W3C home > Mailing lists > Public > public-webapps@w3.org > July to September 2009

Window Modes - best practice and scalability

From: <richard.tibbett@orange-ftgroup.com>
Date: Mon, 17 Aug 2009 11:19:33 +0200
Message-ID: <355A518BC0575547B2A3D6773AAF8EEF2FF125@ftrdmel1>
To: <public-webapps@w3.org>
Hi,

We have been going through the latest Early Draft [1] of the Window
Modes spec and I have followed the latest discussion that I could
discern from the mailing list ([2], [3]).

There was a request for feedback on the window modes spec and I hope
this feedback is useful and timely. 

Initially we were looking at inputing distinguishing mode
characteristics wrt to widget design (i.e. constraining width and height
ranges for the four different widget modes, to allow for normalised
widget design). However, I understand that this is perhaps too
restrictive for us to include at this stage, especially considering the
wide range of display resolutions in use across the tech world and the
pace of display development that is occuring (e.g., the recent
introduction of HD resolutions across devices).

Regarding W3C widget design then and based on the current spec [1], it
seems that the following is the best practice if a developer wants to
create scaling widgets for different devices/resolutions (per
window/widget mode):

/* General Mini Mode Rendering Settings */ 
@media all and {widget-mode: mini} {

}

/* Mini Mode Rendering specific to 'size 1' */ 
@media all and {widget-mode: mini} and min-width:150px and max-width:
300px {

}

/* Mini Mode Rendering specific to 'size 2' */ 
@media all and {widget-mode: mini} and min-width:100px and max-width:
150px {

}


The above could be a good method and/or using good, standard,
liquid/fluid/elastic CSS design can go to some lengths to mitigate the
rendering inconsistency of different devices. 

Is this best practice the generally held opinion of the WebApps group?


My question though is whether we can go one further in our specs to
allow for the scaling of widgets that have been developed at a certain
resolution (with or without the best practice above) to be re-scaled
according to the current device display.

Looking around I don't see any *consistent* methods for scaling a web
page/widget on different devices just yet, and I haven't come across any
related standard even though there are what seem to be a few proprietary
(?) approaches: [4] [5] & [6].

Approaches [5] & [6] follow a 'viewport' approach and have wider
adoption than [4]. [5] & [6] are currently in use in e.g., Opera
Mobile/Mini and Mobile Safari (if anyone could point me to a standard
for this that would be useful...and this may negate the need to include
it in our own specs). Some examples:

<meta name="viewport" content="width = 320" /> 

<meta name="viewport" content="initial-scale=2.3, user-scalable=no" /> 

<meta name ="viewport" content="width=device-width">


So, just wondering if this has been considered/discussed, whether anyone
has a different view of the best practices above for
designing/developing a widget and whether we want to include some
details on window mode design practicalities with some window mode
scaling provision in the spec.

Thanks,

Rich


[1] - http://dev.w3.org/2006/waf/widgets-wm/Overview.src.html
[2] - http://www.w3.org/2009/06/10-wam-minutes.html#item03
[3] -
http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/0729.html
[4] - http://msdn.microsoft.com/en-us/library/ms890014.aspx
[5] -
http://developer.apple.com/iphone/library/documentation/AppleApplication
s/Reference/SafariWebContent/UsingtheViewport/UsingtheViewport.html
[6] - http://dev.opera.com/forums/topic/250293


Safari HTML Reference: Viewport Definition:
http://developer.apple.com/DOCUMENTATION/AppleApplications/Reference/Saf
ariHTMLRef/Articles/MetaTags.html#//apple_ref/html/const/viewport



----------
|        |   Rich Tibbett
|        |   Orange Labs UK :: Service Evolution :: Research Engineer
| orange |   E: richard.tibbett@orange-ftgroup.com 
----------   

"Learning is not compulsory... neither is survival" - W Deming
Received on Monday, 17 August 2009 09:20:15 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 18:49:33 GMT