Update 1.2.3 and 1.2.5 re Audio Description to match their Understanding sections

[nigelmegitt]

Arising from #2441 - the text in §1.2.3:

An alternative for time-based media or audio description of the prerecorded video content is provided for synchronized media, except when the media is a media alternative for text and is clearly labeled as such.

and §1.2.5:

Audio description is provided for all prerecorded video content in synchronized media.

and the definition of audio description:

narration added to the soundtrack to describe important visual details that cannot be understood from the main soundtrack alone

implies normatively that some alternative must be provided to pass the criteria. It is clear from the Understanding section, and from the notes in the audio description definition, that this is not intended, but the normative text does not match the intention.

The goal of these sections is that when playing the media there is a non-visual mechanism that allows the content of the video to be perceived: it is clear that audio description is merely one technique for achieving that goal. If, as allowed for in the definition of audio description's notes and in the relevant Understanding sections, there is already a non-visual mechanism present, such as descriptions in the main audio soundtrack, then no additional alternative to the visual content is required.

My preferred fix for this is quite big, so may not be convenient or palatable: it is to restructure the wording of these sections to specify the desired goal, not the technique, and have audio description as an alternative audio track be one technique, and descriptions in the main soundtrack be another technique. Further techniques such as a timed text track including the descriptions would also be listed.

An alternative, possibly less disruptive fix would be to amend the existing wording to:

§1.2.3:

An alternative for time-based media or audio description of the prerecorded video content is provided for synchronized media, except when the media is a media alternative for text and is clearly labeled as such, or when the video content is already described within the audio soundtrack.

§1.2.5:

Audio description is provided for all prerecorded video content in synchronized media, where the video content is not already described within the audio soundtrack.

I believe these changes would bring the normative text back in-line with the intent and the Understanding sections.

[mbgower]

@nigelmegitt

It is clear from the Understanding section, and from the notes in the audio description definition, that this is not intended

I believe this is support to read "...that this is intended"?

[GreggVan]

no I think he did mean it is not intended. But what was intended or not (ie did they intend to require an alternative track or to have it be EITHER an alternative track or the main track - similar to open or close captions) is not relevant since WCAG 2 is closed at this time

what we need to figure out is what should be true in WCAG3

and I think what we write for WCAG3 should be

that audio description should be offered EITHER an alternative track or in the main track - similar to open or close captions)

[mbgower]

But what was intended or not... is not relevant since WCAG 2 is closed at this time

To clarify, the WCAG 2 Task Force has processes and a mandate to do several things:

• provide updates to informative guidance
• provide errata to clarify/correct normative guidance
• explore normative errata which due to differences of opinion are rated as class 3 changes

We already have materials that fit into each of these categories.

In the case of Audio Description, there is clearly disagreement on what is normatively required. For some people, a language change we propose is merely clarifying what they believe is already normatively present. For others, the same change may be regarded as changing the existing requirement. As a consequence, almost any change we proposed (in either direction) will be a class 3 change. But that does not mean the Task Force or Working Group is excluded from proposing changes to get alignment and clarity.

[bruce-usab]

implies normatively that some alternative must be provided to pass the criteria

I don't think that's correct, but I agree with @GreggVan (et al.) and that it is a worthwhile exercise to contemplate how the SC might have been written.

An alternative, possibly less disruptive fix would be to amend the existing wording to...

I like your proposal, but I would put the "already described" bit at the front. That would be something like:

For time-based media, at least one of the following is true:

• important visual details can be understood from the main soundtrack alone; or
• additional narration is available between pauses in dialog; or
• there is an alternative for time-based media; or
• the media is an alternative for text and clearly labeled as such.

and

For all prerecorded video content in synchronized media, either the video content is already described within the audio soundtrack or audio description is available, unless the prerecorded video content is a media alternative for text and is clearly labeled as such.

[GreggVan]

@mbgower

I think we have info to make your job easier

I have since learned (Thanks AWK) that we have SUFFICIENT techniques for meeting the SC both by
a) adding an alternate audio track and
b) just putting the AD in the main audio track.
(analogous to open and closed captions)

As a result -- we have documentation as to what the intent was. That it was intended to mean that the descriptive audio could be EITHER in the main audio track OR in an alternate audio track (since both are listed as sufficient for meeting the SC)

HENCE - that make this a purely informative note -- and something easily within your mandate as I understand it.

