- From: Robert Sanderson <azaroth42@gmail.com>
- Date: Thu, 1 Jul 2010 14:35:18 -0600
- To: public-media-fragment@w3.org
- Message-ID: <AANLkTinl7PCcuIlJ6jkggNgezrPOkqVWolcv6Rp-y34y@mail.gmail.com>
Dear all, It's great to see that the Media Fragments specification is stabilizing, and to hear at WWW2010 that there is a commitment to extensibility. As mentioned at WWW2010, we have many use cases for media fragments that are not amenable to inclusion by value within the URI. For example, a long SVG path might be used to express a non rectangular area within an image, and possibly in a time series in order to have a moving area within a video. Equally non "traditional" web resources can benefit from this approach, such as 3d objects or scenes, database slices and so forth. The proposal which we suggested last October was to add a new type of fragment, which has a URI as its value. This URI can then be dereferenced and interpreted in the context of the base resource. For example, an image with an SVG overlay fragment might be: http://www.example.com/imgs/img1.jpg#ref=http://www.example.com/fragments/frag1.svg This is a no-fuss way to include unlimited extensibility as the standard does not have to specify the contents of the resource being pointed to. It adds at most a single section to the document and no significant barriers to understanding. The original objection of being able to non rectangular images being technically impossible seems to be unfounded, given the presentation at WWW2010 at least which demonstrated a possible rendering of a video where the fragment area was highlighted, rather than generating a completely new resource. Many thanks for your consideration! Rob Sanderson (for the OAC project, http://www.openannotation.org/)
Received on Thursday, 1 July 2010 20:36:03 UTC