Re: Child element order differences between SVG 1.1 and SVG Tiny 1.2

Erik Dahlström:
> > It seems for container elements, however, such as <g>, the <desc>,
> > <title> and <metdata> elements can go anywhere.
> 
> Yes, the spec explicitly says this for container
> and text elements, see last paragraph of
> http://www.w3.org/TR/SVG11/struct.html#DescriptionAndTitleElements.

Ah OK.

> > In SVG Tiny 1.2, the RNG doesn’t require <rect>’s descriptive element
> > children to come first.  Is this behaviour we want to backport to 1.1
> > when publishing the Second Edition (by virtue of starting from the 1.2T
> > RNG and adding on to it)?  I think it is reasonable.
> 
> I think there's good reason for recommending <title>, <desc> and
> <metadata> to be the first child elements. This is a content
> conformance rule only. I'd like to encourage authors to put these
> elements before other child elements. So if we decide to remove that
> rule from the DTD we should put a strong recommendation for authors in
> the spec instead.

OK.

Or we could go the other way, and modify the 1.2T RNG to require the
descriptive elements to be the first children.  I don’t mind either way.
What do others think?

> SVT12 states:
> "It is strongly recommended that authors use at most one 'title' and
> at most one 'desc' element as an immediate child of any particular
> element, and that these elements appear before any other child
> elements (except possibly 'metadata' elements) or character data
> content."

That same sentence is in that last paragraph of 5.4 in SVG 1.1 you
mentioned, too.

> ...and then goes on to explain what happens when there are multiple
> <title> elements etc.

I think it’s good to call out these restrictions due to the schema in
the prose of the spec, since not everyone is going to delve into the
schema.

-- 
Cameron McCormack ≝ http://mcc.id.au/

Received on Saturday, 18 April 2009 01:06:58 UTC