Jeremy Barnett described a project with contractors and designers spread around the world and people threatening to leave the site. His first question was not how to win a case. It was whether the right documents could reach the right people so mediation could begin.
That distinction moved the blockchain discussion toward a more immediate decision. Can a team assemble enough trusted information to work on the problem before it becomes a formal legal dispute?
A valid claim can still be too costly to pursue
In the episode, I described my public-agency experience as a schedule manager. Calculating potential liquidated damages was part of my work, and pursuing them could cost enough that the claim no longer made practical sense.
That experience does not define every claims process. It does show why usable information matters before escalation. Jeremy Barnett named verified records and automation as two parts of his dispute-avoidance approach. The purpose was to assemble relevant documents and focus mediation. Better records can support that conversation without guaranteeing that disputes will decline.
Code still needs a legal boundary
When the conversation turned to smart contracts, Jeremy Barnett was direct: “They're neither smart nor are they contracts.” He described code that enforces a condition people chose in advance. Governing law still applies, and the legal work changes rather than disappears.
Construction also changes after an agreement is signed. New people join, field conditions emerge, and responsibilities develop. Jeremy's discussion of a contract that grows with the project was an early concept, not a finished system. It exposed the limit of treating a changing project as a fixed set of instructions.
A building's history has to outlive handover
Jeremy Barnett applied the same thinking to building records. In 2017, the consortium was exploring whether a component register could preserve certificates, dates, and warnings throughout a building's life. He clearly described that work as exploratory.
He also drew a hard line between starting a record with a new building and trying to reconstruct one later. Old plans, changed spaces, and missing details make the second task much harder. The useful owner decision is therefore not whether a ledger sounds advanced. It is which information must remain available when the building changes hands or use.
Start where the truth breaks down
Jeremy Barnett's closing guidance was to look for recurring work where an accurate record would help, where somebody else needs the information, and where a record could trigger an action. Then he asked, “Are we doing stuff the computers could be doing?”
Choose one recurring claims or compliance process. Write down which record has to be trusted, who needs to receive it, and what action it should support. Any blank in that short chain is design work to resolve before automation begins.
Dispute avoidance does not begin with software or a courtroom. It begins when people can find, trust, and use the record while there is still time to work on the problem together.
