search

w3c / ttml2

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

Issue 1204 positioning example #1207

Closed nigelmegitt closed 4 years ago

nigelmegitt commented 4 years ago

Closes #1204 by adding a note and an example.

During w3c/ttwg#107 @skynavga objected to moving the region specification in the last example from the body element to the div element and instead suggested adding an explanatory note.

This pull request does that, and also adds a further example document showing content in two regions, overlapping temporally. The sample renderings are generated by sandflow/imscJS based on the source added at spec/images/sources/example-complex-layout.ttml - this necessarily includes both tts:origin and tts:position because that library currently does not support tts:position (since it is not part of IMSC). The example in the specification document omits the tts:origin.

nigelmegitt commented 4 years ago

@skynavga please could you be more specific about the issue with the note? Is the problem only with the second part, for example? What did you have in mind as an alternative when you proposed this compromise in our call on Thursday?

As for the additional example, this is an informative example in an informative section, so it would be helpful to know more about why you think it is a bad idea to add it, given that it has no substantive impact and does, in my opinion, help to address the implementer feedback I received. That feedback included a statement that the examples are typically treated as examples of common practice, and are therefore misleading. My previous proposal was to amend the existing example, and you also objected to that.

skynavga commented 4 years ago

@nigelmegitt upon additional consideration, I have concluded that (1) the current example does not warrant any comment since it only makes use of @region on body; had it also made use of @region on a descendant, then a comment would be in order; (2) the comment attempts to paraphrase a quite complicated piece of semantics in a introductory context where such details are intentionally excluded and doing so would only serve to confuse the reader; (3) the TTML2 specification is not a user guide, but a technical specification, and, as a consequence, authoring information is intentionally minimized;

nigelmegitt commented 4 years ago

@skynavga I'm not sure, but it seems that you are dismissing the issue rather than accepting a proposal to resolve it?

Absent a better user guide, implementers do in fact read the introduction hoping to glean some kind of usage hints: that was the feedback I received that caused me to raise #1204. As things stand, the examples do not give a realistic sense of any typical usage that I'm aware of. That being the case, it seems reasonable to make some change to address the point.

If you feel that the note is going too far, I'd accept removing it if we could agree to add the new example, since the new example pulls the previous threads together in a "here's what it might look like in practice" sense, i.e. it provides what those readers are hoping to find, without going too far down the road of producing a user guide.

skynavga commented 4 years ago

@nigelmegitt my larger issue is that this is not a typo fix, and this is not a wide review comment resolution; this is a new issue that is proposed to be resolved in CR at a point in the process after we have agreed to only entertain typo fixes and wide review comments; so if you want to move this to 3ED, then feel free to do so

nigelmegitt commented 4 years ago

this is not a wide review comment resolution

Ah, this is where we differ. We are not in the Process phase where we must demonstrate wide review, i.e. prior to entering CR, but comments are still welcome, as stated in the CR SOTD, and this is exactly the kind of implementer feedback that we need to accept and respond to positively.

Arguably this feedback was primarily provided against TTML1, but to my mind the best value in terms of future-facing fixes is to fix within TTML2. This also gives the opportunity to show how, for example, using tts:position is a good choice for region positioning.

So I am treating this issue as a wide review comment.

skynavga commented 4 years ago

@nigelmegitt this issue is clearly not related to a change made in 2nd edition, or even in TTML2, it is clearly not related to a substantive bug, and it is clearly not a typo fix; therefore, it is an attempt to introduce a change that is outside the scope of what can be considered a response to the normal review process of a CR; as such, I find it to be NIGO (not in good order) as a wide review comment, and object to it for consideration in 2ED at this time

nigelmegitt commented 4 years ago

what can be considered a response to the normal review process of a CR

This is predicated on the idea that the only valid comments are those made to this specific version of the document. No such qualification is made in the SOTD where the invitation to comment is provided.

nigelmegitt commented 4 years ago

@palemieux @cconcolato @skynavga please could you re-review given the changes from 12th May?

@skynavga I could not derive any actions that would resolve your objection, other than not merging, but perhaps you could look again and consider reconsidering?