W3C home > Mailing lists > Public > www-svg@w3.org > September 2005

Re: [SVGMobile12] out-of-range values

From: Ian Hickson <ian@hixie.ch>
Date: Tue, 6 Sep 2005 20:55:29 +0000 (UTC)
To: Chris Lilley <chris@w3.org>
Cc: Craig Northway <craign@cisra.canon.com.au>, www-svg@w3.org, WG SVG <w3c-svg-wg@w3.org>
Message-ID: <Pine.LNX.4.62.0509062019260.25386@dhalsim.dreamhost.com>

On Tue, 30 Aug 2005, Chris Lilley wrote:
> On Saturday, August 20, 2005, 11:24:24 PM, Ian wrote:
>> On Fri, 19 Aug 2005, Craig Northway wrote:
>>>> Ok. I'm not sure exactly if this is enough (since header names do not 
>>>> change the normative conformance criteria) but I will recheck this 
>>>> when the next last call draft is released.
>>> This section only expresses conformance criteria for clamping colour 
>>> and opacity values.  There is no conformance criteria expressed for 
>>> User Agents regarding the clamping of other types of values.
>> Oh. So the text of the section _has_ been changed?
> The title has been clarified, yes.

That doesn't answer my question.

>> In the last public version, there are conformance criteria in that 
>> section that apply to more than just colour values.
> Could you point them out? Apart from the title, I don't see any.

The title is not a conformance criteria.

There are three conformance criteria in section C.5 in the last SVG Tiny 
1.2 draft. The first one is a "MUST"-level requirement (the RFC2119-term 
capitalisation is mine; the part I cut out is just explanatory and thus 
not relevant here):

 * When out-of-range values are provided, the user agent SHALL defer any 
   error checking until after presentation time, [...].

This requirement is phrased in such a way as to apply to any value 
(regardless of the title of the section).

The next two are "SHOULD"-level requirements (non-normative examples and 
commentary removed for clarity):

 * It is recommended that user agents clamp color values to the nearest 
   color value [...] which the system can process as late as possible [...].

 * Opacity values out-of-range [...] and should be clamped to the range 0 
   to 1 at the time which opacity values have to be processed [...].

These difficulties (e.g. the working group not understanding what a 
conformance criteria is), and the fact that the working group has made 
very substantial changes to SVG Tiny 1.2 in response to the large volume 
of comments received on the last draft, are the reason I will not be 
satisfied with the response I have received to my comment (or more to the 
point, to all my comments in general) until I have had the opportunity to 
review the actual new text in the context of the entire specification, and 
all its changes, in a new last call draft.

Please confirm that the SVG working group does indeed intend to follow the 
spirit of the W3C Process Document (and common sense in general) and issue 
a new last call draft of SVG Tiny 1.2 as the next SVG Tiny 1.2 publication 
so that we can check the specification no longer has the number of issues 
that were present in the last release.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Tuesday, 6 September 2005 20:55:41 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:04 UTC