Various new imaging techniques have been used at the C2RMF, in order to digitize paintings and objects. These techniques are used for extensive and detailed non-destructive analysis of works of art in different ways and include multispectral imaging, panoramic viewing of objects, 3D laser scanning of painting surfaces and 3D photogrametric modeling of objects. These techniques are valuable tools in the restoration and conservation arsenal, allowing us to perform regular and accurate monitoring of works of art, to measure their conservation state and compare these with previous analyses. These techniques, however, result in huge quantities of data, which needs to be visualized and disseminated in several ways. The resulting images must also be accessible to various partners around the world via the Internet, but the data needs to be protected and the vast data sizes involved require careful handling and management. This paper describes several new developments that have been made at the C2RMF by our team in order to make extremely high resolution images available on line. We will examine as a case study the use, the manipulation and the share of high resolution colorimetric images between members of the museum research community. The system that has been developed is a light-weight client-server architecture that efficiently streams image data to the client, allowing the user to quickly view very large images even over a slow Internet connection. These developments have been released as Open Source software in the IIPImage project.

This submission describes the very first step in the long-term project of building a national database of digitized cultural and scientific heritage. The main idea is to create an object-oriented set of metadata for describing any kind of digital object and to define translations to other metadata standards.

: The necessity of a subject-machine-object interaction (SMOI) reference model and its application in electronic publishing of the objects of our culture is considered. The components of such a model are listed. The proposed relatively full Reference Model of the SMOI components allows designers to divide the interaction task into subtasks more easily. It allows an easier choice of standardisation fields and conversion of an Object's models description and presentation.

Indexation of documents is a very determinant step for the information retrieval process and it is often a manual task which highly depends on the indexer knowledge. We propose to improve the manual indexation of documents thanks to a semi automatic semantic annotation process.

Among the new possibilities that the Semantic Web has enabled, the authoring task is considered as a key moment for semantic representation of knowledge to enhance publishing needs. New systems are emerging, delivering annotation and authoring issues based on ontologies. If we consider Semantic Web technologies as of today, what would be the skills needed from authors in order to produce documents compliant with the W3C recommendations and authoring environments that make full use of these possibilities? In this paper we present an assessment of ontology-based authoring tools, based on a specific criteria we developed to identify systems features and compliance with current recommendations. We have learned that, according to these systems, authors are being asked to perform additional knowledge structuring task that he/she is not familiar with. There is an eminent need in tackling these questions in the knowledge representation world, especially if we seek to give support to authors in developing more structured and well represented content knowledge, based on the potential that currently developed ontology technologies and solutions have. To support this need, we have proposed some requirements and a model of architecture of ontology-based authoring environment for the Semantic Web, composed of modules that integrate current technologies, making full use of existing features and proposing the development of new trends.

This paper introduces OAI static repository technology as a method by which publishers can expose their article-level metadata and thereby participate more fully in the wider information environment. It begins by discussing the value of exposing metadata via OAI and the potential role that static repositories might play in lowering the barriers to achieving this, and reports on the progress of a study which is evaluating the applicability and effectiveness of the static repositories approach.

The Information Retrieval System "Manuscript" that is being developed at the Laboratory of Computer-Aided Philological Research, Udmurtia State University is intended for storing, editing and processing electronic copies of manuscripts. By retaining all the peculiarities of the ancient treasures, the Manuscript system provides a thorough input of texts/manuscripts under study while preserving the integrity of the original electronic copy of the manuscript, text transcription, and transliteration for the purpose of creating reference materials or electronic and printed publications of the original manuscripts. The Manuscript system provides a correct input of texts/manuscripts under study by retaining all essential peculiarities of the manuscripts. Computer-aided processing of manuscripts is done taking into account formal and implicit properties of texts and their fragments. The system allows creating text transcriptions and transliterations, preparing dictionaries, lists of words and other units of interest (syntagmas, fragments) for digital and printed editions. The manuscript digital copy exists as a typed electronic text. A specialized text editor "OldEd" was created to ensure data input and correction directly in the database. It helps the user to work effectively with visualized parts of manuscripts, relationships, their properties and values. The editor represents the edited text in the form of geometrical, linguistic, functional and others hierarchies. In the mode of operation over the text, a text can be edited and divided into fragments. In the mode of viewing the hierarchies, the editor allows creating new units (including texts), viewing, adding and deleting relationships between text and dictionary units and modifying their properties (see http://io.udsu.ru/pub/rd/ for more detail).

Firenze University Press was born in 2003 with the mission to provide dissemination of the intellectual output of the University of Florence, Italy. Its strategy aims to enhance visibility of editorial production and to attract authors by a high-quality professional support coupled with low-cost production and the opportunity to keep copyright. Firenze University Press has also a role in innovating publishing models in the Italian academic environment and market. In the next two years, its efforts will concentrate on adopting recent scholarly communication paradigms coupled with advanced technologies, in order to offer better services to the academic community. The adoption of the Open Access model for electronic versions of published works is sustained through sales of paper copies. The implementation of an integrated system for the management of all editorial processes will allow economies of scale and a more efficient use of resources. The platform is based on a modular architecture and makes use of available open-source software, developed in advanced contexts and spread in large communities all over the world; details of the system are described in the paper. These choices enhance sustainability and allow extended customization opportunities. The exploitation of technologies, linked to the use of XML and recent protocols for document treatment and preservation and for metadata exchange, will keep Firenze University Press abreast of the most advanced trends in electronic publishing. This project provides an example of sustainable solutions for up-to-date small publishers and university presses.

