- From: Mounir Lamouri <mounir@lamouri.fr>
- Date: Mon, 28 Jul 2014 20:01:04 +1000
- To: Bruno Racineux <bruno@hexanet.net>, public-webapps@w3.org
On Fri, 25 Jul 2014, at 14:34, Bruno Racineux wrote: > Just took a peak at the latest spec [1], since Chrome Canary breaks my > code > made for the previous spec > and I have to update to a dual screen.orientation object|string context > (It > was previously a string). > > Good to see the new 'natural' keyword and angles. > > 1. One minor comment on the "Best Practice 1" box for the phrase: > "Never assume any kind of relationship between the screen orientation > type > and the screen orientation angle" > > "Any kind" seems too strong a statement, since there is a relationship > between type and angle during the length of a browsing context/runtime as > mentioned afterwards. > > I suggest adding a "permanent" relationship qualification and/or > "cross-devices" dimension in that sentence. Fixed: https://github.com/w3c/screen-orientation/commit/d1adfa1b1419d534c8b124331a70c7a322451008 > 2. I am noting that the definition 1 and 2 of "reading current screen > orientation type" is going to require that IEMobile and iOS change the > reporting of their current screen.width and screen.height to the correct > orientation, and stop reporting portrait only. I suppose that's a good > thing > even if it might break a few scripts. Do you mean that they do not update the screen.width and screen.height when the screen rotates? -- Mounir
Received on Monday, 28 July 2014 10:01:38 UTC