Working on FediTest, we learned quickly that a simple pass/fail is not a very good report for many tests of Fediverse protocols. So our reporting is more complex.
We distinguish not only between PASS and FAIL but PASS and a variety of non-PASS conditions.
(cont'd...)
For example: there are implementations that do not conform to what the standard says is a MUST, but it's unlikely that will cause any interoperability issues.
Often they have good reasons for doing what they do, such as reporting "504 Gateway Timeout" -- a case the WebFinger standard does not consider.
How do we report that? So that's why you see the 4x4 matrix at the top of the report and the different shades of red.
(cont'd...)