In this paper, we report on our experience with the creation of an automated, human-assisted process to extract metadata from documents in a large (>100,000), dynamically growing collection. Such a collection may be expected to be heterogeneous, both statically heterogeneous (containing documents in a variety of formats) and dynamically heterogeneous (likely to acquire new documents in formats unlike any prior acquisitions). Eventually, we hope to be able to totally automate metadata extraction for 80% of the documents and reduce the time needed to generate the metadata for the remaining documents also by 80%. In this paper, we describe our process of first classifying documents into equivalence classes for which we can then use a rule-based approach to extract metadata. Our rule-based approach differs from others in as far as it separates the rule-interpreting engine from a template of rules. The templates vary among classes but the engine is the same. We have evaluated our approach on a test bed of 7413 randomly selected documents from the DTIC (Defense Technical Information Center) collection with encouraging results. Finally, we describe how we can use this process to generate an OAI (Open Archive Initiatives) - compliant digital library from a stream of incoming documents.

By the end of this year visitors to the Flemish town of Mechelen can discover the city with the help of the new CityInMyPocket walking guide. Instead of following a person or a book, people can pick up a CityInMyPocket digital walking guide and go sightseeing at their own speed. So, leave your heavy guide books and many flyers at home. CityInMyPocket will tell you just as much, and even more. CityInMyPocket is like having a friend who lives in the neighbourhood showing you around. The goals are formulated, but how did we manage? In order to succeed, we started investigating how people usually visit a city. Are people just strolling around the city or do they strictly follow the walk mentioned in their walking guide? What are the information needs during their stay? In parallel with the user analysis, we looked at similar projects and applications to see how they approached the problem. Based on the results of the user needs and the tests of existing applications, we defined the concepts of the CityInMyPocket walking guide. Here, we can distinguish three main building blocks: the map application, the search engine and the route planner. When the overall concept was ready and the information needs where specified, we started building the information model based on the Topic Map standard. For the map information we explored the power of the SVG standard. The moment the first data set was prepared, we implemented the map application. The remaining concepts were built around the central map application. This paper describes in more detail how we managed to build the CityInMyPocket walking guide.

The JISC Information Environment Service Registry (IESR) contains descriptions of collections of resources available to researchers, learners and teachers in the UK, along with technical service access details. This paper describes the data model and metadata description schema of IESR, and the services IESR provides to disseminate its records. There is a particular focus on the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) interface, including a possible use scenario. IESR's position within a wider information environment, its relationship to other initiatives, and future possible service registry directions are discussed.

The recent journal price increases and state-wide budget cuts have forced librarians at the University of Maryland (UMD) to make outright cancellations to journals. The implementation of in-house system for preserving scholarly work of faculty was necessary not only because of budgetary stringency. The ongoing “scholarly communication crisis”, the recognition that the library system needs to be changed, the need for archiving e-journal content has prompted the construction of Institutional Repository. This poster aims to identify issues encountered during the design process of a digital repository and suggests possible solutions. New roles of the users of this newly implemented system are described. The main elements of the service are outlined, outreach methodologies are discussed, and faculty concerns are addressed.

The article considers the library cooperation problem and some aspects of the relationships libraries coalition - library - end users. The authors describe the successful model of the first libraries coalition in Bulgaria - Bulgarian Information Consortium. Problems of creation, management practice, activities and achievements are presented and discussed. The role of electronic resources is viewed and analyzed from the perspective of library collaboration. The usage of these resources is shown as a powerful driving factor for libraries development and cooperation. It allows them to overcome behavioral, organizational, and economic problems and to start their real modernization. The libraries cooperation in Bulgaria begins just by shared electronic publications usage and continues mainly on this basis. We do believe that this is a one direction way and there will be no return journey for Bulgarian Librarian society.

Some aspects of digitizing old printed maps and suitable forms of storing are pointed to discussion. Questions arising about practical use of a resolution level, which is necessary for a proper identification of names on maps, and how to identify them, are considered. A case study is presented.

Originally established to merely publish electronic theses and dissertations (ETDs) as well as postdoctoral theses, the edoc server at Humboldt University Berlin (http://edoc.hu-berlin.de/) has become the service platform for scholarly publications of all types. It is an integral part of the university's information infrastructure which also contains a teaching management system, a virtual library portal, and a media portal. During the past years the edoc server has become the central place for electronic publishing within the university. On the basis of a hosting concept, the Electronic Publishing Group which is in charge of the edoc server also provides its publishing services for external authors, editors and institutions likewise. And being a DINI certified document and publication server, the technological components and the organisational workflows developed for the edoc server exemplify other scholarly publishing servers. As a joint group formed by the Computer and Media Service and the University Library of Humboldt University the Electronic Publishing Group deals with all issues associated with scholarly electronic publishing, including technical, library, and legal aspects. Combining the development of publishing tools and services with its routine operation the group consists of permanent and project funded staff. This paper describes the edoc server at Humboldt University and details on both the technical and non-technical parts of the entire publishing service.

