Lessons-from-Collaboration-Insights-with-Juan-Jos-van-der-Linden-on-Model-Driven-Data-Engineering
Reflections on five years of collaboration with Juan José van der Linden, exploring the crossroads of data modeling, architecture, and practical methods that shaped our shared vision of model-driven data engineering.
Summary
Juan José van der Linden and I collaborated for five years at ASN (formerly De Volksbank).
Our conversations often revolved around bridging business and IT through clear, practical modeling approaches.
Key themes: the role of semantics, practical metadata tools, ensemble and 3NF modeling trade-offs, the art of exposing data in consumable layers, and the balance between theory and pragmatism.
His work at DataDenken and insights continue to inspire how I refine Model-Driven Data Engineering, Business-Friendly Mapping, and Business-Oriented Data Modeling.
This article distills those lessons into practical insights that remain relevant for today’s data engineering challenges.
The Value of Working Together
For five years, I had the privilege of working closely with Juan José van der Linden at ASN (formerly De Volksbank). These years were not just about delivering projects, but also about shaping a shared perspective on data architecture and the practical challenges of implementing it in a business environment.
Juan José’s ability to look critically at methodologies — whether Data Vault, 3NF, or ensemble modeling — helped me to sharpen my own ideas about Model-Driven Data Engineering. His mix of pragmatism and creativity was essential in bridging theory and practice.
Key Themes from Our Collaboration
1. Semantics at the Core
We discussed extensively how semantics should drive data modeling. Business terms and concepts need to be captured in a way that both IT and business stakeholders can understand. This is not just about naming, but about ensuring shared meaning across systems and models.
2. Pragmatism in Modeling
Instead of rigidly following one methodology, Juan José always looked for practical ways to deliver value. Whether it was 3NF business models, dual SCD2 historization, or ensemble concepts, the goal was never dogma, but clarity and usability.
3. The Art of Exposure Layers
We often debated how to design exposure views and delivery layers. The principle was simple: business users should be able to consume reliable data without being exposed to technical complexity. This philosophy aligns perfectly with my work on Business-Friendly Mapping.
4. Theory Meets Practice
What stood out most was Juan José’s ability to challenge theoretical models with practical questions. Every idea was tested against real-world constraints: performance, governance, adoption by the business.
Tools and Applications Juan José Proposed
Over the years, Juan José introduced me to a number of tools and approaches that supported our way of working:
- Ellie.ai — for collaborative and business-friendly data modeling. It showed us how modern tools can make data modeling more interactive and less IT-centric.
- Miro — as a canvas for workshops, brainstorming, and even early-stage modeling. This inspired my later work on integrating metadata and Python automation into Miro boards.
- DuckDB — a lightweight analytical database that allows for fast prototyping and proof-of-concept work. Perfect for exploring metadata-driven pipelines.
- Notion — as a knowledge and documentation hub. We experimented with capturing models, mappings, and processes in Notion for greater accessibility.
- draw.io & VSCode integration — practical for keeping diagrams version-controlled and close to metadata scripts, which later became part of my metadata-driven architecture approach.
These tools were not just technical choices. They were enablers of collaboration, making the modeling process more transparent and accessible for both business and IT.
Why It Still Matters
The conversations and experiments with Juan José laid part of the foundation for my current methodology. They taught me that model-driven approaches are only successful when grounded in communication, shared semantics, and practical tooling.
Our collaboration remains a reminder that methodology alone is not enough — you need tools, conversations, and a pragmatic mindset to bring it to life.
🔗 Learn more about Juan José van der Linden’s work: