of enterprise knowledge
infrastructures, these processes are supported by publishing, discovery, collaboration,
learning as well as integration and search services [Maier 2007, 318]. The underlying
resources that support these services can be structured into personal information,
personal communication, resource information, and specific information like text,
images, audio and video which are examined in the following.
3.1. Personal Information
Personal information contains address books, calendars, task lists, wish lists, schedules,
notes, etc. There are many software systems available that manage all or part of this
information. In former days, paper-and-binder personal organizers have been used to
administer personal information. These were small books containing a calendar, an
address book and ruled pages for arbitrary notes. Nowadays, this information is held on
our computers and synchronized with personal digital assistants (PDAs) to be
conveniently available at any place.
Meta-data standards and specifications for personal information include vCard,
vCalendar, FOAF, etc. vCard automates the exchange of personal information. It is a set
of meta-data elements that covers and extends information typically found on a
traditional business card [vCard]. vCalendar defines a transport and platform-
independent format for exchanging calendaring and scheduling information in an easy,
automated, and consistent manner [vCalendar]. It captures information about event and
to-do items that are used by applications such as personal information managers and
group schedulers. FOAF stands for Friend Of A Friend [FOAF]. It is based on XML and
RDF and is intended to manage online communities. The FOAF namespace provides a
set of properties and classes, focusing initially on people, documents, organizations,
images.
3.2. Personal Communication
Communication is the process of exchanging information, e.g., in order to share
knowledge and experiences. To a great extent, personal communication is done via e-
mail messages. Other forms of personal communication include chat rooms, instant
messaging, bulletin boards, and newsgroups. Personal communication may support voice
and video.
An important standard for personal communication is the simple mail transfer
protocol (SMTP or ESMTP), a text-based protocol for transferring messages. It is an

5
application layer protocol that uses TCP/IP to transport and route its messages. SMTP
uses commands like HELO or AUTH and status codes to communicate conditions
between sender and receiver. SMTP is based on pure ASCII text. The standards
describing mechanisms for transmitting text and other information are laid out in
RFC2822, RFC2045, etc. Well-known elements of the RFC2822 standard include the
destination address fields to, cc and bcc which specify the recipients of messages
[RFC2822]. Originator fields contain the fields from, sender and reply-to. Identification
fields contain a message identifier. Informational fields contain a subject and comments.
3.3. Resource Descriptions
The term electronic resource denotes a wide range of varying formats, such as (hyper)
text documents, images, audio or video files. In spite of the format-specific differences
between types of resources, a substantial part of their description can be dealt with in a
generic, type-independent form. Standardization efforts consequently aim at designing a
generic set of meta-data fields (Section 3.3.1) and standardized mark-up languages
(Section 3.3.2) that are used to describe resources independent of their types and formats.
3.3.1. General Fields
The Dublin Core Meta-data Initiative (DCMI) began in 1995 with a workshop in Dublin,
Ohio. Librarians, digital library researchers, content providers, and text mark-up experts
started to improve discovery standards for information resources. The original Dublin
Core (DC) emerged as a small set of descriptors drawing global interest from a variety of
information providers in arts, sciences, education, business, and government [ISO 2003].
DC is intended to co-exist with meta-data standards that offer other semantics. It is
expected that descriptive records will contain a mix of elements from various meta-data
standards. DC’s simplicity is both a strength and a weakness. Simplicity lowers the cost
of creating meta-data and promotes interoperability. But it does not accommodate the
semantic and functional richness supported by complex meta-data schemes. The use of
richer meta-data schemes in combination with DC is encouraged. Richer schemes can
also be mapped to DC for export or for cross-system searching. Conversely, DC records
can be used as a starting point for the creation of more complex descriptions [ISO 2003].
3.3.2. Mark-up Languages
Text mark-up languages are used to tag content and turn unstructured and thus non-
processable text into semi-structured text. Based on SGML, especially XHTML and
XML have gained wide recognition as standard languages for exchanging data on the
Web [XHTML, XML] or between application systems within and across organizations.
With respect to resource description, mark-up languages provide a form of meta-
language which can be used for partners exchanging resource descriptions to agree on
the format in which meta-data is serialized and interpreted. XML has long been widely

6
recognized as the most important base standard in this respect. Whereas Dublin Core
standardizes the contents of a generic set of meta-data fields, mark-up languages
standardize the format in which the standardized meta-data fields are serialized, in this
case into XML elements. Using XML, a number of institutions have developed standards
and started initiatives in order to provide comprehensive frameworks for definition and
exchange of meta-data. The most prominent one is W3C’s Semantic Web initiative.
