Abel Maciel ended the 2017 conversation with a plain challenge: look back at the data your practice creates and find how it could benefit the next cycle of work.
That challenge changes the purpose of a model. The question is no longer only whether the current team can produce it. The question is whether somebody can understand and use its information when another decision arrives.
Define the information need before building the model
Abel Maciel described the employer's information requirements as the place to state what information the building owner needs and whether the model will support facilities management. He also described a BIM execution plan that sets how the team will collaborate and when it will exchange information.
Those two records put purpose before volume. They ask what the information is for, when it should move, and how the team will use it. A large model cannot answer those questions for the people working in it.
“BIM is very much a way of working as well as a technology,” Abel Maciel said. His explanation focused on a federated model, a common data environment, and information shared across the integrated team.
That framing makes coordination visible. Separate disciplines can keep their own models, but the exchange rules determine whether those parts support a shared decision. The model is useful because the people have agreed how information will move, not simply because the file contains more detail.
Automation was the boundary, not the promise
Abel was careful about the limits. In 2017, he said automating an entire complex construction project with a smart contract was not feasible. He discussed provenance, compliance, and product certification as possible narrower uses, while repeatedly describing the work as early.
The same caution appeared in his field examples. Tablets occupied the hands people needed for construction, interfaces could show too much information, and the technology was not integrated well enough to provide a settled answer. Better data does not help if the person at the review point cannot use it in the way the work requires.
Give one data set a next user
Abel Maciel said construction teams can generate large amounts of data, store it, and only remember its value when a similar problem returns. He connected future machine-learning work to the ability to identify ways that existing data could be reused.
Take one information set your team already produces. Name the person or team that could need it next, the decision it should support, and how that person will know where it came from. This does not require a new platform. It requires a clear next use.
The point is not to keep more data. It is to decide, before handover, which information should continue helping someone make a decision.
