Research & novel technology
From a novel idea to a testable prototype
Choose the uncertain claim, build the smallest useful experiment, and distinguish a result from a release.
Start with the uncertain claim
A novel technology project starts with a question whose answer affects whether the product is worth building. Write that question in terms you can test.
For example: can this approach turn a specific input into a useful output within an agreed time and error tolerance? Define the input, comparison, and pass criteria before choosing an implementation. This is an illustrative research question, not a claim about an existing Racoben system.
Keep the experiment bounded
Choose representative samples you have permission to use. Include difficult cases and a simple comparison approach. Record the environment and conditions so someone else can reproduce the check.
Make the first prototype narrow enough to expose the uncertainty. A polished interface is useful when it tests a user interaction; otherwise it may hide whether the underlying approach works.
Report what the result supports
Explain what was measured, what passed, what failed, and what remains unknown. An experiment on samples supports a conclusion about those conditions. It does not establish that a production service is secure, reliable, or commercially viable.
The next stage may be another experiment, a pilot, or stopping. Stopping an unsuitable approach can be a valuable outcome when the decision is supported by evidence.