Publications

This article investigates the post‑2020 evolution of education quality and digital competencies in Central and Eastern Europe (CEE) in the wake of the abrupt shift to online and hybrid learning during the COVID‑19 pandemic. We triangulate comparative evidence from PISA 2022, the Digital Economy and Society Index (DESI)/Digital Decade monitoring, the EU Digital Education Action Plan (2021–2027), and the DigComp/DigCompEdu frameworks to map cross‑country disparities in infrastructure, teachers’ digital readiness, and institutional capacity. Findings indicate partial convergence across several indicators—particularly connectivity and digital public services—yet reveal persistent gaps in basic digital skills, pedagogical integration of technology, and equitable access, with the southeastern EU subregion lagging behind. Urban–rural divides, uneven curriculum implementation, and insufficient large‑scale teacher upskilling remain binding constraints. Methodologically, we offer a structured regional snapshot, then contrast Romania’s trajectory with EU averages, and synthesize lessons from the United States and China to highlight similarities (connectivity versus capability paradox) and differences (scale, governance, and industrial digitalization). Policy analysis emphasizes the need for people‑centric investments that complement infrastructure: standardized DigComp‑aligned assessments, micro‑credentials for students and adults, robust teacher training pipelines, and outcome‑based funding tied to actual technology use in schools, firms, and public services. We argue that narrowing foundational skill deficits (especially among youth, rural communities, and older adults) is a precondition for translating connectivity into sustained improvements in learning outcomes and productivity. The paper concludes with a practical roadmap that prioritizes equity‑oriented delivery, sector‑linked upskilling, and consolidated governance to accelerate convergence within CEE and position the region for the EU’s 2030 Digital Decade targets.

Java programming, a high-level, general-purpose language renowned for its" Write Once, Run Anywhere"(WORA) capability, has (re) gained notable traction as developers increasingly integrate artificial intelligence (AI) generative tools into their workflows. Java’s platform independence, robust security features, and extensive libraries have made it a preferred choice for a wide range of applications, from mobile apps to large-scale enterprise systems. The advent of AI generative tools, such as ChatGPT, GitHub Copilot or Amazon CodeWhisperer, has further enhanced Java programming by automating mundane tasks, improving code quality, and fostering creativity in the development process. In today’s world, solid knowledge related to AI code generation tools is a must for all developers and software engineers. AI tools for generating Java code have also started an entire new set of debates related to copyright issues. Currently, the relevant legal frameworks, at international level, are not harmonized and in some cases even antagonistic.

The integration of ChatGPT, a state-of-the-art NLP (Natural Language Processing) model developed by the company OpenAI, with the Java programming language provides significant advancements in coding generation and software development. ChatGPT has the ability to generate useful and correct, human-like, text responses, a fact that offers developers a robust tool for automating tasks such as code generation, documentation, error diagnosis, and test case creation. Java, renowned for its extensive ecosystem and backend capabilities, is an ideal choice for leveraging ChatGPT's potential. This paper explores methodologies and best practices for integrating ChatGPT with Java, focusing on API interaction, error handling, and performance optimization. Developers can employ HTTP libraries such as OkHttp and frameworks like Spring Boot to create intelligent and scalable applications.

Complex systems are everywhere; countless examples of behaviors fall into the complex systems paradigm, from physical and natural sciences to social and economic sciences. Given the nature of these systems, where the whole is greater than the sum of its constituents, scientists must have adequate tools for investigating complex systems. Recently, Agent-Based Models (ABM) have become a de facto tool for creating idealized computer simulations to investigate pattern formation, perform root-cause analysis, or simulate alternative scenarios within the domain of complex systems. This paper introduces a miniature framework for developing and analyzing agent-based models where agents and the environment can follow vague rules. The proposed tool is applied to a sample simulation, providing a proof-of-concept example of how Fuzzy Logic and Fuzzy Inference can model complex systems with vague rules.