This paper describes the automated process to create structured master and access copies for the digitised works at the BND - National Digital Library. The BND created during 2004 and 2005 nearly half million of digitised images, from more than 25.000 titles of printed works, manuscripts, drawings and maps. The resulting of the digitisation process is a group of TIFF image files representing the surfaces of the original works, which needs yet to be processed in order to be stored and published. Doing that manually would be a very complex and expensive task, with risks for the uniformity of the results, so it was need to develop an automated solution. To create the technical metadata, apply image processing actions and OCR, create derived copies for access in PNG, JPG, GIF, and PDF, we developed a tool named SECO. To create the master copies for each of those works, for preservation, and access copies in HTML, we developed a tool names CONTENTE, which exists as a standalone tool and as a library. Finally the copies are deposited and registered at the BND repository through the service PURL.PT, which assures also the WEB and intranet access control. This complex process is fully automated through several XML schemas for the control of the processes, description of the results (including the OCR outputs), descriptive metadata (in Dublin Core, MARC XML, etc.) and rights and structural metadata (in METS).

This paper concerns the document multi-structuring issue. For various use objectives, many distinct structures may be defined simultaneously for the same original document. For example, a document may have a first structure for logical content organisation (logical structure), and a second structure to express a set of content formatting rules (physical structure). We have already proposed a generic model, called MSDM, for the multi-structured documents, in which several important features were established. In this paper, we address the encoding problem of this kind of documents. We present a new formalism, called MultiX, which allows encoding the multi-structured documents efficiently. This formalism is based on the MSDM model and uses XML syntax.

Along with the long-term preservation of digital publications the next important goal is public access and in this regard user-centred design of digital repositories. Several repositories worldwide have shown that users are their most critical element. Repositories as such are valuable only if used. The success of the repository is often influenced by the content and the design of the interface. The lack of content is the main reason for unsuccessful repository as well as design-centred interface. Many digital repositories' interfaces are design-centred rather than user-centred. Quite a few users claimed that the search interface was too complicated and distracting and that specific jargon, wording and explanation within the certain interface did not help them at all. This demonstration presents the results of end-user survey carried out in the reUSE project, in which we examined the usability of three different digital repositories from Austria, Estonia and Germany. The reUSE is a cooperative project from eContent scheme and involves national and university libraries and universities from Austria, Estonia, Germany and Slovenia. Evaluation of the digital repositories was one of the main goals. Overall aim of the evaluation was to make user-centred repositories which will be at the same time most efficient in technical and organizational regards.

The European Accessible Information Network (EUAIN) is currently examining issues relating to accessible content processing. With the support of key European publishers, it has been possible to begin to identify key trends in accessible content processing which are likely to be of some importance in the coming years. This paper also describes some related standards activities and identifies a need for accessibility to be embedded within content creation and production processes at the earliest stages.

Collection management policies of libraries were mainly shaped by in-house use of materials in the past. Yet the emergence of electronic journals and their availability through publishers' or aggregators' web sites is changing this practice considerably. Libraries are no longer limited with the “one source - one user” model. This paper investigates the potential impact of the availability of electronic journals through the web sites of publishers or library consortia on document delivery services. It reviews the relevant literature first and reports the findings of the study with regards to the provision of articles through electronic journals and its impact on the Turkish national document delivery service. It then compares the number of articles supplied through the service with that downloaded from publishers' web sites by the users of the consortium of Turkish academic libraries. Following questions are addressed: Is document “delivery” using traditional or electronic means becoming a withering practice in libraries? Would document delivery services exist as we know them today in the age of electronic journals, big deals, and library consortia? The paper discusses the implications of electronic journals available through big deals on national document delivery services along with some conclusions.

The growing momentum of the world wide open access movement has splurged a parallel trend of development of open source and low cost electronic publishing systems and digital repository services. These growth and development strongly reflect the desire of the scholarly communities to take back control of aspects of scholarly communication and to improve processes for the dissemination and impact of their research. The development of journal publication systems makes it easier for university presses, libraries, societies, and individual researchers to launch and run open access journals. Researchers are now less dependant on the decisions of large publishers as to which journals should be published (decisions that are often made on financial, rather than academic, grounds), how the journals should be priced, access policies, etc. Digital publishing systems also allow scholarly communities to experiment with innovative formats, to contribute to sharing and re-using digital research data, and overcome the limits of traditional publishing.

The paper presents in brief a methodology for development of tools for knowledge-based search in repositories of digitized manuscripts. It is designated to assist the search activities in collections that may enlist XML documents which should be catalogue descriptions or marked-up full texts of mediaeval manuscripts. The suggested methodology is directed to the development of software environments that will be able to deal with relatively complex user queries containing words or phrases that are considered as domain concepts. The emphasis in this methodology falls on two main types of activities: development of proper ontologies describing the conceptual knowledge relevant to the chosen domain(s) and development of proper intelligent agents for search and processing purposes that are able to retrieve and filter documents by their semantic properties. Some considerations related to the implementation of our methodology are presented as well. The first version of a software tool for knowledge-based search in repositories of digitized manuscripts is discussed. Some results of the application of the tool to a collection of approximately 800 descriptions of mediaeval Bulgarian manuscripts stored in Bulgaria are analyzed.

This contribution presents the concept and analyses the path of diffusion of an innovative publishing idea that originated in one speciality in physics and is now about to spread into other fields, including the social sciences. We discuss the conditions of success and fostering as well as hampering factors on the road for what we consider to be a truly revolutionary concept of presenting the state of knowledge in potentially any given academic discipline. First, we will present the overall idea of Living Reviews (www.livingreviews.org), which are open access online journals featuring an innovative editorial concept for the publication of high-quality scientific content. Second, we discuss the proliferation of this unique concept. Finally, we shall aim at summarising lessons learned from our shared experience in establishing these publishing ventures in our respective fields. Our conclusions give hints for potential future “cyber-activists” who are considering setting up further Living Review projects.

