J
Joelle Andrews
Guest
Business applications drive the core operations of many organizations, and for those built on Progress OpenEdge, decades of continuous service and adaptation are the norm. But with the benefits of maturity come pressing questions: How do you keep your underlying codebase clean as requirements evolve? What’s the best way to address technical debt and ready your systems for future success, especially as you upgrade to new platforms like OpenEdge 12?
In our webinar, In Conversation: Building a Healthy OpenEdge Codebase, Roland de Pijper from Progress and Tibor Lapikas from Software Improvement Group (SIG) discussed these topics, providing advice, real-world examples and practical recommendations. The webinar focused on approaches for building and maintaining an OpenEdge codebase, using AI as a coding tool and addressing current issues in the industry.
Tibor Lapikas serves as Partner Director at Software Improvement Group (SIG) where he supports clients and partners in embedding high standards for software quality and security. With a strong background in software engineering and consultancy, Tibor has spent over 10 years helping enterprises build secure, reliable and maintainable systems, focusing on turning practical software quality management into business results.
Roland de Pijper is a Senior Principal Consultant at Progress Professional Services and has been with Progress since 1995. Based in Rotterdam, Roland is a recognized advisor to customers worldwide, helping them adopt the latest OpenEdge technologies and navigate modernization journeys without sacrificing business continuity.
Early in the conversation, Roland points out that OpenEdge applications often have a long legacy. “Our databases work solid, and we are always backwards compatible. So even your Version 4 code will still run in OpenEdge 12.” This remarkable compatibility provides freedom but also risk. Years of urgent changes, incremental enhancements and business-critical patchwork can leave parts of the codebase tangled and hard to maintain.
When code quality drops, so does your ability to innovate. Issues linger, delivery slows and technical debt creeps in, eventually taking time away from new development and threatening your ability to meet business needs.
Tibor describes technical debt as “any maintenance that you did not do on your codebase.” He explains two common types:
Roland reinforces that real business pressure intensifies the problem: “When does the customer need it? Right now. We have to implement it as soon as possible and code quality will suffer from this.” Best practices may be skipped for the sake of delivery, and, over time, these compromises compound into significant technical debt that becomes harder to unwind.
Measuring and benchmarking code quality is essential, especially when organizations want to guide improvements and track progress over time. Tibor introduces SIG’s method of quantifying maintainability with a star-rating system that ranges from one to five. “Three stars is market average,” he explains, giving teams a clear point of comparison with their peers. The general recommendation: aim for four stars to strike a good balance between agility and long-term maintainability.
OpenEdge users can experience this objective code quality approach through the Application Quality and Security Management (QSM) service. The QSM service, powered by the analytical expertise of SIG’s Sigrid platform, offers deep, data-driven insights into an OpenEdge codebase’s structure, maintainability and security. By benchmarking OpenEdge code against one of the world’s largest industry databases, QSM delivers transparent and actionable guidance for business and development leaders. Instead of guessing where to focus or relying on gut feeling, teams receive a clear roadmap for tackling technical debt and prioritizing the changes that will have the most impact, with guidance and support from Progress Professional Services and the global software assurance standards established by SIG.
Crucially, Tibor cautions against trying to overhaul everything in a codebase once: “If you want to go from two and a half to four stars all at once, that's an enormous effort.” Instead, teams should focus on improving areas with the biggest returns, applying higher quality standards to new code and targeting hotspots that see the most frequent changes or challenges.
Both speakers agree that gradual, ongoing improvement is the key. Roland suggests making use of sprint cycles: “Reserve some percentage of your time in every sprint. Try to prioritize those issues and, every sprint, take some extra time just to fix those kinds of issues. If you are working in a part of your code and know something needs to improve, apply the boy scout rule: leave the code a little better than you found it.”
Their advice echoes a common-sense mantra: don’t try to boil the ocean. Set realistic goals, focus on high-impact areas and empower developers to make incremental improvements as part of their daily work.
