- From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
- Date: Wed, 19 May 2010 18:55:27 +1000
- To: raphael.troncy@eurecom.fr
- Cc: Media Fragment <public-media-fragment@w3.org>
2010/5/19 Raphaël Troncy <raphael.troncy@cwi.nl>: > Dear all, > > I took this action last week, in order to notify all some of the changes we > have discussed regarding the output of test cases dealing with non-existing > and illegal fragments. Note that these changes have already been committed > in the Wiki by Silvia at > http://www.w3.org/2008/WebVideo/Fragments/wiki/TemporalDimension as she has > announced today. > > The resolution is as follow: in case of an empty fragment, e.g. #t=a,a where > a>=0, or an non-existing fragment, e.g. #t=a,b where a>b, we consider that a > media fragment aware UA knows beforehand that this will be an empty fragment > (resp. an non-existing fragment) and SHOULD therefore: > - a) if the UA does not have the setup data for this media, issue a Range > request with just "include-setup", and receive the setup data in a 206 > Partial Content response. > - b) if the UA has already the setup data for this media, do nothing (no > HTTP request issued). > > The situation where no request is issued might be perceived as irritative. > The rationale is that this is what currently happens when one click on a > non-existent fragment of an HTML page. > > If you disagree with this resolution, then please shout! > Next step will be to modify the WD with this modification. Anyone? > Tracker, this is ACTION-163. I can report: this has now already been committed (was working on it as you wrote the email). Cheers, Silvia.
Received on Wednesday, 19 May 2010 09:03:42 UTC