- From: Marcos Caceres <w3c@marcosc.com>
- Date: Wed, 6 Jun 2012 16:02:25 +0100
- To: Tobie Langel <tobie@fb.com>
- Cc: Scott Wilson <scott.bradley.wilson@gmail.com>, Arthur Barstow <art.barstow@nokia.com>, ext Anant Narayanan <anant@mozilla.com>, public-webapps <public-webapps@w3.org>, "public-native-web-apps@w3.org" <public-native-web-apps@w3.org>
On Wednesday, June 6, 2012 at 2:58 PM, Tobie Langel wrote: > On 6/6/12 3:35 PM, "Marcos Caceres" <w3c@marcosc.com (mailto:w3c@marcosc.com)> wrote: > > > On Wednesday, June 6, 2012 at 9:51 AM, Tobie Langel wrote: > > > > > Mozilla's proposal seems to essentially target applications distributed > > > through app stores. We'd like to see a solution that also enables > > > providing meta data to bookmarked apps similar to how meta tags work in > > > iOS. > > > > > > > > I've not much experience with the iOS meta tags, so is there anything > > missing in W3C Widget's config.xml or in Moz's JSON? > > > Not in the configuration file itself (what Mozilla calls the Web > Application Manifest), but it is not specified how it is delivered to the > client in that case. Ok, well, at least we potentially have most of the bits we need. > > Would be cool if one could just do: > > > > <!-- gimme config in accepted format (xml or json) --> > > <meta config="/config"> > > > > That way, there is no need to repeat meta tags in every page... just > > repeat it onceŠ > > > Absolutely, or: > > <html manifest=/path/to/config.webapp> > > and combine appcache and config into a single format. The AppCache > manifest format works beautifully in JSON (and I'm sure it would do > equally well in XML). See how the sample manifest files provided in the > HTML5 spec[1] would look like in JSON: https://gist.github.com/2881982 yep, that could also work… though I wonder if it's too late to be swapping manifest formats. > > or doing something like a fav.ico equivalent that is loaded > > automagically. > > > I'd rather we avoided simultaneously adding an extra http request to all > the web pages in the world **and** filling server logs with garbage 404 > requests. True. Spectacularly dumb idea. Is there a way to undo what I said? :)
Received on Wednesday, 6 June 2012 15:02:58 UTC