ABOUT
Four Decades Inside Enterprise Systems - Now Exploring What Generative AI Makes Possible.
I have spent more than four decades designing, developing, supporting, maintaining, and modernizing mission-critical enterprise systems. Today, I am exploring how Generative AI can assist organizations in understanding those systems, recovering the business knowledge embedded within them, and approaching modernization responsibly.
That experience taught me that an organization’s greatest asset is not its software—it is the business knowledge embedded within its systems. Business rules, policies, decisions, exceptions, processes, and operational knowledge can accumulate over decades, often becoming difficult to separate from the technology that carries them.
.jpg)
Four Decades Inside Enterprise Systems
My career in computing began in 1976 and has taken me through more than four decades of enterprise systems work—from mainframe applications and databases to large-scale public- and private-sector systems. Across those years, I have designed, developed, supported, maintained, analyzed, and modernized systems whose value extended far beyond the technology itself.
Working with these systems taught me that the code is only part of the story. Embedded within programs, data, documentation, procedures, and years of accumulated change are business rules, policies, decisions, exceptions, dependencies, and operational knowledge. Much of that knowledge may never have been completely documented outside the systems themselves.
Generative AI has opened a new question for me: can these technologies help us investigate legacy systems at a scale that was not previously practical—connecting code, data, documentation, business rules, dependencies, and human knowledge—while keeping validation, authority, and accountability firmly in human hands?
An Investigative Approach to Generative AI
My interest in Generative AI begins with questions, not conclusions. I am investigating where GenAI can genuinely assist the understanding and modernization of enterprise systems, where its limitations become significant, and where human expertise must remain central.
The guiding principle is simple: we follow the evidence; we don't make the evidence follow our enthusiasm. Propositions should be investigated, challenged, refined, and, where the evidence requires it, rejected. Generative AI may help us investigate more broadly and connect information more quickly, but possibility should never be confused with proof.
GenAI can assist discovery by investigating, connecting, comparing, questioning, documenting, testing, and surfacing ambiguity. Human expertise and governance remain essential to establishing authenticity, completeness, meaning, authority, and accountability. The objective is not to replace human judgment, but to determine where GenAI can responsibly extend our ability to understand complex enterprise systems.
Where the Investigation Leads
This work is not intended to begin with a predetermined answer. It begins with propositions about legacy systems, institutional knowledge, Generative AI, human validation, governance, and modernization—and asks what the evidence will ultimately support.
If the investigation demonstrates that GenAI can responsibly help recover, organize, connect, and preserve business knowledge embedded within enterprise systems, that evidence may point toward better approaches to modernization—and perhaps, over time, toward better ways of developing, maintaining, and supporting enterprise systems throughout their lifecycle.
The propositions published on this site are therefore invitations to investigation and discussion. They will evolve as evidence develops, questions are answered, new questions emerge, and experience challenges what we initially believed might be possible.