Re: term name for prefix redefinition?

I agree with all of this.

On Thu, Nov 15, 2012 at 5:37 PM, Niklas Lindström <lindstream@gmail.com>wrote:

> On Thu, Nov 15, 2012 at 7:15 PM, Ivan Herman <ivan@w3.org> wrote:
> > Guys,
> >
> > I think we agreed on our last telco that, in RDFa+HTML5, a processor
> should raise a warning if a prefix definition redefines either an initial
> RDFa context entry or a prefix 'inherited' from an ancestor. (I trust Manu
> to do this change in the editor's draft.) To make things complete, we should
> >
> > - agree on a term name that I would add to http://www.w3.org/ns/rdfa#.
> I propose
> >
> > rdfa:PrefixRedefinition dcterm:description "warning; to be used when a
> prefix, either from the initial context or inherited from an ancestor node,
> is redefined in an element";
> >     rdfs:subClassOf rdfa:Warning .
> >
> > (I guess it is perfectly fine if we decide on that in an email...)
>
> + 1
>
> > - If we have an agreement on all that, we should also add some tests
> before we forget. I am not sure it is worth adding a test on whether a
> warning is issues on _all_ predefined prefix-es; we have a series of test
> cases for those, but I believe it is enough if a test looks at at least one
> predefined prefix case, that should be enough to prove that the mechanism
> works for a specific implementation
>
> Sounds good.
>
> > The slightly touch issue: formally, this extra warning is relevant for
> HTML5 only; but I do not believe that it should be a processor error if
> such a warning is issued for other cases, too. In other words, the tests
> should be added to the HTML5 branch of the test cases only, but the other
> branches should be silent...
>
> +1
>
> As you say, processors are allowed to warn about this in all
> circumstances. As 7.6 in Core says: "Other implementation-specific
> rdfa:Info, rdfa:Warning, or rdfa:Error triples may be generated by the
> RDFa Processor." And in 7.4.1 there is an explicit Note stating that
> "[in] general it is a bad practice to redefine prefix mappings within
> a document". So it's very reasonable to add this generally in an
> implementation, for all cases.
>
> But as you also say, this is speced for, and the test is for, HTML5 only.
>
> Best regards,
> Niklas
>
>
> > Thoughts?
> >
> > Ivan
> >
> > ----
> > Ivan Herman, W3C Semantic Web Activity Lead
> > Home: http://www.w3.org/People/Ivan/
> > mobile: +31-641044153
> > FOAF: http://www.ivan-herman.net/foaf.rdf
> >
> >
> >
> >
> >
> >
>
>


-- 
Shane P. McCarron
Managing Director, Applied Testing and Technology, Inc.