3.4. Specific Standards
In addition to the generic standards, specific standards focus one or a limited number of
specific types of resources. Successful standardization initiatives have been working on
specific standards for texts, images and audio/video files. These are presented in the
following subsections along the lines of a general description of the specific resource
type and a brief description of widely known exemplary standards.
3.4.1. Texts
The advantage of text with respect to its treatment in solutions for semantic content
management is that its content is fully accessible to text processing, searching and
retrieving. In addition to meta-data on text, thus the content itself can be structured
according to guidelines and standards so that not only the text document in its entirety
can be accessed, but also individual parts of the document, e.g., a certain section,
paragraph or embedded resources, such as figures, tables, formulae etc.
Examples for encoding schemes that focus on text documents are the Text Encoding
Initiative’s [TEI] encoding scheme or the Open Document Format maintained by the
OASIS organization. The Open Document Format defines an XML schema for office
documents, including, but not limited to text documents, spreadsheets, charts and
graphical documents like drawings or presentations. Thus, text documents are not the
only object targeted by the Open Document standard. The schema aims at high-level
information suitable for editing and automatic processing of documents. The standard
comprises pre-defined meta-data elements as part of a separate XML element usually
located at the beginning of a document. Meta-data elements can be extended by users in a
specified format requiring name, data type and value (user-defined meta-data) as well as
in an unformatted way (custom meta-data). Pre-defined meta-data elements contain the
elements defined in the Dublin Core standard [Durusau et al. 2006].
3.4.2. Images
Finding specific images is much harder than finding text, unless additional, textual,
information is available. This additional information can be embedded in image files or
stored in separate files. For example, users can upload their images to a server and then
describe the contents of these images. Other users can then search these descriptions and
find the corresponding images. Nowadays, most digital cameras embed various pieces of

7
information into images they generate. This information includes current date/time,
shooting conditions, e.g., whether a flash was used, GPS coordinates and camera
settings, e.g., shutter, aperture. Adobe Photoshop, the leading image editing software, can
also embed pieces of meta-data into image files, such as author, title or copyright notice.
Examples of image files containing meta-data include Exif (exchangeable image file
format) and Tiff (tagged image file format). Exif is the format for camera-embedded
meta-data [EXIF]. Exif is a standard for storing interchange information in image files,
especially those using jpeg compression. Most digital cameras now use the Exif format.
Tiff is a wide-spread format for high color depth images, besides jpeg and png. The Tiff
format is supported by a wide range of applications [TIFF], e.g., Adobe Photoshop.
3.4.3. Audio and Video
Searching audio and video information is comparable to searching images. Without
additional textual information it is hardly possible to create, e.g., multimedia directories
or audio/video catalogues. Meta data specific to audio and video includes the original
medium, e.g., reel-to-reel, sampling rate and bit depth, digitization date or digitization
software used. Audio meta-data includes title, artist, track number, comment, year, genre
or album title. Video meta-data includes title, directors, actors, format, language or
studio.
Examples of formats include MPEG-7 and MP3. MPEG-7 is officially called
"Multimedia Content Description Interface", a means of attaching meta-data to
multimedia [Thom et al. 1998]. MPEG-1, -2, and -4 have been designed to represent
information itself. MPEG-7 is meant to represent information about information, i.e.,
MPEG-7 supports finding multimedia content. MPEG-7 is supposed to make the web
more searchable for multimedia content [MPEG-7]. Meta-data may be used for more
advanced access to underlying content, by enabling automatic multimedia presentation or
editing. MP3 stands for MPEG-1 Audio Layer-3 and is a standard technology and format
for the compression of sound sequences into small files without losing too much sound
quality [MP3]. Rippers and encoders can be used to convert songs from CDs to MP3
files. ID3 is the meta-data container that is most often used in conjunction with MP3. It
allows information such as title, artist or album to be embedded in the MP3 file.
