How the research works
Four rules I work to. All four are visible in use in the logs.
Facts and conclusions stay separate
Every log entry sets out what a source actually supports, then what I concluded from it. Opinion is labelled as opinion.
Every source records who produced it
Government, academic, vendor, campaign group. Tagging them shows when the evidence for a claim comes mostly from people with something to gain from it.
Looking for what would prove it wrong
Anything that could change the project's direction is tested against the evidence that would undermine it, not just the evidence that fits.
Nothing gets quietly changed
Corrections are new entries that point back at what they replace. The reasoning stays readable, mistakes included.
How a claim earns its place
- A claim worth using turns up
- Go to the original source, and record who produced it
- Look for what would prove it wrong, not only what backs it up
- Log what the source actually says, and what it does not
- Use it, with the citation. Or retract it, in the open.
The path every claim takes before I rely on it.
None of this is aspiration. The logs carry retractions, corrected figures and claims that could not be verified, all left where they are. A project asking for better evidence from government should be able to show its own.