Decision Checklist
In the previous chapters, we organized the two patterns within the golden rules 7+1 (a system where seven internal roles consult in parallel and an external audit confirms up to three rounds). The first is the "standard pattern" (the full three-round review), and the second is the "shortcut pattern" (going directly to external audit).
Using the adoption criteria for each pattern, this chapter brings together five questions for deciding which pattern to choose. When you face a task, you can select a pattern simply by answering these questions in order.
How to Read the Checklist
Answer all five questions with either "yes" or "no."
If even one answer leans toward the "foundation side" (touching the design's base, having broad impact, or having no precedent), choose the standard pattern. Only when all five answers lean toward the "shortcut side" should you use the shortcut pattern.
The conditions for using the shortcut pattern are intentionally strict. Even if you feel uncertain when looking across all five questions, choosing the standard pattern is the default. Feeling uncertain is a sign that there are still elements that could pull toward the foundation side.
The Five Questions
1. Does this task introduce a new premise into the design?
- Yes → Standard pattern
- No → One vote for the shortcut side
A "premise" (here meaning: an underlying rule or policy that serves as the basis for design decisions) is something whose addition can affect later decisions as well. Adding "apply this rule to this task" also changes the decision criteria for other tasks. Tasks that add a new premise are best suited for the three-round review where all seven internal roles contribute their perspectives.
2. Does this task change an existing premise?
- Yes → Standard pattern
- No → One vote for the shortcut side
Changing a premise carries more impact than adding one. Previous decisions are built on top of that premise, so a change can alter the basis of past decisions as well. Any task that changes a premise is always confirmed using the standard pattern.
3. Is the scope of impact for this task reversible (correctable after the fact) or limited?
- Yes (reversible or limited) → One vote for the shortcut side
- No (irreversible action or wide-ranging) → Standard pattern
Tasks that can be corrected allow you to address problems after they appear. On the other hand, tasks that cannot be undone, or whose impact extends broadly to other areas, require accuracy to be raised before the fact.
4. Is there a similar precedent, and have the blind spots (perspectives that are easy to overlook) already been identified?
- Yes (precedent exists, blind spots already known) → One vote for the shortcut side
- No → Standard pattern
If a precedent exists, the perspectives that came up in past reviews can be reused. However, even with a precedent, a task does not shift to the shortcut side unless the blind spot review has been done thoroughly.
5. Does this task involve any irreversible actions (publishing, sending externally, deleting, and so on)?
- Yes → Standard pattern
- No → One vote for the shortcut side
Information that has gone outside cannot be recalled. The same applies to deleted data. Tasks involving irreversible actions call for the standard pattern regardless of how the other four questions are answered. This is treated as the strongest criterion among the five.
If Even One Question Points to Standard, Choose Standard
If even one of the five questions pulls toward the "standard pattern," we do not use the shortcut.
This design is intentionally weighted toward the safe side. The shortcut pattern can save time, but the condition for using it is that all five questions confirmed the shortcut side. We do not choose it based on a feeling that "it will probably be fine."
The same applies when you feel uncertain. Tasks that leave you uncertain still have something in them that could pull toward the foundation side. That is exactly when the standard pattern's three-round review works most effectively.
We have built in a conservative default (the basic setting of choosing the safe side when in doubt): "when in doubt, choose the standard pattern." Changing this requires first building a track record of all five questions confirming the shortcut side.
Using the Five Questions as a Log
Running every decision through the five questions leaves a record of the reasoning behind each one.
Records build up such as: "Question 2 pulled toward the foundation side, so we chose the standard pattern" or "all five pointed to the shortcut side, so we used the shortcut pattern." This log can be used when checking for consistency across similar tasks. If the decision changed between one task and the next, having a record makes it possible to check which question had a different answer.
The checklist is a tool for moving pattern selection from instinct to documented judgment. Each time it is run, more reasoning accumulates and a trail of how decisions were made is left behind.