Re: What parts of a spec should be normative?

It doesn’t make sense for definitions referenced by normative content to be non-normative, nor is the presence or absence of RFC keywords relevant.

Consider for example an informative section that defines “authorized ingredients” as any kind of vegetable or fruit that weights less than 500g. 

Now consider a normative section that states that a conformant food processor MUST be able to process authorized ingredients in less than 60s.

The normative statement is meaningless without the definition. Yet the definition doesn’t contain any RFC keywords.

One way of looking at this is to assume that RFC keywords are there implicitly, and that when you write:

> an ‘authorized ingredients’ is any kind of vegetable or fruit that weights less than 500g

you’re implicitly saying something along the lines of:

> an ‘authorized ingredient’ MUST be either are fruit or a vegetable. It MUST weights less than 500g.

> These words [the RFC keywords] can be used as defined here, but using them is not required.  Specifically, normative text does not require the use of these key words. They are used for clarity and consistency when that is what's wanted, but a lot of normative text does not use them and is still normative.

Hope this helps.

—tobie

---
[1] https://www.rfc-editor.org/rfc/rfc8174

Ven 22 sept 2023, à 16:29, Jeffrey Yasskin a écrit :
> Hi editors,
> 
> I left a comment in https://github.com/w3c/vc-data-model/issues/1285 that a particular section shouldn't be marked non-normative, and the editors there disagree with me. Do we have a document somewhere that gives guidelines for what should be normative vs not? In this case it's "just statements that use RFC 2119 keywords" vs "and definitions those statements depend on", but I'm sure there have been other disagreements like this.
> 
> Thanks,
> Jeffrey