ComSIS has entered the third year of its publishing. In the two past years, we have put a lot of effort and enthusiasm to move ComSIS towards a high-standard, high-quality journal in such a broad area of computing. So far, we have published 34 papers in four regular issues, and one special issue focusing on e-Learning. Now, we may say that we have preserved continuity, and that the beginners' problems are mostly behind us.

According to ComSIS policy and the agreement of all the other members of the Consortium, ComSIS office has been moved for the next two years to the Faculty of Technical Sciences at the University of Novi Sad, and I am honored and privileged to take the position of Editor-in-Chief. Due to generous support from the former Editors-in-Chief, Professor Branislav Lazarevic, and Professor Vladan Devedžic, this transition was smooth.

Since the beginning of this year, the ComSIS management team includes Vladan Devedžic (Vice-Editor-in-Chief), Miro Govedarica (Managing Editor), and Dinu Dragan, Srdan Popov, and Slavica Aleksic (Editorial Assistants). I believe that their remarkable experience will help ComSIS to fulfill its mission. Having in mind that a lot has to be done in near future, I hope that we will continue with special issues and further improvements of the journal quality, as well as provide world-wide bibliography indexing. It will be a great pleasure for us if the readers, reviewers, and authors recognize our intention and efforts, and further support us in this mission.

This issue of ComSIS contains one invited, and five regular research papers. The invited paper comes from Stellan Ohlsson from the University of Illinois, Chicago, and Antonija Mitrovic from the University of Canterbury, New Zeland, the distinguished authors that have devoted a lot of their time and effort to one of the hot areas in artificial intelligence: knowledge representation and learning systems. The authors emphasize that formal representation of knowledge is the constituting idea of AI. The ambition to explore the technological potential of such representations makes AI a bridge between technology and the symbolic traditions in philosophy, psychology and other cognitive sciences. The authors discuss state constraints, a type of knowledge unit originally invented to explain how people can detect and correct their own errors. They also consider the differences between constraints and other representational formats, the advantages of constraint-based models and the types of domains in which they are likely to be useful.

Data Mining is also an attractive discipline in computing that is in a close relationship to AI. Zengyou He, Xiaofei Xu, and Shenchun Deng, in their paper "Improving Categorical Data Clustering Algorithm by Weighting Uncommon Attribute Value Matches", consider the technique of clustering that has been extensively studied and used in many domains. They present an improved Squeezer algorithm for categorical data clustering by giving greater weight to uncommon attribute value matches in similarity computations.

Formalizing the process of user interface design has been a challenge in software engineering for many years. In their paper "Adapting the Unified Software Development Process for User Interface Development", Zeljko Obrenovic and Dusan Starcevic propose an approach to make use of HCI knowledge easier for ordinary software engineers who are usually not familiar with the most recent HCI research results. They consider how existing software developing processes, such as Rational Unified Process, can be adapted in order to allow disciplined and more efficient development of user interfaces.

Metaprogramming introduces a layer of abstraction above the domain language programs. It is especially useful for developing program generators for domains, where a great deal of commonalties exists. Robertas Damaševicius in his paper "On the Quantitative Estimation of Abstraction Level Increase in Metaprograms" estimates the increase of abstraction by evaluating the information content at the lower (domain) and higher (meta) layers of abstraction. The estimation method is based on the Kolmogorov complexity and uses a common compression algorithm.

There is a wide variety of different software development environments on the market that integrate language-based structure editors as their components. Despite that, the ways to enhance the quality of software development, and the productivity of developers, is still an extremely interesting research area. Zorica Suvajdžin, and Miroslav Hajdukovic, in their paper "A Structure Editor for the Program Composing Assistant", introduce The Program Composing Assistant, an interactive generic development environment that provides a structure editor with graphical user interface as a main feature. It is based on an intuitive approach, and aims to integrate important practical aspects of structure editing.

Zoran Živkovic, and Milorad Stanojevic, the authors of the paper "Simulation Analysis of Protected B2B e-Commerce Processes", indicate that further development of Internet-based business applications is jeopardized by increased security risks, threats, and attacks. Efficient security policy measures are needed to minimize these risks. They propose cryptographic security measures as an efficient solution to this problem. The paper presents a simulation analysis of certain trust models (complex PKI architectures) with regard to the security support of B2B applications on the Internet. It also underlines the significance of the up-to-date cryptographic mechanisms: digital signature and digital certificate to deliver the main security services based on PKI.

Finally, on behalf of the ComSIS Consortium, I would like to take this opportunity to thank the reviewers and all the authors for their high-quality work, great efforts, and remarkable enthusiasm.

Ivan Luković