- From: Robin Berjon <robin@robineko.com>
- Date: Wed, 6 Jan 2010 17:29:37 +0100
- To: <richard.tibbett@orange-ftgroup.com> <richard.tibbett@orange-ftgroup.com>
- Cc: <public-device-apis@w3.org>
Hi, it's time to reheat this issue! On Dec 15, 2009, at 10:58 , <richard.tibbett@orange-ftgroup.com> wrote: > Just wondering whether we should settle on a specific approach to > hanging certain APIs off the navigator.device object. To summarise what is in existing drafts and has been proposed, we currently have these abstract variations: 1. Service object, simple method, inside device navigator.device.dahut.graze() 2. Prefixed method, directly on device navigator.device.dahutGraze() 3. Generic unprefixed method, directly on device navigator.device.graze() 4. Service object, simple method, directly on navigator navigator.dahut.graze() I think we should eliminate option (3) because it doesn't scale (if we produce a Unicorn spec, since unicorns graze too we'll have painted us into a corner). The more I think about (2) the less I like it. It makes for a huge device object that doesn't really make much sense as a whole. So I guess the question is how much we mind polluting navigator :) Personally, I don't mind much because it's not a space in which authors normally put stuff so the risks should be low. The downside is that we don't own navigator (the HTML WG does) but I guess we can ask for their review. So I'm going to go with (4), i.e. Doug's proposal which Max already indicated support for. Any other opinions? One reason I'm asking is because once we get our first API out we'll also have to release the Core Device Interfaces, and they depend directly on this decision (I can make the change really quickly whichever option is chosen). -- Robin Berjon robineko — hired gun, higher standards http://robineko.com/
Received on Wednesday, 6 January 2010 16:30:06 UTC