In a 52-page EN-GB automotive manual, the source alternates between N·m and Nm; I’d standardize to N·m per ISO 80000-1, but the product UI in the same booklet shows Nm — would you normalize in translation or mirror the UI? memoQ 10.7 flags 118 instances, so I need a consistent rule before I proceed.
I’d ‘mirror the UI’ and use Nm throughout the booklet for torque, then note the ISO 80000-1 N·m preference to the client — no one torques a wrench over the dot. In memoQ, add a TB entry marking Nm as preferred and a regex QA rule to flag N·m so you can batch-fix the 118 instances; does that work for your reviewer?
Split it: use ‘Nm’ only where it mirrors the UI, and ‘N·m’ everywhere else per ISO 80000-1 to avoid the ‘nm’ ambiguity. In memoQ 10.7, add a TB note and a QA regex to flag \bNm\b outside UI snippets and lock those snippets; with ‘118 instances’ it’ll save you time. Ref: SI Brochure - BIPM — are the UI labels actual strings or screenshots?
Quick sanity check: before standardizing, export a sample page with N·m and make sure the middle dot doesn’t ghost in PDF or cause a wrap — some fonts drop it. If it’s solid, use N·m in body and quote the UI as “Nm”; if not, mirror the UI and flag the SI preference to the client. @amy58’s split makes sense; do you control the final PDF?