Re: entities in SVG (middle term view)

On Monday, October 10, 2005, 12:16:55 PM, Robin wrote:

RB> Salut Andre!

RB> Andre M. Winter - Vectoreal.com wrote:
>> btw, as DTDs seem to get replaced by relaxNG (and as this is another 
>> planet for me), will there be a need of referencing those RNGs from an 
>> SVG file header or will there be an implicit link to it thanks to the 
>> namespace and version declaration in the SVG root element (assuming 
>> using only standard SVG and no extensions)? from my point of view
>> 
>>    xmlns="http://www.w3.org/2000/svg" version="1.0"
>> 
>> should be enough...

RB> There is no standard way of referencing a RelaxNG schema from an 
RB> instance document. The reason for this is because it's considered bad 
RB> practice to associate tightly a document with a schema, since in fact 
RB> depending on the situation one could want to apply different schemata 
RB> from the one the author thought would be the best.

Right. Some editors use processing instructions to do this. Others use a
table of namespaces and start elements.

RB> So yes, the fact that the document begins with an <svg> element in the 
RB> "http://www.w3.org/2000/svg" namespace is sufficient to indicate that 
RB> it's a standalone SVG document (in fact, it's already sufficient today, 
RB> though many still believe one has to specify the DOCTYPE).

RB> It is also likely that we'll be putting an RDDL (http://rddl.org/) 
RB> document at http://www.w3.org/2000/svg that will point to the RelaxNG, 
RB> but that may require that we wait for the TAG to come to some decision 
RB> on it.

No, it doesn't. The TAG, after spending a lot of time discussing it and
even more time trying to design a RDDL 2 that no-one seemed to want,
then noticed OWL as well and decided there was no one true namespace
document format any more. So, people who want to use RDDL 1.0 can carry
on doing so.

Robin is correct that the svg namespace will probably contain a RDDL 1.0
document at some point. But the main constraint is getting time to write
one.


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

Received on Monday, 10 October 2005 16:35:18 UTC