In this paper we present the e-publishing platform created by and for the Living Reviews project. In particular, we give an overview of the Living Reviews project, concept and publishing model, compare our project to related projects, and present the technical infrastructure used to run the "Living Reviews" family of online journals. We will give an overview of the publishing environment, point out the challenges inherent in distributed editorial offices, and describe the interaction between various user groups and the infrastructure. Based on this description of the environment, we will explain the constituents of the infrastructure and how they evolved. Following the history and detailed description of the constituents we give an overview of further plans within our project and applicability of our particular solutions to other publishing projects. Again on this level of detail, we will compare our solutions to related efforts. We will conclude the paper listing lessons learned from our project. In particular, we will describe the challenge of reconciling the technical and the non-technical divisions of the publishing project and how it was met for the Living Reviews family. While mainly providing a case study, we also hope to point out solutions to publishing problems which may be of general interest, and more importantly make a case for necessary standardization of e-publishing software components. We don't see much scientific value in this venture, but value for the scientific community since we present another epublishing platform to use, provide another set of tools to choose from, give another best-practices description when rolling your own.

When content producers are forced to generate and manage content day after day (and, most of the time, in a hurry), as in the case of TV broadcasters, they have different priorities and requirements than other kinds of content providers. However, the requirement for content interchange through optimum content search is always there. The paper presents how the issue of interoperability of metadata and digital rights information is solved in a specific project, the XAC (Xarxa IP Audiovisual de Catalunya, Audiovisual IP Network of Catalonia), where producers of TV programmes are the main involved content owners. After an analysis of how these content providers manage metadata and rights information, a specific proposal fulfilling providers' requirements and International Standards, the XAC Metadata Set (XAC-MS), is described. This is the basis for implementing a system for multimedia content interchange that could be extended to other kinds of multimedia publishing.

In the last few years, the University of Aveiro, Portugal, has been offering several distance learning courses over the Web, using e learning platforms. Experience showed that different editions of a same course, using the same contents and structure, and having similar target learners, had different success rates. What would be the reason for that? A hypothesis was considered: The level of success could be directly related with the remote follow up of the learners' participation in the courses; the best results usually occur when the follow-up is closer. The existing e-learning platforms offer and the standardization works being developed by organizations and consortiums like IMS (IMS Global Learning Consortium, Inc), ADL SCORM (Advanced Distributed Learning Sherable Content Object Reference Model), IEEE LTSC LOM (Institute of Electrical and Electronic Engineers Learning Technologies Standard Committee Learning Object Metadata), ARIADNE (ARIADNE Foundation for the European Knowledge Pool), AICC CMI (Aviation Industry CBT Committee Computer Managed Instruction), etc, don't cover the course monitorization concerns mentioned. Those projects were focused on aspects like contents and its delivery in the context of the execution of the courses' activities. This is even true in the SCORM project that doesn't include any reference to the management of the e-learning processes. Recently, in the context of the IMS Global Consortium, a new project designated IMS LD (Learning Design) is under development, providing a framework for the description of learning units under a three level model. In the most recently defined level, the C level, some functionalities related to notifications were proposed, expressing similar concerns to the ones that triggered our research. However, the extent at which IMS LD takes the functionalities is, from our point of view, not complete. This article describes a proposal of a reference model and functionalities towards a specification of a layer for real-time management of user interactions on LMSs, and its possible integration with the ADL SCORM standard proposal. The paper includes a discussion of the management metadata model for the LMS sub system and how the integration of the management module under SCORM may be achieved.

NARCIS, National Academic Research and Collaborations Information System, is a project in the Netherlands to build a portal for research information which combines structured research information with information from OAI repositories (publication and other scientific results), websites, and news pages of research institutes. The main goal of NARCIS is to create a central place for searching all these types of data. The research data in NARCIS have been collected via the administrative processes of the different participating institutes within the work flow process. Different techniques are being used to combine the current research information and the research results. The idea for the project has been developed within the DIO-platform (National Platform Data Infrastructure Research Information) and has been realised with a subsidy of the Dutch programme DARE (Digital Academic REpositories) coordinated by SURF. SURF is the higher education and research partnership organisation for network services and information and communications technology (ICT). Partners in the NARCIS project are The Royal Netherlands Academy of Arts and Sciences (KNAW), Netherlands Organisation for Scientific Research (NWO), the Association of Universities in the Netherlands (VSNU), and the Information Centre of the Radboud University of Nijmegen (METIS). The implementing of NARCIS took one year.

The article explores Open Access in the context of recent contributions in communication and public relations theory, notably Bakhtinian dialogism. Increasingly since the 1960s communication and public relations theory have shifted their focus away from one-way, asymmetrical to two-way, symmetrical models, thereby progressing to more equitable and interactive, multi-level conceptual models of dialogic and polyphonic communication. It is in this context that Bakhtinian dialogism becomes highly relevant. The influence exerted today by the Russian literary theoretician Mikhail M.Bakhtin (1895-1975) extends far beyond literary theory into fields such as philosophy of language, linguistics, musicology, anthropology, culture studies, classic studies, history, political science and theology, communication and public relations theory, including new media. Open Access is interpreted by using the Bakhtinian concepts of dialogue, polyphony and carnival, which in combination view Open Access as a healthy and sustainable, consumer-led answer to the serials crisis and a revolt against the system that has produced the serials crisis, viz. hegemonic power structures and relations of traditional, subscription-based and commercially motivated scholarly publishing.

