- From: David Singer <singer@apple.com>
- Date: Thu, 22 Oct 2015 14:59:33 +0200
- To: Philip Jägenstedt <philipj@opera.com>
- Cc: Simon Pieters <simonp@opera.com>, "public-texttracks@w3.org" <public-texttracks@w3.org>
this seems like a decent direction. certainly non-header style blocks would be problematic, as the clock is ticking when you encounter them. header style blocks might be ok in some circumstances, but I’d rather restrict here and then only consenting adults who know they can make it work would do it in controlled environments > On Oct 22, 2015, at 9:53 , Philip Jägenstedt <philipj@opera.com> wrote: > > On Wed, Oct 21, 2015 at 4:03 PM, Simon Pieters <simonp@opera.com> wrote: >> On Wed, 21 Oct 2015 15:00:30 +0200, Philip Jägenstedt <philipj@opera.com> >> wrote: >> >>> Oh, right, as part of "the properties corresponding to the background >>> shorthand." If it's the only problematic property, explicitly >>> excluding it would be fine I think. >> >> >> Yes, but it would disallow data: URLs as background-image as well as things >> like CSS gradients, which is too restrictive. Instead I specified that URLs >> other than data: URLs just fail to resolve. >> >> https://github.com/w3c/webvtt/commit/9fdff0abcbcfc5024f103ccc8c54ff94162cd27d > > OK, that seems fine. I had never realized it was possible to have > background images in WebVTT until now, but I see that it actually > works. Is there anything useful that could be done with this, given > that you have no idea what the size of the box the background will > apply to is? > David Singer Manager, Software Standards, Apple Inc.
Received on Thursday, 22 October 2015 13:00:08 UTC