Consider allowing images and text in the same profile

[nigelmegitt]

Via a conversation with a contact at NHK in Japan, I learned that there is a requirement in ARIB-TT to be able to display images inline with text content, for example to insert images of company logos alongside the name. This is apparently in common usage today. However in IMSC currently mixing text and images is prohibited.

[mikedo]

This would also address symbols in common use that are not defined in Unicode. For example, the "CC" (one symbol) defined in CTA 708.

[palemieux]

Propose deferring due to lack of concrete use cases.

[nigelmegitt]

How is the use case in the issue not concrete?

[palemieux]

How is the use case in the issue not concrete?

TTWG has not received concrete evidence of presumed IMSC 1.1 users interested in deploying the feature, and providing test cases and implementations.

[mikedo]

@palemieux - How else would one accomplish the use cases above: 1) graphic symbols (logos): and 2) symbols not commonly defined in the reference fonts (e.g. 708 "CC")?

[palemieux]

How else would one accomplish the use cases above:

@mikedo I do not know. Even if there is a gap, it is however hopeless for TTWG to do the right thing without proponents of these changes actively participating, either directly or through liaison -- including providing test cases and implementations.

P.S.: Re: 708 "CC", isn't this routinely represented by the string "[CC]"? I would think that CEA would suggest adding the symbol to unicode, which is what ARIB seems to have done with its favorite symbols.

P.P.S.: Where does support of generic company logos stop being subtitles/captions, and start falling into chyroning.

[mikedo]

To clarify about logos, I am only repeating what I have independently heard from a Japanese broadcaster. I don't have any personal experience in the US with logos in text.

Although a necessary step towards having standard glyphs for "CC" is getting it added to Unicode, that is not sufficient. It has to be in commercially reference fonts which seems unlikely to happen without more work. And, for the exact same reason, the 4-character string "[CC]" has been used in practice for decades. More of a real issue is that a few 708 graphic symbols, although defined in Unicode, are not supported in the commercial reference fonts. But close substitutes are available. So, this use case is not a hard requirement. There is no fatal present condition, but allowing it would overcome any symbol barrier in the future, e.g. Yiddish symbols.

I agree that not having the proponents present and willing to engage is an issue. I cannot speak to Japanese logo requirements, nor support test materials and reference implementations. Allowing author-defined symbols seems obvious but I also cannot support test materials and implementations. So, I understand any reluctance at embarking on the spec text under these conditions.

[andreastai]

To allow both profiles in one document will certainly be more work than just saying "images are allowed in text profile documents". Currently most of the style features are prohibited in the IMSC image profile and the consequence of being present together with images need to be defined in this case.

[nigelmegitt]

TTWG has not received concrete evidence of presumed IMSC 1.1 users interested in deploying the feature, and providing test cases and implementations.

@palemieux you are setting a bar here that is higher than usual for a requirements document. The time to cull features on this basis is much later in the process. If there is uncertainty about the requirement we can note that within the requirements document.

[css-meeting-bot]

The Working Group just discussed Consider allowing images and text in the same profile imsc-vnext-reqs#4, and agreed to the following resolutions:

• RESOLUTION: Defer this requirement for the time being, absent more detail.

The full IRC log of that discussion

<nigel> Topic: Consider allowing images and text in the same profile imsc-vnext-reqs#4
<nigel> github: https://github.com//issues/4
<nigel> Pierre: A reminder that this is IMSC 1.1 requirements.
<nigel> Nigel: Right so some other version of IMSC in the future could support both image and text simultaneously.
<nigel> Pierre: Exactly.
<nigel> .. Specifically on this issue, it is supported in TTML2, the question is if it is a requirement for IMSC 1.1.
<atai2> q+
<nigel> Nigel: Playing devil's advocate here, our goal is to produce a global subtitle standard, and
<nigel> .. there's a concrete use case in existence already, in use in Japan, so we should support that feature.
<nigel> ack atai2
<nigel> Andreas: I'm not sure that an informal conversation is enough to add this to the requirements list.
<nigel> Nigel: This may not be the most diplomatic approach - it was a genuine conversation from
<nigel> .. someone who came to me as Chair and asked for a feature to be requested.
<nigel> Cyril: We would need to know more detail about the precise layout requirements - is it needed
<nigel> .. in vertical layout for example?
<nigel> Pierre: It is reasonable to ask for more information.
<nigel> Andreas: What does the requirement list mean? Is it the list of requirements we agree to as
<nigel> .. group members, or those collected from other organisations?
<nigel> Pierre: It is the consensus list of requirements to drive IMSC 1.1.
<nigel> Nigel: +1
<nigel> Andreas: Then we need to ask if this is needed for the next version of IMSC.
<nigel> Nigel: [scans ARIB-TT spec] - it looks like they're building this functionality based on SMPTE-TT.
<nigel> Nigel: Shall we defer this then until a future version of the spec, assuming we have more information?
<nigel> Pierre: Yes please go ahead.
<nigel> RESOLUTION: Defer this requirement for the time being, absent more detail.

[nigelmegitt]

Further discussion happened since the previous comment on this issue, and a generic solution to meet this requirement, at least partially, was agreed for IMSC 1.2. See also w3c/tt-reqs#15 which is closely related. This allows any image that can be placed in a font file to be referenced.