4. Top-Level Ontology
We suggest to use a top-level ontology and to merge the vast and increasing number of
meta-data standards and ontologies under a single “roof”, the top-level ontology. We
imagine six dimensions to be used for that purpose, i.e., the dimensions when, what,
where, who, why and how. These dimensions have been identified as being essential for
effective categorization, visualization and navigation of collections of contents in a series
of unstructured interviews about and self-observation of ICT work environments of
knowledge workers [Maier/Sametinger 2002]. We have used the terms isolation, loss and
plurality in order to categorize problems faced with traditional ICT work environments

8
based on the desktop metaphor. Problems arise due to the facts that we have (1) a
hierarchical folder structure on our desktop rather than a flexible means of
categorization, (2) trivial and multiple categorization mechanisms in various applications,
e.g., Lotus Notes, Mozilla Firefox, MS Exchange, MS Internet Explorer, Palm Desktop,
(3) documented knowledge that is only accessible via specific applications, e.g., email
messages, appointments including notes, ideas and protocols or information about
people’s skills and expertise and their current availability, (4) versioning information
only available with specific applications, e.g., DMS, (5) multiple documents that are
different representations of the same contents, and (6) insufficient meta-data about local
and remote documents. Thus, we suggest mapping entries of meta-data standards
provided by a variety of applications and adhering to a variety of standards to the
identified six basic dimensions therefore easing retention, discovery and presentation of
documented knowledge.
Meta-data standard items map differently to our top-level ontology, see Fig. 1. Exif
items, for example, are mostly mapped to the how-dimension. This is not surprising
because when we take a picture with our digital camera, then most of the information that
is available about the taken image is technical information about its creation like shutter
speed o aperture opening. ID3 tags also provide a bulk of technical information.
Additionally, they contain information about persons like composer and interpret.
Fig. 1. Mapping of meta-data standard items to top-level ontology
There is one thing that catches our eye when looking at the suggested matching. It is
the fact that there are mappings for all dimensions except for one, i.e., the why-
dimension. What seems odd at first sight becomes obvious when giving it a second
thought. Information in this dimension cannot be created automatically but requires input
from a human being. For example, a picture may have been taken because there was the
wedding of a relative. The camera will never know this. We will have to add this extra
piece of information in order to keep it and make it accessible in the future. The subject
of an email message is an example for information that we map to the why-dimension.

9
The top-level ontology relates meta-data elements to the six dimensions. In the
following, a case example describes how these mappings can be used in order to respond
to rich user queries. A knowledge worker is engaged in an on-demand learning activity
and needs to find all relevant resources that could answer questions regarding the topic
“knowledge management”. The typical entry point in this case would be the what-
dimension. The knowledge worker would find events in the organization’s group
calendar (meta-data element summary), electronic text documents that contain
documented knowledge on the topic (title, keywords), audio and video presentations with
systematic introductions into the topic (title, genre=learning) as well as personal
information on co-workers whose job title has to do with knowledge management.
Browsing the material, the knowledge worker realizes that “John Brown” seems to be the
organization’s top expert on the topic, so she initiates another query using the who-
dimension in order to find out contact details, events on which she could personally meet
up with the expert as well as text, audio and video documents authored by or featuring
the expert. The results are presented in one or two dimensions. One single dimension can
be selected twice, e.g., a who/who-matrix that shows the relationships between our
knowledge worker and the expert. In the intersection cell, all resources are reported that
contain meta-data on “Sue Ash” AND “John Brown”, e.g., email messages, a document
created by John and modified by Sue (single resources are thus reported as often as
retrieved via multiple paths), a meeting in which both participated or friends and business
acquaintances that they have in common (using the friend-of-friend framework).
In an empirical study of the TOP 500 German companies conducted in 1999/2000,
improving visibility of knowledge and improving access to existing knowledge were the
two most important goals that organizations strive for when implementing KM initiatives
[Maier 2007, 472]. Our simple approach can be seen as targeting these two goals which
have also been proposed as starting points for KM life cyles [e.g., Probst et al. 1998]. In
addition, our approach positively influences the effort needed to improve retention of
knowledge in a variety of forms and formats and to improve acquisition of external
knowledge which is already externalized and available in a documented form. These are
another two goals found less important by organizations in the cited study, but which
have been on the rise since then with a recent stronger focus on the management of
knowledge risks, e.g., concerning intra-, inter- and extra-fluctuation of employees, as
well as on projects, processes and formal as well as informal knowledge cooperations
crossing organizational boundaries [Maier 2007, 136]. The challenge we target with
respect to these goals is that documented knowledge embedded in varying intra- and
inter-organizational contexts needs to be jointly retained and accessed while the effort to
prepare and refine the documented knowledge should be kept as low as possible.
5. Comparable work
Related work includes desktop alternatives like Lifestreams [Fertig 1996], Time-Machine
Computing [Rekimoto 1999] or Presto [Dourish 1999], and tools to display large

10
amounts of information, e.g., TheBrain (www.thebrain.com), Data Mountain [Robertson
1998], DLITE [Cousins 1997], and TopicShop [Amento 2000]. Currently, information
on our computers is still organized in old-fashioned files and folders, but searched by
specialized tools like Google Desktop Search [Google] or Copernic Desktop Search
[Copernic]. Operating systems also have features to better handle large amounts of data.
