search

w3c / ttml2

Timed Text Markup Language 2 (TTML2)
https://w3c.github.io/ttml2/
Other
40 stars 16 forks source link

Inconsistent treatment of feature, feature designation, feature specification. #1115

Open skynavga opened 5 years ago

skynavga commented 5 years ago

Various prose fragments exhibit inconsistencies with respect to the distinction of a feature, as a syntactic or semantic expression or capability, a feature designation, as a label of a set of constraints or requirements as applied to a collection of features, and a feature specification, as an instance of a ttp:feature element, which binds a feature designation to an expression of support for (or use) of the functionality or capabilities associated by the feature designation. The same comment applies in the case of extension, extension designation, and extension specification.

For example, the preamble of Appendix E states

This appendix specifies (1) a set of feature designations, each of which labels one or more syntactic and (or) semantic features defined by this specification, and (2) for each designated feature, whether the feature is mandatory or optional for ...

The first part (1) of this sentence of this prose clearly refers to a something called a feature, which it qualifies as "syntactic and (or) semantic".

In addition to this prose, we also have the definition [feature], which states

A syntactic or semantic expression or capability that is defined and labeled (using a feature designation) in this specification (or a past or future revision of this specification).

So, again, here, we have a characterization of the something as "a syntactic or semantic expression or capability".

There is also a discrepancy in these two cited cases regarding the labeling of such somethings. In the first citation, Appendix E, the label (a feature designation) is associated with one or more features, implying a one to many relationship between features designations and features. In the second citation, each something, i.e., a feature, is labeled, which one might infer means that each somthing is assigned a distinct label (feature designation), implying a one to one relationship between features designations and features.

Then we have the ttp:feature element, and, starting in TTML2 1e, the associated term [feature specification], both of which introduce potential challenges to understanding the overall intent of the specification.

For this issue, I propose to review all uses of these related terms and concepts, and create, as needed, additional clarifications, elaborations, and explanations of an editorial nature to help improve specification clarity and conciseness.

nigelmegitt commented 5 years ago

I'd welcome a more consistent clearer approach here, thanks for raising this @skynavga .