Designing the Register: Fields, Sources, Scoring

Everything to this point has been about seeing opposition clearly. This module is about capturing what you see in a durable, comparable form, so that scattered observations become an intelligence product you can actually reason over. That product is the opposition register: a structured record of opposition and permitting events across the markets you care about, designed so that a pattern which is invisible in any single project becomes obvious across many. Let me be concrete about how to build one, because the design choices determine whether the register earns its keep or becomes a graveyard of stale notes.

Start with the unit of record. The right unit is the event, not the project. A single project generates many events over its life: an application filed, a hearing continued, a moratorium enacted, a study demanded, an appeal filed, a denial issued, a mitigation agreed. If your unit is the project, you flatten this history into a single stale status. If your unit is the event, you preserve the trajectory, and trajectory is where the early warning lives. Each event then rolls up to a project, and projects roll up to a jurisdiction and a market, giving you the three levels at which you will want to reason.

Now the fields. For each event, capture the essentials that let you compare across cases: the date, the jurisdiction at the finest resolution you can get, the project and its rough scale, the event type drawn from a controlled vocabulary, and the outcome state, advanced, delayed, or blocked, that we defined earlier. Controlled vocabularies matter enormously here. If one analyst writes rejected and another writes denied and a third writes voted down, your register cannot count, and a register that cannot count cannot find patterns. Impose the discipline of fixed categories even though it feels rigid, because the rigidity is what makes aggregation possible.

The most valuable field, and the one most often done badly, is the reason. Capture the stated grievances against a fixed taxonomy: water, power and rates, noise, traffic, viewshed, property values, procedural or trust failures, and so on. Record them at the moment of the event, because reasons reconstructed months later are contaminated by hindsight. When you have reason codes recorded consistently across hundreds of events, you can answer the questions that actually matter: what does this jurisdiction reliably object to, which grievances predict a denial versus mere delay, and what would a survivable proposal here need to preempt.

Sourcing determines whether the register is trustworthy, so record it explicitly. Every event should carry its source, a meeting agenda, a local news report, a filed document, a public comment record, and ideally a link and a retrieval date. The register's credibility rests on traceability; an assertion no one can check is worse than a gap, because it invites false confidence. Prefer primary sources, the actual agenda and the actual filing, over secondary reporting, and note when you are relying on a single unconfirmed account.

Layer scoring on top of the factual record, but keep the score separate from the facts. A useful register carries a small number of derived assessments: an opposition-intensity score for a project, capturing how far along the campaign life cycle it has run and whether external professionals have engaged; and a jurisdiction-hostility score, aggregating the outcome history and reason patterns of a place. Keep these scores transparent and rule-based rather than intuitive, so that two analysts scoring the same facts land in the same place, and so that a reader can see why a score is what it is. A score whose derivation is hidden cannot be trusted or improved.

Finally, design for freshness and decay. Opposition is a moving target, and a register is only as good as its currency. Build in a review cadence, timestamp everything, and let old assessments visibly age so that no one mistakes a year-old quiet for present calm. The goal throughout is a living instrument that turns the daily noise of hearings and filings into a structured, queryable, honestly-sourced picture, which is precisely the input the final lesson turns into forward-looking warning.