Scientific research is an important component of knowledge. Much contemporary scientific research along with its associated metadata is available in digital format via various means, such as digital libraries, research management systems, IRs, and publishers’ platforms. Open access to scientific research enhances further development of science.⁴ Maximizing the visibility of scientific research is essential for scientific advancement. Visibility can be enhanced by putting research into IRs that are OAI-PMH interoperable.⁵ The OAI-PMH protocol was primarily developed as a low-barrier method for interoperability between metadata repositories and provides an interoperability framework based on metadata harvesting by defining two classes of participants: data providers that expose metadata and service providers that harvest metadata. The IR's metadata schema has a key role in increasing interoperability of the repository, i.e., maximizing visibility of theses and dissertations that are stored in the repository.⁶

A rich metadata schema enables establishing relations between various systems that contain scientific research. Collaboration between those systems has been discussed in recent years. According to Joint, sharing data between institution repositories and research management systems to avoid duplication of efforts is necessary.⁷ Joint recommends a single point of entry to research articles regardless of whether it is through an institution repository or a research management system. He notes that three institutions (Glasgow University, Southampton University, and Kingston University) have already implemented this approach.

Krause suggests the creation of a virtual library that aims to enable users to gain integrated access to all relevant information in their special scientific field, irrespective of the location of metadata and digital form of documents.⁸ A virtual library includes a single point for creation of queries that are sent to all systems that are part of the virtual library and integrates results retrieved from the systems.

“NARCIS: The Gateway to Dutch Scientific Information,” by Dijk and colleagues, describes the National Academic Research and Collaborations Research System (NARCIS) portal (www.narcis.nl), which provides access to all scientific research information in the Netherlands.⁹ That system is an integration of the Netherlands research management system and the Digital Academic Repositories in the Netherlands (DARENET).¹⁰ Olivier describes collaboration between the research management system and the digital library at Pretoria University.¹¹

The general objective of the CRIS-IR group is “to work out an optimal solution for the interoperability of Research Management Systems on the one hand and Institutional Repositories on the other, on a European scale, taking into account all relevant aspects.”¹² The aim of the Current Research Information Systems and Open Access Repository (CRIS/OAR) interoperability project is to increase the interoperability between research management systems and open access repositories “by defining and proposing a metadata exchange format for publication information with an associate common vocabulary.”¹³

The aim of integrating systems that contain scientific research is to maximize visibility of scientific research and avoid duplicating input of the same metadata in various systems. Many libraries worldwide store metadata in the MARC 21 format. Those libraries have electronic services that enable downloading metadata about bibliographic records related to scientific research, thus the interoperability of a repository of scientific research with those libraries is important for increasing visibility. Frequently, electronic services will enable metadata exchange in DC, but the fact that DC is not strict a standard can cause problem in metadata interoperability. However, the metadata defined by this format are a subset of metadata defined by MARC 21. When searching databases of scientific research, users can better express their information needs if the research is described in a richer set of metadata. A CERIF-compatible data model based on the MARC 21 format makes CRIS systems interoperable with library information systems.¹⁴ In this model some CERIF data are stored in the MARC 21 format data model. As noted earlier, CERIF defines a data model that enables interoperability between CRIS implementations; MARC 21 is a standard for storing data for library systems. That model includes all entities and attributes of the CERIF data model and preserves the existing references between the CERIF data model entities. Furthermore, that model enables input of multilingual data prescribed by the CERIF standard. The MARC 21 format is rich in metadata and enables more detailed description of entities in CRIS systems. A MARC 21 record can store all metadata prescribed by DC and ETD-MS format.¹⁵ An information system based on the CERIF-compatible data model can exchange data with other systems using XML documents (which have XML schemas prescribed by CERIF standard) and can exchange data with LIS based on the MARC 21formats and with IRs based on DC or ETD-MS format.

CRIS UNS is a CERIF-compatible research management system under development since 2008 at UNS. The first phase of CRIS UNS development was the implementation of a system for entering metadata about published scientific research including papers published in journals, papers from scientific conferences, monographs, and papers published in monographs.

CRIS UNS is built on the CERIF-compatible data model based on the MARC 21 format described in the previous section. CRIS UNS was implemented as a web application based on “best-of-breed” open-source components written in Java.¹⁶ The system has three-tier architecture. Three-tier architecture contains a client tier (the presentation logic, including simple control and user input validation), middle tier (the business processes logic and the data access), and data tier (the data server provides the business data).¹⁷ Any web browser supporting HTML 4 and JavaScript can be used for application access.

The server side of the system is executed within the Apache Tomcat (http://tomcat.apache.org) application server. Apache Tomcat is an open-source software implementation of the Java Servlet and JavaServer Pages (JSP) technologies. A servlet is “used to extend the capabilities of servers that host applications accessed via a request-response programming model. Although servlets can respond to any type of request, they are commonly used to extend the applications hosted by Web servers.”¹⁸ JSP technology “provides a simplified, fast way to create dynamic web content.”¹⁹

The presentation tier is developed using the JavaServer Faces (JSF) development environment (www.jcp.org/en/jsr/detail?id=252) and RichFaces (www.jboss.org/richfaces) library. JSF technology simplifies building user interfaces for JavaServer applications, and RichFaces is a library of Ajax-enabled components for JSF. The Apache Lucene (http://lucene.apache.org) library is an open-source information retrieval library written in Java and is used for indexing and searching text contents. Text indexing and query processing include a Cyrillic to Latin transliteration algorithm. All index entries are stored as Latin text, thus enabling the use of both scripts in searching. On the other hand, database contents hold information as it was entered by the user, preserving the correct script. Cyrillic to Latin transliteration is unambiguous. This means that every character of Cyrillic has an appropriate character in the Latin (or Roman) alphabet—and that every word written using Cyrillic can be unambiguously translated to a word using Latin characters. The MySQL (www.mysql.com) database management system is used for data preservation.⁴ The system data model and architecture enable easy integration of the system with LIS and interoperability with other CERIF-compatible national CRIS systems.