The paper gives an overview of recent developments in Sweden concerning the communication and reception of the Open Access concept and the growth and co-ordination of digital academic repositories. The Open Access concept was primarily introduced in Sweden from library circles which also coloured the arguments used. University leaders started to show an interest from 2002 and their association, the SUHF, issued a report favorable to Open Access in 2003. In the consecutive years both the SUHF and the Swedish Research Council signed the Berlin declaration. The relatively swift change of policy towards a support of Open Access by these two important stakeholders was obviously influenced by the international discussion but also by the way the issue was raised in Sweden and by the absence of a national publisher lobby arguing against Open Access. The development of e-publishing within Swedish higher education started to gather momentum in the years following 2000. Universities either chose to particicipate in the DiVA consortium or to implement available Open Source software. In 2003 the national SVEP project was launched to co-ordinate and scale up the development of e-publishing within higher education. Recommendations were issued for metadata descriptions of e-publications, for publication databases and for subject categories. The SUHF gave support to the metadata recommendations for publication databases (local registers of academic publications) which clears the way for a co-ordination between publication databases and freely available full-text material in repositories. Important building blocks of a generalized archiving workflow between a local repository and a national archive were implemented. A resolution service for the use of permanent identifiers and a service for administrating name spaces in URN:NBN were established. A national search service for undergraduate theses was developped and put into operation at the LIBRIS website. The service uses the OAI-PMH to harvest local repositories who have to implement a metada model based on simple Dublin Core. A new development programme to support Open Access in Sweden is now being initiated. It will engage the major stakeholders in Sweden and will have a clear focus on promoting the growth of the volume and diversity of content in academic repositories.

Open access publishing strategies have traditionally been directed towards what has been regarded as a homogenous scientific community of universities, researchers and libraries. However, discipline specific practices in communication and publishing strategies are prevailing in different scientific areas. In this study we, argue that discipline specific publishing patterns may affect the ways that open access strategies can be adopted in different scientific areas. We characterise and identify incentives for publishing open access into factors depending mostly on the social environment and factors mostly depending on personal factors of the researcher. In the case study comparing the field of biomedicine and economics and business administration we were able to find out figures on the proportion, type and channel of open access publishing of scientific articles by Finnish researchers in economics and medicine.

The BRICKS EU Project (http://www.brickscommunity.org) is constructing a distributed, scalable, and safe infrastructure that will provide open-source technology for the European Digital Memory. This infrastructure, known as the “Foundation”, has been implemented through the integration of independent software units called “Bricks”, which are developed on the architecture nodes. A Brick is an independent building block whose functionality is made available through a formally defined interface. Bricks can be put together to provide richer functionality, and may encapsulate Content. In order to test and validate the Foundation infrastructure and demonstrate how to build value-added services overlapping such infrastructure, some significant added-value “Pillar” applications have been developed and made available, addressing the “Greek temple metaphor”. The intention is to use such applications for validating the Foundation services and, at the same time, those applications will constitute the basis of the future advancement of the BRICKS sustainability concept and be an attractive feature for the creation of a future Cultural Heritage Community. The “Fortuna Visiva of Pompeii” Pillar application has been developed as a case study within the “Archaeological site” Pillar. By using most of the Bricks created within the overall framework, demonstrates how the final users will be provided with services designed to enable them to use Pompeii's Archaeological Digital Library and Geographic Information System for learning, research and cultural tourism. Final users have been largely involved from the very beginning in the design phase of the case study applications, ensuring that such application matched the initial requirements (“vertical integration”), thus contributing to bridge the gap between users' community and advanced ICTs in the Cultural Heritage domain.

Dictionaries and Thesauri are valuable resources for Natural Language Processing but are not as widely available as one would hope, especially for languages other than English, and most can only be used for querying online. Our main goal with T2O - Thesaurus to Ontology framework - is to create a multilingual ontology: -- freely available online and for downloading; with a computer-readable format; with a good API; with a structure as rich as possible; reusing all the structured information we can get.

Electronic publishing exploits numerous possibilities to present or exchange information and to communicate via most current media like the Internet. By utilizing modern Web technologies like Web Services, loosely coupled services, and peer-to-peer networks we describe the integration of an intelligent business news presentation and distribution network. Employing semantics technologies enables the coupling of multinational and multilingual business news data on a scaleable international level and thus introduce a service quality that is not achieved by alternative technologies in the news distribution area so far. Architecturally, we identified the loosely coupling of existing services as the most feasible way to address multinational and multilingual news presentation and distribution networks. Furthermore we semantically enrich multinational news contents by relating them using AI techniques like the Vector Space Model. Summarizing our experiences we describe the technical integration of semantics and communication technologies in order to create a modern international news network.

Multilingual aspects are gaining more attention in recent years. This direction is further broadened by a global integration of the European states and vanishing cultural and social boundaries. The spread of foreign languages is even bigger with the information boom caused by an emergence of easy internet access. Multilingual text processing becomes an important area, which brings a lot of new and interesting problems. Their possible solutions are proposed in this paper. The first part of this contribution is devoted to methods for multilingual searching, the second part deals with summarization of retrieved texts. We tested some novel processing techniques: a language-independent storage format, semantic-based indexing, query expansion or text summarization leading to faster and easier retrieval and understanding of documents. We implemented a prototype system named MUSE (Multilingual Searching and Extraction) and evaluated its qualities against the state-of-the-art searching engine - Google. The results seem to be promising; MUSE shows high correlation with market-leading products. Although our experiments were performed on Czech and English articles, the main principle remains the same for other languages.

In ATN sentences and its constituents are generalized. Special signs are introduced to define categories of words and their parameters. ATN is expedient for the frame analysis system, the result of which can be transited as frame transformations based on both cause-and-consequence relations and the fields of concepts. Frame system can be presented as a tree on the top of which the basic data accumulated with the sub-frames.

The topics addressed in this paper are threefold: First, techniques for the semiautomatic normalization of image descriptors in a digital image collection from free text titles and keywords. Second, the efficient construction of thesauri for specific image collections. And third, the optimisation of search mechanisms to deal with the special characteristics of the image collections and with the use made by users through web-based search interfaces. The solutions presented here have been developed in the framework of a commercial project intended to improve image search in a website for selling photographs through the web. The ultimate goal of this project is to improve the customer accessibility to a collection of more than two million photographs.

The internet causes a continuous emergence of novel forms of scholarly communication and collaboration. Electronic publishing provides a means for representing eventual outcomes of these processes, i.e. all types of content such as papers and advanced forms of media. Electronic journals are often chosen as an adequate publishing format because they simultaneously deliver content in a well-known manner but, at the same time, allow extending traditional publishing with innovative features. The initiative Digital Peer Publishing (DiPP) provides technological, organizational and legal frameworks and tools that help to incubate and proliferate such innovative publishing projects. The hosting platform reflects principles of a Service Oriented Architecture. It combines, via Web Services, already established components such as an OAI repository (Fedora) and a Web Content Management System (Plone) with customized workflows for document processing, conversion and distribution. As an open platform it is capable of integrating external tools and services or acts itself as a service provider. It is therefore disposed for supplementing research infrastructures with electronic publishing.

This paper deals with the problems of weakening of human memory due to the increasing use of machine memories in many ways. It enumerates the advantages of machine memory as stated by computer people as opposed to human memory. The paper then briefly discusses the reliance on human memory in such activities as oral history and oral archive, enumerates some institutions working on Iranian history based on oral history or archive, and then presents some suggestions.

The EU TEL-ME-MOR PROJECT integrates the ten New Member States' national libraries into The European Library (TEL - http://telmemor.net). Their resources - electronic catalogues and digital libraries - are being incorporated into the portal via OAI-PMH or Z39:50 protocols. A special workpackage surveys and reports on their research and development activities as well as the underlying technologies and digital data useful for TEL. Recently, the Conference of European National Librarians (CENL) asked for extension of the survey to include all the European national libraries. The poster presents the most interesting results and findings.

This paper presents partial results of a study that reflects on ten years of the annual International Conference on Electronic Publishing – ElPub. The rich content available in the proceedings of the nine previous conferences, plus the material available through ‘pre-10th conference’ provided the data from which a number of aspects were looked at. This includes data on authors, papers, countries, institutions, keynotes, themes, conference organisation, conference websites, conference organisers (the so-called ElPubers!) and so forth. The analysis was based on both qualitative and quantitative (mostly) approaches. Results were, therefore, obtained through a content analysis entirely based on the proceedings (1st to 9th ElPub) and Scix’s database (10th ElPub), using Excel. Several questions have been asked on the quantitative data gathered, in order to provide a well-built picture of the conference. These questions relate to such issues as most productive authors, most actively participating countries, most influential areas of knowledge, most representative type of institutions (academic, governmental or entrepreneurial), most approached topics and so forth. In terms of qualitative data, the study looks at conference and session themes, the variety in the type of sessions, the environment within which topics presented are set in (academic, industrial, governmental, entrepreneurial) as well as its concerned focus, the field of knowledge from which keynotes and organising chairs have come from along with the focus of their speech, the influence of the host country in the number of local authors contributing to the conference at hand and the following ones, the major pattern of authorship (individual or collaborative), the composition of the Programme Committee in terms of country or continent represented (and perhaps the obvious greater presence of natives in it), field of knowledge, seniority and gender. Major conclusions will be presented at the conference. Partial results are provided.

A text parsing component designed to be part of a system that assists students in academic reading an writing is presented. The parser can automatically add a relational discourse structure annotation to a scientific article that a user wants to explore. The discourse structure employed is defined in an XML format and is based the Rhetorical Structure Theory. The architecture of the parser comprises preprocessing components which provide an input text with XML annotations on different linguistic and structural layers. In the first version these are syntactic tagging, lexical discourse marker tagging, logical document structure, and segmentation into elementary discourse segments. The algorithm is based on the shift-reduce parser by Marcu (2000) and is controlled by reduce operations that are constrained by linguistic conditions derived from an XML-encoded discourse marker lexicon. The constraints are formulated over multiple annotation layers of the same text.

A snapshot of the British blogosphere focusing on the blogging habits and technical abilities of a sample of British bloggers. Most of the academic research that has so far been published on blogging behaviour has been based on US bloggers. By investigating British bloggers, who are relatively late in the field as a group, we hope to clarify the role of gender, as opposed to early adoption or subject matter, in the popularity of blogs. Analysis of the blogs has so far focused on technical information such as technical ability of the blogger, which software used, analysis of the blogroll, enhancements to the blog and use of images This sample suggests that male and female bloggers in the UK approach the construction of their blogs differently. However, the finding that women tend to have more outward links means that the facile 'men-filter' and 'women-journal' distinction should not be made.

