Re: [codec] draft-ietf-codec-oggopus and "album" gain

[Ron <[e-mail]>]

On Thu, Aug 14, 2014 at 10:24:48AM -0700, Timothy B. Terriberry wrote:
> >In section 5.2.1:
> >-   If an encoder wishes to use R128 normalization, and the output gain
> >-   is not otherwise constrained or specified, the encoder SHOULD write
> >-   the R128 gain into the 'output gain' field and store a tag containing
> >-   "R128_TRACK_GAIN=0".  That is, it should assume that by default tools
> >+   The encoder should assume that by default tools
> >     will respect the 'output gain' field, and not the comment tag.  If a
> >     tool modifies the ID header's 'output gain' field, it MUST also
> >     update or remove the R128_TRACK_GAIN and R128_ALBUM_GAIN comment tags
> 
> This was causing confusion because it suggested that an encoder should
> _always_ prefer to put the track gain into the 'output gain' field. In
> reality we used to have text saying that using the album gain for 'output
> gain' was preferred, and this was just an alternative (the leading "If"
> clause being the the important part of the sentence, not the SHOULD).
> However, this sentence is entirely redundant with the text from the
> definition of the 'output gain' field: "An encoder SHOULD set this field to
> zero, and instead apply any gain prior to encoding, when this is possible
> and does not conflict with the user's wishes." Removing the sentence avoids
> the confusion.
> 
> I've applied both sets of changes locally (using "An encoder" instead of
> "The encoder"), and they'll be included in the next update (after WGLC).
> 
> I also moved the "assume by default" sentence to the end of the paragraph
> and removed the preceding paragraph break, as the warning about being
> required to update the tags when the header value changes now logically
> follows from the declaration that these tags are relative to that field.

So the patch applied for this part is now:

> - If an encoder wishes to use R128 normalization, and the output gain is not
> - otherwise constrained or specified, the encoder SHOULD write the R128 gain
> - into the 'output gain' field and store a tag containing "R128_TRACK_GAIN=0".
> - That is, it should assume that by default tools will respect the 'output gain'
> - field, and not the comment tag.
>   If a tool modifies the ID header's 'output gain' field, it MUST also update or
>   remove the R128_TRACK_GAIN and R128_ALBUM_GAIN comment tags if present.
> + An encoder should assume that by default tools will respect the 'output gain'
> + field, and not the comment tag.

And I wonder two things about it:

Is that intended to be "An encoder SHOULD assume"?

And if we do want this to be normative, do we need to expand a little on
what 'tools' and 'respect' means there - since that's possibly now less
clear as a statement which stands on its own, rather than a(n intended)
clarification to the text that is being removed.

I agree with the rationale of what's intended by this change, but I'm
less sure if it's now clear to a first time reader whether 'tools'
includes both players and editors, and under what conditions the
assumptions about 'default behaviour' will apply (since if we have the
comment tags, obviously *something* might respect them sometimes).

If we're going to give (normative) guidance there, it possibly wants
to be a little more specific about what exactly it applies to.

  Ron