About TellDear

The question of how people argue — and especially how they argue badly — didn't start with TellDear. It has been with me, honestly, for as long as I can remember. As a teenager I would take apart TV debate performances before I knew that what I was doing had a name. What started as intuitive scepticism became, over the decades, a systematic fascination.

From TellDear to TellDear

TellDear is not my first attempt to build tools for better thinking. Under the name TellDear — a play on Too Long; Didn't Read — I spent a long time working on the idea of using machines to evaluate the argumentative substance of complex texts. The core question was the same as it is today: Can you teach a computer to evaluate the quality of an argument?

TellDear didn't fail because the idea was wrong. It failed because the timing was. The language models of that era could summarise texts, but they couldn't recognise argumentation structures. What was missing was the bridge between human expertise on fallacies and machine-based text analysis. That bridge is exactly what the Atomic Instruction Dataset is — the taxonomy that powers TellDear today.

Schreiber Network

My day job is running Schreiber Network from Brandenburg, Germany — a consultancy specialising in system integrations, middleware, and connecting complex ecosystems of shops, marketplaces, payment providers, and enterprise platforms. Working in that space teaches you quickly that every system speaks its own vocabulary and that "translation" between systems is a craft. That experience feeds directly into TellDear: the AID taxonomy is, at its core, middleware between human argumentation and machine comprehension.

AI Conferences and Public Engagement

I regularly speak at AI conferences such as the AI Impact Days on the intersection of artificial intelligence and human cognition. My focus isn't the technology itself but the question of what happens when we point AI systems at domains where "right" and "wrong" aren't technical categories. Argumentation is one such domain. Manipulation is another.

Why I'm Seeing This Through

TellDear is not a side project that will quietly end up in a drawer. It is the convergence of everything that has driven me for decades: the conviction that better argumentation is not a luxury but a democratic core competency. That technology can help — not by replacing humans, but by showing them where to look more carefully.

The 452 aspects that TellDear catalogues today are just the beginning. The vision is a living, growing reference system that equips journalists, teachers, students, and anyone who wants to read critically with tools that were previously available only to specialists.

I will pursue this intensively.