The conversation then turns to the promise and limitations of AI coding assistants, tools that leverage artificial intelligence to help developers write code, automate repetitive tasks and even suggest tests or improvements. While these assistants are becoming increasingly skilled, not only generating functions but also supporting broader architectural tasks, Roland and Tibor emphasize that AI is just one part of the development landscape.
Human expertise, review and ongoing collaboration remain central to creating high-quality software, with AI serving as a valuable complement rather than a complete solution. “AI agents are not a silver bullet,” says Roland. “Developers need to be in control.” Tibor emphasizes that AI tools are most effective when experienced developers use them wisely and when organizations invest in training and incremental rollout.
Progress is working in this space, delivering solutions like QSM that not only analyze code for quality and security but can be connected to AI agents for direct, automated feedback. Still, the code you generate with AI must be validated to prevent new technical debt or security issues. The most effective approach is “human in the loop”—keeping developers responsible for final acceptance, review and continuous improvement.
Every OpenEdge codebase is unique, but the path to sustained health is remarkably similar:
As Tibor puts it, “AI will take a bigger role in doing the actual work of generating the code, but what happens and what needs to be done will still be mainly driven by the developers themselves.” The upshot? Brighter prospects for OpenEdge teams without having to sacrifice stability, continuity or innovation.
Now is an excellent time to get started. Even small changes, guided by the right metrics and tools, can make a big difference over the long term. Don’t hesitate, and don’t go it alone. Contact the Progress Professional Services team to learn more about Application Quality and Security Management to support the health of your OpenEdge codebase.
Continue reading...
In our webinar, In Conversation: Building a Healthy OpenEdge Codebase, Roland de Pijper from Progress and Tibor Lapikas from Software Improvement Group (SIG) discussed these topics, providing advice, real-world examples and practical recommendations. The webinar focused on approaches for building and maintaining an OpenEdge codebase, using AI as a coding tool and addressing current issues in the industry.
Meet the Experts
Tibor Lapikas serves as Partner Director at Software Improvement Group (SIG) where he supports clients and partners in embedding high standards for software quality and security. With a strong background in software engineering and consultancy, Tibor has spent over 10 years helping enterprises build secure, reliable and maintainable systems, focusing on turning practical software quality management into business results.
Roland de Pijper is a Senior Principal Consultant at Progress Professional Services and has been with Progress since 1995. Based in Rotterdam, Roland is a recognized advisor to customers worldwide, helping them adopt the latest OpenEdge technologies and navigate modernization journeys without sacrificing business continuity.
Why Code Quality Matters for OpenEdge Teams
Early in the conversation, Roland points out that OpenEdge applications often have a long legacy. “Our databases work solid, and we are always backwards compatible. So even your Version 4 code will still run in OpenEdge 12.” This remarkable compatibility provides freedom but also risk. Years of urgent changes, incremental enhancements and business-critical patchwork can leave parts of the codebase tangled and hard to maintain.
When code quality drops, so does your ability to innovate. Issues linger, delivery slows and technical debt creeps in, eventually taking time away from new development and threatening your ability to meet business needs.
Recognizing and Managing Technical Debt
Tibor describes technical debt as “any maintenance that you did not do on your codebase.” He explains two common types:
- Code-level complexity: Functions with too many decision points, deeply nested logic or unclear structure. “It happens line by line… one small ‘quick fix’ at a time,” Tibor notes, “but if you keep deferring the cleanup, eventually your kitchen is very messy and it's going to take a very long time to clean up.”
- Architectural complexity: Over time, teams add modules and cross-dependencies without clear separation, making it risky to change one area without unexpected impact elsewhere.
Roland reinforces that real business pressure intensifies the problem: “When does the customer need it? Right now. We have to implement it as soon as possible and code quality will suffer from this.” Best practices may be skipped for the sake of delivery, and, over time, these compromises compound into significant technical debt that becomes harder to unwind.
