Skip to content

Require profile signaling - #10

Merged
palemieux merged 2 commits into
masterfrom
issue-5-profile-signaling
Nov 9, 2017
Merged

palemieux merged 2 commits into
masterfrom
issue-5-profile-signaling

Conversation

@palemieux

Copy link
Copy Markdown
Contributor

Closes #5

@palemieux palemieux self-assigned this Nov 9, 2017
@palemieux
palemieux requested a review from skynavga November 9, 2017 04:03

@nigelmegitt nigelmegitt left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've asked for more detail here. I'm concerned that it introduces a change that makes otherwise conformant documents non-conformant, but possibly there are ways to work around that.

Comment thread index.html Outdated
<h3>Profile Signaling</h3>

<p>The [[!ttml-imsc1.1]] profile to which an Instance Document conforms SHALL be identified using the profile signaling
mechanisms specified in [[!TTML2]].</p>

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this mean that an otherwise conformant EBU-TT-D or IMSC 1 document that does not identify the profile is required to be a non-conformant IMSC 1.1 document? Or does this line allow for the possibility of a content profile signalled by the document interchange context? If that is the intention, then we can not actually impose requirements on the document interchange context since it is by definition out of band.

@palemieux palemieux Nov 9, 2017 •

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I had not considered this possible interpretation of the requirement.

Could we say the following?

The [[!ttml-imsc1.1]] profile to which an Instance Document conforms SHALL be identified using the profile signaling mechanisms specified in [[!TTML2]], unless compatibility with IMSC 1.0.1, EBU-TT-D and SDP-US processors is desired.

This matches w3c/imsc#267 .

@nigelmegitt nigelmegitt Nov 9, 2017 •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, but since that desire is not something that can be mandated or tested, the SHALL doesn't really exist, so we can reword equivalently to:

The [[!ttml-imsc1.1]] profile to which an Instance Document conforms SHOULD be identified using the profile signaling mechanisms specified in [[!TTML2]], unless compatibility with IMSC 1.0.1, EBU-TT-D and SDP-US processors is needed.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It can be tested: unless IMSC 1.0.1, EBU-TT-D and SDP-US signaling is used, TTML2 signaling must be used.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah right, in that case you're not testing the desire (which is a human emotion as far as I know!), you're testing the presence of signalling.

However there is no requirement for any profile signalling at all in EBU-TT-D, so how does this test work?

You might say that unless there is a determination from the TTML2 profile resolution mechanism that IMSC 1.0.1, EBU-TT-D or SDP-US applies, then the document needs to include explicit profile signalling, but again, that might depend on the document interchange context, which is out of scope.

I think the best we can do here as part of the requirements is to say that content profile signalling shall be recommended. And leave it at that. Then in the IMSC 1.1 document we say it is a SHOULD and add a note that the explicit signalling of content profile can cause non-conformance with some profiles of TTML, and if such conformance is needed then the non-conformant syntax should not be included.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We could say: TTML2 signaling must be used if any TTML2 feature is used,

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, that would work - assuming you mean "any feature in TTML2 but not in TTML1".

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

assuming you mean "any feature in TTML2 but not in TTML1".

Yes.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looking at this thread, I am now thinking that the requirements should avoid getting in the weeds, and should merely say:

TTML2 profile resolution semantics and signaling should be used

This mirrors w3c/imsc#3 and the PR at w3c/imsc#267

@palemieux

Copy link
Copy Markdown
Contributor Author

However there is no requirement for any profile signalling at all in EBU-TT-D, so how does this test work?

conformsToStandard is not required?

@nigelmegitt

Copy link
Copy Markdown
Contributor

conformsToStandard is not required?

Correct. Cardinality is 0..*.

@palemieux

Copy link
Copy Markdown
Contributor Author

Correct. Cardinality is 0..*.

Ok. In that case, the profile would need to be specified through the document interchange context, just as when @ttp:profile is not specified on a IMSC 1.0.1 document.

@nigelmegitt nigelmegitt left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approved in meeting 2017-11-09

@palemieux
palemieux merged commit 2758d05 into master Nov 9, 2017
@palemieux
palemieux deleted the issue-5-profile-signaling branch November 9, 2017 17:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants