August 2026
On reporting intervals
Why a fixed sampling interval flatters slow processes and penalises bursty ones, and what to do about it when you cannot change the collector.
Applied research, at a readable scale.
North Harbour Labs is a small independent research practice. We take on questions that need a careful answer rather than a fast one, and we write up what we find so that someone else can check it.
What we work on
Measurement and instrumentation, mostly. How a thing is counted usually matters more than the count, and that part is often left undocumented. A lot of our work is going back to the definition and asking whether it survives contact with the data.
We publish short notes rather than long reports. If a finding fits in two pages, it gets two pages.
Recent notes
June 2026
Baselines that drift
A short note on comparison baselines that quietly move, and a cheap check that catches most of the cases.
March 2026
Small-sample intervals in practice
What we actually do when the sample is too small for the method the textbook recommends.