Published results from the system are available to anonymous users via the Internet. Moreover, the system is in accordance with the CERIF standard and meets requirements prescribed by the Republic of Serbia Ministry of Science and Technological Development in the field of scientific research results evaluation. Therefore the system data model is extended with necessary entities.²⁰ The system is implemented as a web application that enables authors to input metadata about their own research without knowledge of the CERIF standard and the MARC 21 format.

The first step in this project was analysis of various systems that contain metadata about theses and dissertations. The following are international initiatives:

• NDLTD (www.ndltd.org) is an international organization that aims to create a worldwide network of ETDs. Each digital repository that is a network member has to enable metadata exchange in the ETD-MS format (developed by DNLTD) in accordance with OAI-PMH.²¹
• DART-Europe E-Thesis Portal (www.dart-europe.eu/contributors/how.php) aims to collect details of the open access research theses stored in Europe's digital repositories (doctoral and master theses). It collects metadata in DC using OAI-PMH.
• DRIVER (www.driver-community.eu) is an international organization co-funded by the European Commission with the goal of creating a network of freely accessible digital repositories with content across all academic disciplines. Each digital repository that is a network member has to enable metadata exchange in DC in accordance with the OAI-PMH protocol.

In addition, many academic and research institutions and research communities may implement and manage the following approaches to collecting, preserving, accessing, and disseminating research:

• IRs are online systems that collect, preserve, and disseminate the intellectual output in digital form of an institution. IRs may use open-source software, such as DSpace (www.dspace.org) and Fedora (http://fedora-commons.org), or hosted, proprietary software, such as Digital Commons (http://digitalcommons.bepress.com) and SimpleDL (www.simpledl.com). Many IRs support the exchange of data in DC via OAI-PMH.
• A CRIS is a database of other information system for storing data on current research (e.g., data about institutions, researchers, research projects, equipment, published results, etc.). The European Union encourages the development of national research management systems in accordance with the CERIF standard.²² CERIF-compatible research management systems are called CRIS. Due to specific local or national requirements, CRIS systems are built on different modifications (or extensions) of CERIF data model.²²
• A library information system (LIS) is a software system for acquiring, cataloging, and circulating library holdings. LIS are built on various bibliographic standards; most are based on MARC 21 formats.

Across these systems, different standards and protocols—CERIF, OAI-PMH, DC, ETD-MS, and MARC—enable interoperability.

After analysis was completed, a comprehensive metadata set was defined to develop a repository that is compatible with all previously mentioned systems.Then the authors extended the CRIS UNS data model to store all metadata about ETDs as well as the ETDs as digital objects. Finally, the authors expanded CRIS UNS with a module for storing ETDs along with associated metadata. An object-oriented method was used for the module modeling. Object-oriented modeling creates models using object-oriented diagrams (class diagram, sequence diagram, etc.), which is the starting point for implementing a system using object-oriented programming language. The modeling was carried out using the Sybase PowerDesigner tool that supports OMG's Unified Modeling Language (UML) 2.0 (www.omg.org/spec/UML/2.0). The module model can be obtained by contacting the authors. The implementation was realized using “best-of-breed” open-source components written in Java. After the authors developed the module, it was verified and tested on EDTs by researchers in the Faculty of Sciences, UNS. After migration of the existing dataset containing ETDs along with associated metadata from the DIGLIB UNS system to the CRIS UNS system, UNS researchers verified the migrated data about their theses and dissertations and supplied additional data; the CRIS UNS metadata set is richer than the DIGLIB UNS metadata set. These steps are covered in detail in the following sections.

This paper describes the implementation of a digital repository of EDTs within the CRIS UNS system. Metadata about theses and dissertations are stored in the MARC 21 bibliographic format. The implementation is based on open-source components. The system architecture allows an easy transition to other bibliographic standards and easy integration with LIS based on the adopted bibliographic standard.

The system can exchange ETDs metadata with other CRIS systems, IRs, the NDLTD network members, and LIS. Interoperability with previous stated systems maximizes visibility of ETDs from the repository without duplicate entry of ETDs metadata in various systems. Metadata are entered once, but metadata are stored in various systems across the Internet. High international visibility of theses and dissertations of researchers from University of Novi Sad enhances the further development of science and raises public awareness of UNS research.

The system for inputting of ETDs has been verified and tested on a dataset containing EDTs by researchers at Faculty of Sciences, UNS. The addition of ETDs from additional fourteen UNS faculties is in progress. After this process is finished, further effort to consolidate data will be necessary; this will include such activities as removing duplicated items and consolidating scientific fields and disciplines. After this step, web services for data exchange will be made available for public access. Finally, an audit will be performed to assess whether the visibility of scientific research from UMS has increased after this repository implementation.