Re: [w3ctag/design-reviews] Web Publications review (#344)

> Perhaps my concern was the description _generally_ of WebIDL in that way, rather than that being a description of _how WebIDL is being used in this specification_.

Yes, this might be useful to clear up. We don't specifically say in the introduction that we're not exposing this data.

> I think partial is a bad practice because it breaks the ability of a reader of the specification to look at the definition of the interface and understand what methods it has on it -- whether than reader is a user or an implementor of the specification. It's essentially a syntax for monkeypatching another specification (or another section of a specification) with no indication in that other section that it is incomplete and patched elsewhere.

We do say that the member definitions are defined in each relevant section where [we declare the `PublicationManifest` dictionary](https://w3c.github.io/wpub/#webidl-wpm).

As @iherman has already written, the idea was to provide an encapsulated view of the expected expression of each member property, rather than require the reader to refer to the [IDL index](https://w3c.github.io/wpub/#idl-index) any time they want to cross-reference the Web IDL expression.

I'm not qualified to counter you on how WebIDL use should be preferred, but personally I didn't think this was problematic when it was proposed as the WebIDL specification says:

> Note: Partial interface definitions are intended for use as a specification editorial aide, allowing the definition of an interface to be separated over more than one section of the document, and sometimes multiple documents.

That's, in a nutshell, all we were trying to achieve. Is there another way to do what we wanted with WebIDL, or are you saying that we should _only_ have the IDL index?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/344#issuecomment-491055296