Repository navigation
Can we make the glossary (or parts of it) normative #78
Description
The glossary is very helpful, but we can't use it for normative (parts of) documents. Can we make some of the definitions normative and get this on the rec track somehow?
If not, could we move some of the useful bits here to the WHATWG Infra spec, as it provides a bunch of text related things that we use in throughout the web platform (and specs) in a normative manner.
for example:
https://infra.spec.whatwg.org/#code-points
All that stuff is super useful.
Activity
marcoscaceres commentedon Jul 4, 2024
We would probably be interested in citing the auto (direction), string direction, and maybe base direction definitions.
xfq commentedon Jun 19, 2025
Meeting discussions:
Options: https://docs.google.com/document/d/1jGup6Z9dO0OAeSfM6zZhTeT_E5eM9lDOVWo_Ec37mn0/edit
xfq commentedon Aug 7, 2025
Regardless of how we proceed, I think we should first consider classifying the terms into normative and non-normative, and then consider the following for normative terms:
I'll think about it and come back to this issue.
xfq commentedon Aug 9, 2025
I will try to complete the transformation of the glossary in #84 / #85, and hope to transfer the document to REC-track in the next rechartering.
xfq commentedon Dec 10, 2025
One thing we need to consider: how are we going to prove interoperable implementations of definitions?
marcoscaceres commentedon Dec 10, 2025
I don’t think that matters. These are concepts most of the time.
Where it’s testable, it’s on the other specs to show conformance. But not this spec.
xfq commentedon Dec 12, 2025
We discussed this at the working group teleconference yesterday. We hope the glossary is a living CR, which may not go to REC, so there is no need to prove whether it has two independent interoperable implementations, because that is only required when exiting CR.