Suggested Note

NOTE: As documented by the sufficient techniques for the SC, the audio describing the visual information can be provided via an alternate audio track, or it can be provided directly in the main audio track. This is analogous to closed and open captions.

(the last sentence can be included as instructional - or left off as desired.

[nigelmegitt]

@bruce-usab I like your proposal too.

---

I believe this is support to read "...that this is intended"?

@mbgower as @GreggVan noted, I did mean "not intended" as written.

---

@GreggVan an informative note will not address the issue, which is that the normative text is wrong, or at least misleading. My goal here is to achieve normative text that says what is intended, which I believe it does not do now. There is already adequate informative text here - if you don't think there is, that would be another issue, or potentially you could reopen #2441.

---

There is a wider problem here which makes it important to address this issue: §1.2.3 now reads (to me, anyway) as implying that to achieve accessibility in video media content providers must tick the box "has an additional alternate audio track including audio description", leading to a target setting approach of wanting to achieve 100% audio description. But this would be entirely the wrong result, and a disservice to users. Audio description is often a "sticking plaster" approach to address failings in the original media production, and it would be better all round if more content were made that did not require an additional audio track, where possible.

An unrealistic and idealistic target would be 0% audio description, as an alternate audio track, because in all cases the main soundtrack is already accessible. In reality the proper target is somewhere >0% and <100%, because we do not have production and scripting techniques that allow all genres of content to be accessible without an additional alternate audio description track.

[bruce-usab]

@GreggVan is your Suggested Note an addition for Understanding or proposed as an additional note for the WCAG TR doc?

@nigelmegitt I do not think there is anything like consensus that the normative text is wrong. Even “misleading” I don’t think is quite the right adjective. It’s misunderstood and there is disagreement about normative text requires. It’s also a fair argument that the bar should be higher.

[nigelmegitt]

I do not think there is anything like consensus that the normative text is wrong.

@bruce-usab There is certainly not consensus that the normative text is correct!

[GreggVan]

@nigelmegitt
There is a difference between "the text is wrong" and "the text is being read by some people wrongly"

The text is correct. It is indeed not as well worded as it could be -- and can be misread -- and should be worded differently in the future. But there is no evidence that it is wrong. The wording does not say that a separate audio track is required.

AND - if one looks at the sufficient techniques - there is a general technique that was created to specifically say that if the AD is in the main track - no additional track is needed.

audio description
narration added to the soundtrack to describe important visual details that cannot be understood from the main soundtrack alone

it says narration added to THE soundtrack -- not "an alternate" soundtrack. The use of "main" soundtrack later can create confusion - but as noted - the sufficient technique makes this clear that it does not need to be separate.

so while it can be worded better - it is not wrong.

That said -- if we are making editorial changes to WCAG2 - then I would suggest the following changes to clarify this and other controversies.

audio description
narration added to the soundtrack where there are gaps in the video dialog to describe important visual details that cannot otherwise be understood from the soundtrack alone
NOTE: the narration can be added to the default audio track or to another audio track selectable by the viewer.
NOTE: the amount of audio description in a video is limited by the availability of suitable gaps in which to add it, and usually does not include all important visual details.

Since this does not change the definition of AD but only makes it clearer - it would be an editorial change.

[nigelmegitt]

@GreggVan indeed there is a difference. In this case I am arguing that the text is wrong. Here's how I parse the text alongside the definition:

An alternative for time-based media or audio description of the prerecorded video content is provided for synchronized media, except when the media is a media alternative for text and is clearly labeled as such.

and the definition of audio description:

narration added to the soundtrack to describe important visual details that cannot be understood from the main soundtrack alone

1. There's a "main soundtrack".
2. An extra thing is provided that is supplementary to that main soundtrack.
3. The extra thing is either timed text describing the video content or an audio track that includes the main soundtrack and some more narration.

I'm not sure how you can read it any other way, from the normative text alone. It's fine to have additional informative text like the sufficient techniques and the "understanding" sections, but the normative text basis has to agree with them, and here, it does not.

it says narration added to THE soundtrack -- not "an alternate" soundtrack. The use of "main" soundtrack later can create confusion - but as noted - the sufficient technique makes this clear that it does not need to be separate.

Right, but the §1.2.3 text has already made clear that an additional resource needs to be provided.

Since you concede that "main soundtrack" causes confusion, it seems like a point of agreement that there are legitimately multiple ways to read this, and that's a bad situation for normative text, particularly if it's referenced in legal contexts.