The Text Encoding Initiative is an international and interdisciplinary standards project established in 1987 to develop, maintain and promulgate hardware- and software-independent methods for encoding humanities data in electronic form. Initially the TEI was jointly sponsored by three established international professional associations (the Association for Computers and the Humanities, the Association for Computational Linguistics and the Association for Literary and Linguistic Computing), which established a small management committee, and appointed two editors to co-ordinate the enthusiastic participation of more than a hundred scholars worldwide. Its remit was to attempt a complete definition of current practice and to produce recommendations or Guidelines for the creation and usage of electronic texts in key linguistic and literary disciplines. The first research phase of the TEI came to an end in 1994 with the publication of TEI P3, which over the next few years was to become the reference standard for the building of the digital library. At the start of the current century, the TEI re-established itself as a membership consortium, jointly hosted by four institutions, two on either side of the Atlantic, and managed by a Board of Directors and its technical work overseen by an elected Council. There are currently nearly 100 members of the Consortium, among them universities, research libraries, academic and other publishers, both non-profit and commercial, as well as scholarly societies and research projects concerned with the design, production and delivery of structured electronic text. One does not, of course, need to be a member of the Consortium to use the TEI, and indeed there are thousands of users worldwide. In 2002 the first major revision of the Guidelines, known as TEI P4, was published. This was a "maintenance release", seeking only to bring the Guidelines up to date with changes in the technical infrastructure, most notably in the use of the W3C's Extensible Markup Language (XML) as its means of expression, rather than the ISO standard SGML used by earlier editions. Since then, work has been proceeding on a complete revision of the TEI Guidelines, TEI P5, initial releases of which are now available from Source Forge. In this presentation we will sketch some of the changes introduced with this new release, in particular the major new module for the description of manuscripts and other kinds of primary sources, based chiefly on the work of the EU-funded MASTER project (1999-2001) and the TEI Medieval Manuscripts Description Work Group (1998-2000), but with significant input also from complementary work done by other agencies, notably the Repertorium of Old Bulgarian Literature and Letters project (1996- ). We will also show how P5's modular construction facilitates the customisation of the TEI scheme for use in a wide variety of literary and linguistic study, and speculate about its likely implications for future encoding work consequent on the internationalisation effects currently under way. The session is intended as "an orientation and workshop", meaning that there will be both a theoretical and a practical aspect. Previous experience with TEI mark-up would be a decided advantage, but is not required.

The recent work of the Text Encoding Initiative (TEI) in developing an XML standard for manuscript description has provided an opportunity to explore methods of making such information more widely available through a web-based system. The emergence of native XML databases as a viable storage mechanism has made web delivery of XML-encoded material easier and more efficient. The presentation discusses a prototype system developed as part of a research project carried out at the University of Copenhagen's Arnamagnæan Institute in the summer of 2005. The Arnamagnæan Manuscript Collection, which derives its name from the Icelandic scholar and antiquarian Árni Magnússon (1663-1730), comprises the single most important collection of early Nordic manuscripts in the world, in all some 3000 items, the bulk of them Icelandic. The collection is divided between sister institutes in Copenhagen and Reykjavík. In 1997, representatives of the two Arnamagnæan institutes began looking into the possibility of reuniting the two halves of the collection virtually, in particular through the development of a searchable web-based catalogue. From January 1999 through June 2001, the Arnamagnæan Institute in Copenhagen participated in the MASTER project, an EU-funded project whose goal was to define and implement a general purpose standard for the description of manuscript materials using XML. Since then, encoding work has proceeded apace at both institutes, if at times only slowly, and there are now at least minimal records for the entire collection. There has not, however, until now, been a suitable system for searching and browsing these records, and the goal of the research project was to create a such a system. Using PHP and the eXist native XML database, a three-tier web database application was developed. Users are provided with a facility for executing queries on the database through a web form designed to allow complex query formulation involving many different criteria. The web system is multi-lingual and places an emphasis on usability, standards-compliance and the use of open-source software. The realisation of this resource demonstrates a method by which other institutions may undertake similar projects involving XML-encoded source material. Because the manuscript records are encoded in standardised TEI XML and have a known and consistent structure, the potential exists for integrating records from this project with those from other collections, thereby creating a larger and more complete catalogue. Indeed, this could easily be the first step toward a (virtual) union catalogue of manuscripts in European repositories. And by allowing users to combine many different search criteria, it also has the potential to change the way people do research into manuscripts. The presentation will comprise a demonstration of the web resource, including examples of XML source material, the eXist database system, application code, and the web interface.

True accessibility requires minimizing the scanning time to find a particular piece of information. Sequentially reading web pages do not provide this type of accessibility, for instance before the user gets to the actual text content of the page it has to go through a lot of menus and headers. However if the user could navigate a web page based through semantically classified blocks then the user could jump faster to the actual content of the page, skipping all the menus and other parts of the page. We propose a transcoding engine that tackles accessibility at two distinct, yet complementary, levels: for specific known sites and general unknown sites. We present a tool for building customized scripts for known sites that turns this process in an extremely simple task, which can be performed by anyone, without any expertise. For general unknown sites, our approach relies on statistical analysis of the structural blocks that define a web page to infer a semantic for the block.

