- From: Simon Pieters <simonp@opera.com>
- Date: Wed, 26 Jun 2013 01:44:18 +0200
- To: "public-texttracks@w3.org" <public-texttracks@w3.org>, "Rick Eyre" <rick.eyre@hotmail.com>
On Wed, 26 Jun 2013 01:14:33 +0200, Rick Eyre <rick.eyre@hotmail.com> wrote: >> The DOM construction rules are only relevant for getCueAsHTML(), not for >> rendering the cues. The rendering rules don't use a DOM at all and don't >> mention "title" attributes or tooltips. > > Okay I see. > > What's the reason for having getCueAsHTML() separate from the actual CSS > specifications applied to nodes when rendering them as subtitles? Why > not just use getCueAsHTML() for subtitles as well? IIRC Hixie didn't want to define the rendering in terms of a DOM because it would be cheaper to create CSS boxes directly, or some such. I'm not sure it was a great idea though. >> The spec is unambiguous, as far as I can tell. The voice isn't rendered >> by >> default, and there's no tooltip. >> >> I think the voice is basically supposed to be a semantic version of >> colored lines, which are commonly used for separating voices. However >> it's >> currently up to the author to provide CSS to get the colors. >> >> We could change the spec, of course, if there are good reasons. > > My first thought is that there should be some kind of default display > for voice text. If it's something that people are going to be wanting to > use a lot, which I would think it is, then it would make sense to have > some kind of default that people don't have to worry about it. I don't really disagree. However, if the default is something that people *don't* want, then that might be worse, since authors might not figure out how to undo it and instead avoid using voices. -- Simon Pieters Opera Software
Received on Tuesday, 25 June 2013 23:42:34 UTC