- From: Ian Hickson <ian@hixie.ch>
- Date: Mon, 8 Jan 2007 23:59:51 +0000 (UTC)
- To: Cameron McCormack <cam@mcc.id.au>
- Cc: public-appformats@w3.org
- Message-ID: <Pine.LNX.4.62.0701082355010.22379@dhalsim.dreamhost.com>
On Thu, 7 Dec 2006, Cameron McCormack wrote: > > 2.14 The prefetch Element > > To make sure the images are always available, it uses a couple of > prefetch elements. > > It doesn’t ensure that they will be available, because the resource > may not be available at all, or it still may not download in time before > the binding is applied. > > I’m a bit worried also that there is no normative text that explicitly > says how to process prefetch elements. There is only: > > The prefetch element can be used to list resources that may be > pre-loaded for performance reasons. > > which does not say what the UA should do when it encounters a prefetch > element if such elements are supported. Should the resource be fetched > when the binding document is loaded? Or when the binding is first > applied? This is not defined, intentionally, so that UAs can do whatever they think is best. Interoperability is not required here; UAs could in fact prefetch even without this element, it's just making their life a little easier. > Ok, so this example is a little far fetched. So are most of the > examples in this spec. They’re just examples! > > This also does not have a tone appropriate for a technical > specification. I suggest something like: > > While this example is contrived, it does demonstrate the use of the > prefetch element to increase the chance that the green.png image will > be available to display when the button with label “green” is clicked. I have received a lot of feedback suggesting that the lighter tone actually helps people read the spec. I've used the second part of your suggested text to clarify the first paragraph, but left the above "light" text. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Tuesday, 9 January 2007 00:00:00 UTC