Handle disagreement and uncertainty
Distinguish factual, predictive, value and resource disagreements; separate verification from tradeoffs.
Identify the disagreement
“We should release Friday” can involve test completion, failure risk, the priority of speed versus quality, and available support. Do not collapse them into a single question of who is right.
Use evidence for facts and criteria for tradeoffs
Test completion can be checked in records. Acceptable rework is a tradeoff. List agreed facts and preserve disputed items. Another data point may not resolve a difference in values.
Attach an action to uncertainty
Do not invent a “90% safe” estimate. Specify unknowns, checks and decision timing. A trial needs a scope, stop condition and owner. Delay or reduced scope are options when prerequisites are unmet.
Preserve disagreement without obscuring the decision
Record decision authority, criteria, unadopted concerns and a review trigger. Silence is not agreement, and understanding another position does not require pretending consensus.
Fictional case: one side prioritizes the date; another fears an untested feature. Both confirm the core path is tested and export is not. Compare delaying, disabling export or limiting the trial, including costs and authority; do not automatically compromise by testing less.
Check your understanding
Reference explanation
This may be a tradeoff disagreement rather than a factual error. Make criteria and constraints explicit.
Apply this to a specific conversation
Records stay in this browser, separately by language and lesson. The tool organizes your inputs; it does not infer intentions, prove agreement or send messages. Mark unknowns explicitly.
My communication record
My learning status
Mark statuses manually; visits do not establish mastery. Languages share statuses while lessons keep their own answers. This backup contains statuses only, not answers or notes.