Measuring Code Quality: The Star Rating Approach
Measuring and benchmarking code quality is essential, especially when organizations want to guide improvements and track progress over time. Tibor introduces SIG’s method of quantifying maintainability with a star-rating system that ranges from one to five. “Three stars is market average,” he explains, giving teams a clear point of comparison with their peers. The general recommendation: aim for four stars to strike a good balance between agility and long-term maintainability.
OpenEdge users can experience this objective code quality approach through the Application Quality and Security Management (QSM) service. The QSM service, powered by the analytical expertise of SIG’s Sigrid platform, offers deep, data-driven insights into an OpenEdge codebase’s structure, maintainability and security. By benchmarking OpenEdge code against one of the world’s largest industry databases, QSM delivers transparent and actionable guidance for business and development leaders. Instead of guessing where to focus or relying on gut feeling, teams receive a clear roadmap for tackling technical debt and prioritizing the changes that will have the most impact, with guidance and support from Progress Professional Services and the global software assurance standards established by SIG.
Crucially, Tibor cautions against trying to overhaul everything in a codebase once: “If you want to go from two and a half to four stars all at once, that's an enormous effort.” Instead, teams should focus on improving areas with the biggest returns, applying higher quality standards to new code and targeting hotspots that see the most frequent changes or challenges.
Practical Steps for Sustainable Improvement
Both speakers agree that gradual, ongoing improvement is the key. Roland suggests making use of sprint cycles: “Reserve some percentage of your time in every sprint. Try to prioritize those issues and, every sprint, take some extra time just to fix those kinds of issues. If you are working in a part of your code and know something needs to improve, apply the boy scout rule: leave the code a little better than you found it.”
Their advice echoes a common-sense mantra: don’t try to boil the ocean. Set realistic goals, focus on high-impact areas and empower developers to make incremental improvements as part of their daily work.
Embracing AI Agents—With Care
The conversation then turns to the promise and limitations of AI coding assistants, tools that leverage artificial intelligence to help developers write code, automate repetitive tasks and even suggest tests or improvements. While these assistants are becoming increasingly skilled, not only generating functions but also supporting broader architectural tasks, Roland and Tibor emphasize that AI is just one part of the development landscape.
Human expertise, review and ongoing collaboration remain central to creating high-quality software, with AI serving as a valuable complement rather than a complete solution. “AI agents are not a silver bullet,” says Roland. “Developers need to be in control.” Tibor emphasizes that AI tools are most effective when experienced developers use them wisely and when organizations invest in training and incremental rollout.
Progress is working in this space, delivering solutions like QSM that not only analyze code for quality and security but can be connected to AI agents for direct, automated feedback. Still, the code you generate with AI must be validated to prevent new technical debt or security issues. The most effective approach is “human in the loop”—keeping developers responsible for final acceptance, review and continuous improvement.
The Path Forward: Practical Takeaways
Every OpenEdge codebase is unique, but the path to sustained health is remarkably similar:
- Raise awareness of technical debt and its impact
- Measure and benchmark code quality with clear goals
- Focus improvements on new code and key hotspots
- Prioritize ongoing, incremental fixes alongside new features
- Use AI and automation to help, not replace, thoughtful developers
- Reach out to solutions experts and leverage assessment tools when deeper architectural changes are needed
As Tibor puts it, “AI will take a bigger role in doing the actual work of generating the code, but what happens and what needs to be done will still be mainly driven by the developers themselves.” The upshot? Brighter prospects for OpenEdge teams without having to sacrifice stability, continuity or innovation.
Now is an excellent time to get started. Even small changes, guided by the right metrics and tools, can make a big difference over the long term. Don’t hesitate, and don’t go it alone. Contact the Progress Professional Services team to learn more about Application Quality and Security Management to support the health of your OpenEdge codebase.
Continue reading...