Walking is becoming increasingly popular as a leisure activity across Europe. Outdoor equipment has modernized, gained flexibility and lost weight. GPS devices are gaining popularity. More and more websites are created for walkers. On the other hand information about walking paths is still very diverse in terms of source, structure, availability and language. This is why it is still difficult for outdoor enthusiasts to get all the trail information they need. The project WalkOnWeb wants to overcome this type of problems by defining a new publishing model for walking and tourist information. This paper describes the information structures defined by the WalkOnWeb project to solve this problem in an electronic publishing model. It explains what information layers the project defines for a walk and describes the information model in detail, including the different ontologies developed and used in the project. The specific focus of this paper is on the walk and navigation ontologies, including how both are used to create language independent navigation instructions. In this authoring paradigm authors do not describe navigation instructions literally in textual form but they write navigation instructions in a rigorously structured way, selecting possible types of instruction from a list and using standardised concepts taken from the ontologies. This paper describes how this works and how these structures are then used in a prototype to generate a textual instruction in a certain language and walking direction.

Interfaces for users on the web are usually created using standard HTML-technologies. Although dynamic text-based features are widely available on the web (e.g. by using DHTML) there are still some limitations in the user experience when using dynamic graphical elements. In this article we focus on a GUI-prototype using Scalable Vector Graphics (SVG) as an interface to an existing Digital Library. First we describe the current situation of the running system and explain some of the limitations in the GUI. Thereafter we compare the document-format with other electronic formats. An overview of the prototype and a detailed description of the used framework will be given in the main section of this article. Finally we are going to show an improved version of a search task using an SVG-category viewer. Currently supported SVG-recommendation (SVG 1.1) by the SVG-plugins have some limitations. The upcomming release will solve some of the problems and introduce new features related to data-handling and database access. An outlook of these new features and lessons learned in implementing the prototype will close the article.

This paper shows an approach for converting bitmap images of text glyphs into a vector format which is suitable for being embedded in XML representations of digitized documents. The focus is on a contour based vectorization method as the output can be easily transformed into SVG glyph descriptions. A concrete implementation is described and the results are discussed with special regard to the visual quality. The work is related to the development of a system for processing documents which are not suitable for current OCR methods. This is especially important in the field of retrospective digitization of historical works.

In this paper we present HyperJournal, an Open Source web application for publishing on-line Open Access scholarly journals. In the first part (sections 1, 2 and 3) we briefly describe the project and the software. In sections 4 and 5, we discuss the weaknesses of the current publishing model and the benefits deriving from the adoption of Semantic Web technologies, outlining how the Semantic Web vision can help to overcome the inefficiencies of the current model. In the last two sections (6 and 7), we present two experimental applications, developed on top of HyperJournal, with the purpose of demonstrating how the technologies can affect the daily work of scholars. The first application is a tool for graphically visualizing the network of citations existing between articles and their authors, and for performing bibliometric measurements alternative to the ISI Impact Factor. The second is a tool for automatically extracting references from non-structured textual documents, which is part of a tool-chain for the extraction of hidden semantics.

Scientific journals have always been an important tool in scientific research, and the advent of new technologies has fostered their development into e-journals that has brought about new editorial techniques and methods. Thus, e-journals have become the easiest and fastest means to meet the needs of researchers in their research works as the Internet and its services represent a tremendous opportunity for communication, edition and information retrieval. The Algerian 'savoir-faire' in this field has led to the setting up of an e-journals repository with a double experience that has paved the way to using the open source editorial system SPIP. This paper deals with the Algerian experience starting from 1999 by creating a national database for scientific journals that had to be accessible online for the researchers' community. Even if it was relatively underestimated, it encouraged our team to go ahead and look for new tools to enhance the data base, its content and also the web site. After many tests on a few electronic publishing software such as LODEL and GREENSTONE, SPIP appeared to be the best one to meet our needs especially that it includes Arabic. Furthermore, we are keen to start the process of adhering to interoperability international standards in order to make our contents more accessible. In addition, a study has been initiated on the archiving and long term preservation issue, particularly with an XML solution.

Most of the studies conducted on future business models for electronic scholarly publishing have concentrated on the major publishing areas of North America and Western Europe, and on large publishers (both for-profit and not-for-profit). This paper considers the prospects for electronic scholarly publishing in a smaller country away from these two parts of the world. Malaysia is a medium-sized developing country in Southeast Asia, with the ambition to become fully developed by 2020. The government has invested heavily both in ICT infrastructure and in educating the population in its use, and as a result Internet usage is quite high, and most academics and research workers can access the Internet both at work and at home. A number of journals are published by not-for-profit organisations in Malaysia, but their sales are small and their financial positions precarious, and few of them are available electronically yet. This work is based on a large-scale questionnaire survey of Malaysian scientists, and interviews with managers of university presses and other not-for-profit publishers in Malaysia, designed to reveal attitudes to online electronic journals among their potential authors and potential publishers. Those academics who published frequently in printed scholarly journals were significantly more positive in their attitude towards online journals, and this effect was particularly strong if they published in international journals and in English. Interviews with representatives of the presses of universities and research institutes revealed that most Malaysian journals are small, publishing only 20-30 papers per year, and have low print runs, typically about 300. They gain relatively little income from subscriptions, many copies being distributed though exchanges, and are subsidised by their host institutions. Only one journal among this sample of presses had a parallel electronic version; the remainder were print-only. An important feature of Malaysia is its very high value on Hofstede's Power Distance Index (PDI), which implies that a high level of deference to one's superiors is usual in Malaysia. The results from the interviews suggested strongly that the effect seen in scholarly publishing is a consequence of senior managerial figures in universities not supporting online publishing. Those researchers with a more international orientation are the ones most likely to go against the cultural trend.