This is done by adding search capabilities like Apple’s Spotlight [Apple] or Windows
Vista’s Instant Search [Microsoft]. Tools like Google’s Picasa [Picasa] and Apple‘s
iTunes [iTunes] provide interesting approaches for organizing large amounts of data with
the drawback, that extra software has to be used which is not integrated into the user’s
desktop.
6. Conclusion
We argue that our computers would greatly benefit if their access mechanisms would be
extended to use a top-level ontology and to have dimensional views on documents and
electronic resources. In addition, the benefit would be much greater if all resources on
our computers would share such an access mechanism. This means that we do not need
any longer, for example, applications administrating images and songs or, in a
knowledge management perspective, lessons learned, learning objects, contributions to
newsgroups, Wikis and Weblogs, personal experiences and the like. Rather, we would
have a simple file access mechanism with additional small tools that provide the
functionality that applications now offer in addition to administering file access.
This paper has shown that existing meta-data standards can be mapped to a simple
six-dimensional top-level ontology which enables convenient integration between a
variety of heterogeneous types of resources and their specific meta-data standards. The
main advantage of this approach is its simplicity that does not require the time-
consuming mapping of sophisticated ontologies and the implementation of costly
ontology management middleware. Access to electronic resources fostered by the use of
a top-level ontology thus provides an easy mechanism to enhance the ICT work
environment of knowledge workers and thus the support for weakly structured, highly
complex, knowledge-intensive business processes with their unpredictable knowledge
requirements.
References
Amento, A., Terveen, L., Hill, W., Deborah Hix, D. (2000): TopicShop, in: Proceedings
of the 13th annual ACM symposium on User interface software and technology.
November 2000
Apple: Tiger Developer Overview Series: Working with Spotlight.
http://developer.apple.com/macosx/spotlight.html, last visited May 22, 2007
BernersLee, T., Hendler, J., Lassila, O. (2001): The Semantic Web. A new form of Web
content that is meaningful to computers will unleash a revolution of new
possibilities. In: Scientific American, May 17th, 2001, http://www.sciam.com/

11
Carr, L., Miles-Board, T., Woukeu, A., Wills, G., Hall, W. (2004): The Case for Explicit
Knowledge in Documents, in: Proceedings of the 2004 ACM Symposium on
Document Engineering, New York, 2004, 90-98
Copernic: Copernic Desktop Search, http://www.copernic.com/en/products/desktop-
search/index.html, last visited May 22, 2007
Cousins, S., Paepcke, A., Winograd, T., Bier, E. A., Pier, K. (1997): The Digital Library
Integrated Task Environment (DLITE), in: Proceedings of the 2nd ACM international
conference on Digital libraries DL 97, Philadelphia PA, USA, July 1997
Daconta, M. C., Obrst, L. J., Smith, K. T. (2003): The Semantic Web. A Guide to the
Future of XML, Web Services and Knowledge Management, Indianapolis (IN),
USA 2003
Dourish, P., W Edwards, W., K., LaMarca, A., Salisbury, M. (1999): Presto: an
experimental architecture for fluid interactive document spaces, in: ACM
Transactions on Computer-Human Interaction (TOCHI), Vol. 6, No. 2, June 1999
Durusau, P., Brauer, M., Oppermann, L. (2006): Open Document Format for Office
Applications (OpenDocument) v1.1, Committee Specification1, 19 Oct 2006,
http://www.oasis-open.org/committees/office, last visited March 12, 2007
EXIF: JEITA. Exchangeable image file format for digital still cameras: Exif Version 2.2,
http://www.exif.org/Exif2-2.PDF, April 2002. last visited May 22, 2007
Fertig, S., Freeman, E., Gelernter, D. (1996): Lifestreams: An Alternative to the Desktop
Metaphor, in: Proceedings of the CHI '96 conference companion on Human factors
in computing systems: common ground, Vancouver (BC), Canada, April 1996
FOAF. The Foaf Project. http://www.foaf-project.org/
Google: Google Desktop Search, http://desktop.google.com/, last visited May 22, 2007
Gruber, T. R. (1993): A Translation Approach to Portable Ontology Specifications, in:
Knowledge Acquisition, Vol. 5, No. 2, 1993, 199-220
ISO (2003): Information and documentation — The Dublin Core meta-data element set.
