Re: Feedback about Vorbis Comment Chapter Extension

On Sat, Sep 22, 2012 at 11:45 AM, R.L. Horn <[email protected]> wrote:
> On Thu, 20 Sep 2012, Martin Leese wrote:
>
>>> (I do have to ask, why "-->"?  That's an awfully clumsy token.)
>>
>> To match WebVTT
>
> Oh...an srt legacy.  I should have seen it.
>
> That makes me wonder about the VLC position on WebVTT, particularly "only
> nested cues" files
> (http://dev.w3.org/html5/webvtt/#webvtt-file-using-only-nested-cues).  If
> they've embraced that, then they're in no position to complain, apart from
> the fact that the chapter extension spec. needs to be fleshed out a bit.
> After all, reformatting the comments for a WebVTT parser is trivial.

No, I don't think they like it very much.

> In fact, one of the criticisms was entirely spurious anyway.  The one
> about "infinite trees" applies to WebVTT, but not to the chapter
> extension.

Here's what I would suggest.

We need a mapping of WebVTT into Ogg anyway (something I've been
meaning to do for a while, but hey...)

So, what if we used WebVTT for complex, hierarchical chapters and just
went with the simple Matroska-style chapters for vorbiscomment
headers?

There will be problem then when we have both vorbiscomment headers
like this AND WebVTT in a Ogg file, but we can solve that in the
WebVTT encapsulation spec.

Cheers,
Silvia.