Every modern organization is an archive. Tickets, commits, reviews, approvals, incidents and releases all leave time-stamped traces. Yet when a new project starts, its plan is still mostly a negotiation between optimism and experience.
Predicting the work does not mean predicting the future in general. It means asking a narrow question with a checkable answer: given this team, this kind of task and this structure of dependencies, what range of outcomes should we expect — and how often is that range right?
What would make it useful
A prediction is useful when it changes a decision and when its confidence means something. A forecast that says “80% likely by March” should be right about four times in five. Without that property, a number is decoration.
What this journal will do
We will publish the questions, the methods and — eventually — the scorecards, including the predictions that failed. This first note only states the ambition. The work of earning it has not started yet.
This is a concept note. It reports no experiments and no results.