Loosen coupling between usesForced and forced bound parameter (#609).

[nigelmegitt]

Closes #609 by changing the definition of the usesForced named metadata item in the way proposed there.

[skynavga]

Remove all scoping semantics that are not document scoped.

[nigelmegitt]

I believe this change already meets the agreed relaxation of the "must be specified as a child of the head element" to express only significance - see #609 (comment)

Rather than expressing a boolean significance based on the location, it instead relates the position to the scope of significance.

[palemieux]

What use case would require usesForced to be specified anywhere but <tt>?

[nigelmegitt]

What use case would require usesForced to be specified anywhere but <tt>?

@palemieux An author may wish to indicate specific content elements for example <div>s that contain forced content, perhaps because they have been introduced or modified as part of the authoring process and require some additional quality control, for example.

The counter-case is also relevant here. Rather than leaving a possible syntactic option open to differences in interpretation which would not be interoperable, I would prefer to state a baseline interpretation now, which would be open to review.

[palemieux]

An author may wish to indicate specific content elements for example

s that contain forced content, perhaps because they have
been introduced or modified as part of the authoring process and require some additional quality control, for example.

Isnt't this true about any change? Should there be a usesStyle named metadata item? I really fail to see an actual use case for usesForced on anything but head. This is of course from the perspective of IMSC1.1. I do not object to adding it in TTML2 for potential future use, as long as it is prohibited in IMSC1.1.

[nigelmegitt]

Isn't this true about any change?

Maybe, and ttm:item is extensible to cover that.

Should there be a usesStyle named metadata item?

Nobody has asked for one so far, so no.

I really fail to see an actual use case for usesForced on anything but head. This is of course from the perspective of IMSC1.1. I do not object to adding it in TTML2 for potential future use, as long as it is prohibited in IMSC1.1.

I'm not going accept a restriction on metadata in IMSC 1.1 just because you don't see the use case for it @palemieux , when it has no impact on processing. It just isn't dangerous enough to prohibit.

However I'm not really clear about the use case for the usesForced metadata item being defined in TTML2 at all, regardless of scope. Can you or @skynavga or anyone else here recall why it was added? Could we simply remove it and resolve this discussion with "less is more"?

[palemieux]

Could we simply remove it and resolve this discussion with "less is more"?

Yes!

Can you or @skynavga or anyone else here recall why it was added?

I did not ask for it, and do not know of an outstanding requirement for it.

[nigelmegitt]

Closing in favour of #621.