Under the current plan, mainframe SMEs are only consulted if the Java team reports a defect. What edge cases are most likely to be missed under this approach?
Select an answer to reveal the explanation.
Short Explanation
Waiting for a defect report only catches the loud failures. It's the quiet, wrong-answer cases that need someone who knows what right looks like, so you need SMEs involved before that.
Full Explanation
The correct option is cases where the transformed code runs without error but silently produces a result that differs from the legacy program's intended business outcome. Compile errors, startup failures, and exceptions are all visible technical failures the Java team will naturally notice and report; a silently wrong business result is not, and only SME review of intent reliably catches it. The option about cases where the transformed code fails to compile is incorrect because a compile failure is loud and technical, and does not require business-meaning input to detect. The option about cases where the transformed service fails to start is incorrect for the same reason, because a startup failure is an obvious technical signal, not a silent one. The option about cases where a unit test throws an unhandled exception during a build is incorrect because an unhandled exception during a build is also a visible technical failure that the development team would already notice. For the exam, defect-triggered SME involvement misses silent business-logic errors specifically, which is why ongoing involvement matters.