Re: Closed: Action 130 Remove postpone from resource priorities

Hi,

The use case that Brian introduces *1 , and I supplemented *2 is one of a functionality of Level Of Details *3 called in SVGWG.

*1: http://lists.w3.org/Archives/Public/public-web-perf/2013Sep/0087.html
*2: http://lists.w3.org/Archives/Public/public-web-perf/2013Sep/0103.html
*3: http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Input#Level_of_detail_control

The number of the nested and tiled SVG file resources under the loading control that we assume in Level Of Details is enormous.
#For example, the number of resources becomes around 6 million when we prepare 10 mile square map tiles (on the equator) for whole earth.

Therefore, with this use case, UA where functions for this use case are not implemented appropriately will runaway, by the waste of the computing and network resource by uncontrolled loading.

In a draft of former resource priorities, I thought that 'postpone' met the requirements of the function for this use case. On the other hand, about 'lazyload', it will be essential for the smart processing such as 'postpone' that is not specified in the specifications to be done in UA. It may cause the compatibility issues between browsers.

>http://www.w3.org/2013/11/14-webperf-minutes.html#item03
>Anne: I think we should consider starting with lazyload. For example, the user agent could interpret lazyload to mean postpone based on data.

Will such a critical use case be in the scope of resource priorities? May you call the property for such a use case a 'hint'?

Regards, 
Satoru Takagi

W3C AC Representative, SVGWG Member

KDDI CORPORATION, R&D Strategy Department,
Technology Development Division

> Hi,
> 
> On April 17, 2014 at 3:04:04 PM, Tobin Titus (tobint@microsoft.com) wrote:
> > Per agreement yesterday, postpone was removed from resource priorities:
> > http://www.w3.org/2010/webperf/track/actions/130
> >  
> > The updated spec is here:
> > https://dvcs.w3.org/hg/webperf/raw-file/tip/specs/ResourcePriorities/Overview.html  
> 
> Can I kindly ask that you please move the document to GitHub? Having it on mercurial makes it extremely difficult/inconvenient for the community and implementers to contribute changes. Could you also please add a link to the bug tracker in the spec (or just use the GH issue tracker once you move the spec)? 
> 
> I'm also still concerned about how much HTML/SVG monkey patching is happening in this spec. I think Hixie raised a similar issue. Can I ask that you remove section 4.3 as it seems to add little value. Just focusing on the attribute itself, and what it's supposed to do when used would be great. Then HTML/SVG can wire it in properly to work as intended (and even put it on the right interfaces).  
> 
> Please note that this is the second time I've raised this issue [1]. If we had an issue tracker, it would make it better to make sure feedback doesn't fall through the cracks.
> 
> Thanks!  
> 
> [1] http://lists.w3.org/Archives/Public/public-web-perf/2014Mar/0028.html
> 
> -- 
> Marcos Caceres
> 
> 
>

Received on Thursday, 24 April 2014 09:21:46 UTC