This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
(Feedback from Christian Vogler) IIUC, the PAC specifies the row number not as an absolute number but relative to the previous row number. So, the percentages in the table should be added on to a previous "line" specification, rather than be an absolute number. Similarly, the PAC indent position is also a relative number given on top of an existing indent, so positions need to be added on.
Is this in response to the comment I made about positioning of roll-up captions, at http://dvcs.w3.org/hg/text-tracks/raw-file/default/608toVTT/608toVTT.html#roll-up-captions ? If so, this referred to the positioning of the proposed region. Right now the proposal reads: Region: id=rollup width=80% height=n scroll=up regionanchor=0%,100% viewportanchor=10%,90% start=bottom Instead, it needs to be: Region: id=rollup width=80% height=n scroll=up regionanchor=0%,100% viewportanchor=10%,yy% start=bottom where yy is the percentage taken from the most recent PAC, which is calculated according to the table in http://dvcs.w3.org/hg/text-tracks/raw-file/default/608toVTT/608toVTT.html#positioning-in-cea-608 I don't think the PAC base row is relative in CEA 608 ever. We also need a note stating that as soon as a new PAC appears, the entire region is supposed to be repositioned at the new PAC, which isn't easy to support with the current proposed region specs. CEA-608 specifies that the entire content of the region needs to be moved, but I'm not sure how prevalent that is in practice, as opposed to just starting with blank content in the captions after the move.
Ah ok. Thanks for clarifying. Regions currently cannot be re-positioned, but I know we want this feature. I think we can do so by adding region changes inside the WebVTT file, e.g. WEBVTT Region: id=fred width=80% height=3 regionanchor=0%,100% viewportanchor=10%,90% scroll=up 00:00:05.940 --> 00:00:10.610 region:fred WHEN I GET A SICK BIRD, 00:00:07.040 --> 00:00:11.700 region:fred THAT JUST STOPS EVERYTHING Region: id=fred viewportanchor=10%,10% 00:00:09.410 --> 00:00:12.910 region:fred FROM MOVING FROM MY PLACE 00:00:10.610 --> 00:00:14.100 region:fred TO ANYWHERE ELSE OR BEYOND This has, however, other implications. E.g. it makes encapsulation of the file more difficult (e.g. the WebVTT in WebM spec).
We should also clarify that a region move - irrespective of how it ends up getting done in practice, if at all, happens only if the row of the PAC is different from the previous one. Cue some ancient CEA-608 cruft here: in roll-up captioning style, the PAC specifies the baseline of the region, as well as the indent of the current line, which *usually* stays constant from one caption to the next, but I don't think there is a promise about this, so it depends on the specifics of the captioning file that is being converted. The safest, but not semantically most meaningful, way to convert might be to handle the indent on a line-by-line basis. The indent can be calculated via normal means described in the positioning section. The conversion procedure would need to check, though, if the row has changed relative to last time. If not, just indent and proceed as usual with the cue. If yes, reposition the region.
You know, if it's just for converting canned files, we can simply create another region to handle change in positioning: WEBVTT Region: id=fred width=80% height=3 regionanchor=0%,100% viewportanchor=10%,90% scroll=up Region: id=fred2 width=80% height=3 regionanchor=0%,100% viewportanchor=10%,10% scroll=up 00:00:05.940 --> 00:00:10.610 region:fred WHEN I GET A SICK BIRD, 00:00:07.040 --> 00:00:11.700 region:fred THAT JUST STOPS EVERYTHING 00:00:09.410 --> 00:00:12.910 region:fred2 FROM MOVING FROM MY PLACE 00:00:10.610 --> 00:00:14.100 region:fred2 TO ANYWHERE ELSE OR BEYOND == As for the indent: it is possible to have a cue indent inside a region. So, we don't need to handle any additional indents on the region level.
Hmm. If we do that, what happens to the text displayed in the first region? It needs to be deleted from the screen at some point.
(In reply to comment #5) > Hmm. If we do that, what happens to the text displayed in the first region? > It needs to be deleted from the screen at some point. Yes, indeed - we would need to split that cue and replicate it in the other region. Not a good solution.
(In reply to comment #1) > Is this in response to the comment I made about positioning of roll-up > captions, at > http://dvcs.w3.org/hg/text-tracks/raw-file/default/608toVTT/608toVTT. > html#roll-up-captions ? > > If so, this referred to the positioning of the proposed region. Right now > the proposal reads: > > Region: id=rollup width=80% height=n scroll=up regionanchor=0%,100% > viewportanchor=10%,90% start=bottom > > Instead, it needs to be: > > Region: id=rollup width=80% height=n scroll=up regionanchor=0%,100% > viewportanchor=10%,yy% start=bottom > > where yy is the percentage taken from the most recent PAC, which is > calculated according to the table in > http://dvcs.w3.org/hg/text-tracks/raw-file/default/608toVTT/608toVTT. > html#positioning-in-cea-608 I've tried to apply this idea, but I don't think it's necessary. The indent can continue to be on the cue and not on the region. The idea is that it's possible for a cue to have an indent within the region that may not be used by other cues in that region. So, I've made a clarifying note for this. > We also need a note stating that as soon as a new PAC appears, the entire > region is supposed to be repositioned at the new PAC, which isn't easy to > support with the current proposed region specs. CEA-608 specifies that the > entire content of the region needs to be moved, but I'm not sure how > prevalent that is in practice, as opposed to just starting with blank > content in the captions after the move. I've also specified that a move can be replicated by creating a new region and copy the text and that it's only necessary if the PAC base row changes. https://dvcs.w3.org/hg/text-tracks/rev/653204801736