Source Potato Research
Measure what software actually does.
Evidence-led investigations into the information software reads, combines, and transmits from real computers.
How we approach research
Open enough to question. Clear enough to reproduce.
Good research should earn trust by showing its work. We intend to be candid about our methods, our limitations, and what participation actually means.
Open methods
We explain what we are investigating, how evidence is gathered, what our tools can observe, and where the method has limits. Conclusions should never require blind trust.
Reproducible by design
We preserve methods, versions, timestamps, evidence standards, and test conditions so other people can examine the work, repeat it, and challenge the result.
Transparent participation
People who contribute should understand what is collected and why, see the information associated with them, download it, and choose when their participation ends.