Fixes for “curated” branch about webref HOT 2 CLOSED

Hi @SofyaPovarova. Thanks for reaching out!

The curation process in Webref is really intended to fix (hopefully temporary) inconsistencies that emerge from time to time in the original specifications, for instance situations where the same CSS property is defined in two different specifications. It is not meant to create data that the original specification does not already expose in a machine-readable way.

To improve the quality of the data that appears in Webref, our general approach is not to do more curation in Webref, but rather to improve the specifications themselves, by raising issues or preparing pull requests against them. We try to reduce the amount of manual maintenance that we need to do to keep the data in Webref up-to-date. A side goal is also to improve the quality of the specifications. There may be other ways to achieve the same results.

Said differently, we don't want the data to "be closer to the original specifications", we want the data to be "in the original specifications".

I'll reply to the individual examples that you raised. Both of them are illustrative of improvements that could be brought to the original specifications. If you could help with that, that would be terrific! Things are slightly easier with IDL specifications, as they use Web IDL which is easy to extract. The value definition syntax is used less systematically in CSS specifications, making it harder to extract machine-readable information all the way down to values.

from webref.

Many thanks for the detailed explanation! I hadn't considered fixing value definitions in specs, but I agree that's the right approach

from webref.