ISO Standard 15836-2003: http://www.niso.org/international/SC4/n515.pdf,
February 2003, last visited March 12, 2007
iTunes: Apple’s iTunes http://www.apple.com/itunes/overview/, last visited May 22,
2007
Maier, R. (2007): Knowledge Management Systems. Information and Communication
Technologies for Knowledge Management, 3rd edition, Berlin
Maier, R., Hädrich, T., Peinl, R. (2005): Enterprise Knowledge Infrastructures, Berlin
Maier, R., Peinl, R. (2005): Semantic Description of Documents in Enterprise
Knowledge Infrastructures, in: Jantke, K. P., Fähnrich, K-P., Wittig, W. S. (eds.):
Marktplatz Internet: Von e-Learning bis e-Payment. Proceedings of the 13th
Leipziger Informatik-Tage, LIT 2005, September, 21-23, 2005, Leipzig, 358-366
Maier, R., Sametinger, J. (2002): Infotop - An Information and Communication
Infrastructure for Knowledge Work, ECKM2002, The 3rd European Conference on
Knowledge Management, Trinity College Dublin, Ireland, September 24-25, 2002
MP3: MP3: ISO/IEC JTC1/SC29/WG11 N, MPEG-1 - Coding of moving pictures and
associated audio for digital storage media at up to about 1,5 Mbit/s, June 1996.
http://www.chiariglione.org/mpeg/standards/mpeg-1/mpeg-1.htm. last visited May
22, 2007

12
MPEG-7: ISO/IEC JTC1/SC29/WG11N6828, MPEG-7 Overview,
http://www.chiariglione.org/mpeg/standards/mpeg-7/mpeg-7.htm.. last visited May
22, 2007
Nonaka, I., Takeuchi, H. (1995): The Knowledge Creating Company. New York
Picasa -Google's photo organizer, Picasa, http://picasa.google.com/, last visited May 22,
2007
Probst, G., Raub, S., Romhardt, K. (1998): Wissen managen: Wie Unternehmen ihre
wertvollste Ressource optimal nutzen, 2nd edition, Wiesbaden
Rekimoto, J. (1999): Time-machine computing: a time-centric approach for the
information environment, in: Proceedings of the 12th annual ACM symposium on
User interface software and technology, Asheville, North Carolina
RFC2822 (2001): Internet Message Format, Network Working Group, Request for
Comments: 2822, http://www.ietf.org/rfc/rfc2822.txt, last visited May 22, 2007
Robertson, G., Czerwinski, M., Larson, K., Robbins, D. C., Thiel, D., van Dantzich, M.
(1998): Data mountain, in: Proceedings of the 11th annual ACM symposium on User
interface software and technology, November 1998
Schimkat, R.-D. (2003): Techniken und Aspekte zur Realisierung proaktiver
Informationssysteme. Osnabrück
Swetland A. G., Baca M. (Eds.) (2000): Introduction to Meta-data: Pathways to Digital
Information, Getty Information Institute
TEI. Text Encoding Initiative (2007), http://www.tei-c.org/ last visited March 12, 2007
Thom D., Purnhagen H., and the MPEG Audio Subgroup (1998): MPEG Audio FAQ
Version 9 MPEG-7, October 1998; see http://www.chiariglione.org/mpeg/faq/mp7-
aud/mp7-aud.htm, last visited May 9, 2005
TIFF: Adobe. TIFF 6.0 Specification,
http://partners.adobe.com/public/developer/en/tiff/TIFF6.pdf, last visited May 22,
2007
vCalendar: http://www.imc.org/pdi/vcaloverview.html, last visited December 5, 2005
vCard: Version 2.1, http://www.imc.org/pdi/vcard-21.rtf, last visited December 5, 2005
Wolff, E. N. (2005): The Growth of Information Workers, in: Communications of the
ACM, Vol. 48, No. 10, 37-42
Microsoft: Features Explained: Instant Search,
http://www.microsoft.com/windows/products/windowsvista/features/details/instants
earch.mspx, last visited May 22, 2007
XHTML: XHTML 1.0, The Extensible HyperText Markup Language, 2nd edition, A
Reformulation of HTML 4 in XML 1.0, W3C Recommendation, 26 January 2000,
revised 1 August 2002, last visited May 22, 2007
XML: Extensible Markup Language (XML) 1.0, 4th edition, W3C Recommendation, 16
August 2006, edited in place 29 September 2006, last visited May 22, 2007