Begin with a real question.
A recurring uncertainty or a specific system may reveal a gap worth examining. Commercial potential alone cannot supply the question.
Method
A useful tool needs a defensible way of seeing.
Twicefield's instrument work begins with a question, makes the method explicit, and gives it software or another usable form only when that form improves examination. The result should help someone else understand their own system without inheriting an unsupported conclusion.
Make the framework explicit.
Name the claim, evidence types, authority, relevant boundaries, and the distinctions that could change the answer.
Test a repeatable method.
Describe the steps and stopping rules so the examination can be applied with materially similar discipline by someone else.
Build only where the tool helps.
A utility, comparator, inspection surface, or connecting layer should make the method more exact, less costly, or easier to use.
Label maturity and limits.
A sketch, method, prototype, fielded instrument, reusable utility, and supported product carry different claims.
Let the work stop.
A study may end in a negative finding. A commissioned examination may answer its question. A promising instrument may remain internal.
Instrument lineage
Question.
Method. Useful form.
Question
What must be known, and for which claim or decision?
Framework
Which fields, evidence, assumptions, and authority matter?
Method
How can the comparison be performed and repeated?
Instrument
Would a tool materially improve this method? If so, what can it actually do?
Status
What has been tested, what remains unknown, and who may responsibly use it?
The boundary
A tool can make evidence easier to examine. It cannot create missing history, confer authority on weak records, or turn a reconstruction into the event itself.
A low-risk first step
A serious question can begin before its instrument exists.
Describe the question or system. It may call for an answer, a method, an instrument, or no work here.
Describe the situation