Re: [SVGMobile12] "6.7 Property inheritance"

On Friday, May 20, 2005, 1:44:20 PM, Ian wrote:


IH> Section 6.7 Property inheritance is very confused.

IH> * It should be titled "Property inheritance and computation", or the 
IH> section should be split into two, since it currently covers both.

We agree and have renamed the section accordingly.

IH> * It contains the statement "The definition of each property indicates 
IH> whether the property can inherit the value of its parent", which is 
IH> meaningless, since properties can always inherit the value of the parent
IH> element.

Agreed and removed.

IH> * The statement "In SVG, as in CSS2, most elements inherit computed 
IH> values" is incorrect, as in CSS2, inheritance is always from the computed
IH> value. The sentence following that one is therefore wrong as well.

Agreed and fixed.

IH> * Computed values of absolute lengths should just be an absolute length,
IH> there should not be any specific units associated with them. Percentages
IH> should not be computed to absolute lengths; percentages should compute to
IH> percentages (computation doesn't depend on layout). In general I would 
IH> recommend linking straight to the CSS2 definition of computation, since it
IH> already covers all these cases.

Computation of percentages in SG does not depend on box model layout in
any case.

We did consider linking to the CSS2 definition of computation, but found
a number of problems. Missing out the examples, which are of course
informative, it says:

>> Some values are inherited by the children of an element in the
>> document tree. Each property defines whether it is inherited or not.

>> To set a "default" style property for a document, authors may set the
>> property on the root of the document tree. In HTML, for example, the
>> HTML or BODY elements can serve this function. Note that this will
>> work even if the author omits the BODY tag in the HTML source since
>> the HTML parser will infer the missing tag.

>> Specified percentage values are not inherited; computed values are.

The first sentence is not something that can be conformed to. The second
one is fine. The next paragraph is entirely about irrelevant workarounds
for historical HTML parsing. The last sentence is fine. So overall, its
a fairly small amount of normative content.

IH> * Each property should define its own computed value, in the style of the
IH> "Computed Value:" line found in CSS specifications.

We agree, have reviewed all the property definitions, and ensured that
they all define their computed values.

Please let us know shortly if this comment does not satisfy the comments
you made above.

IH> * The links are to an obsolete version of CSS2; CSS2 has sinced been 
IH> revised. CSS2 revision 1 has many old errors corrected. CSS2.1 is at the
IH> same stage as CSS2.0 in the REC track -- namely, the stage after last call
IH> (note that CSS2.0 never exit the then-equivalent of CR since it was 
IH> published before the formal CR process was established).

We are unable to believe that a Working Draft is the same level of
maturity as a W3C Recommendation.

However, based on your comment, we made a last call comment on CSS 2.1
asking for greater clarity from the CSS WG on whether CSS 2.0 is planned
to be rescinded or not.



-- 
 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 W3C Graphics Activity Lead
 Co-Chair, W3C Hypertext CG

Received on Tuesday, 8 November 2005 14:17:20 UTC