Re: feedback on rdfcore syntax changes: pls allow unqualified about=, etc

Hi

On Tue, 22 Oct 2002, Brian McBride wrote:

> At 16:28 22/10/2002 -0400, Dan Brickley wrote:
>
> [...]
>
> >According to M+S '99 REC (which has about='' examples)
>
> is that only elements in the rdf namespace?  Ah no. there is an example in
> another namespace.

Another as well, that's a shame. It seems that the resource= and about=
usage has taken off. I know of no apps that are using non-prefixed
attributes to encode RDF properties (thankfully). The about= examples do
seem to have been copied/followed though...

>
> >this is OK.
> >According to RDF Core, it isn't. The new RDF syntax spec doesn't make
> >clear why such documents are no longer considered RDF, only that they are
> >not. Perhaps there is a case based on parser complexity, efficiency etc.,
> >but I've not yet seen it made strongly enough to justify the backwards
> >compatibility hit.
>
> As I recall, a concern was about defining names not in the rdf namespace, e.g.
>
>    <foo:bar about="...">

Yup.

> We talked about deprecation but were advised against by DanC.

I remember. Seemed so reasonable at the time. Seems the cleanest way out
now, given that the '99 REC encourages a usage we don't like, and that
having two ways of writing this forever seems excessive.

> In practise
> parsers will be tolerant of this input

In practice, XML documents that are written 'about=' are not RDF/XML
documents, and are assigned no semantics by the RDF MT. Parsers may
transform these formally meaningless XML documents into similar RDF
documents and hence graphs, but I don't really want to encourage this. If
we think the XML inside XMP files isn't RDF, and is meaningless, we should
clearly say so; if we think it is RDF, ditto. Formally outlawing it but
giving the nod'n'wink to scrapers/extractors/convertors seems to be the
worst of both worlds.

>	 unless they are set to strict mode
> and that seemed like a reasonable migration strategy.  We could add a note
> to they syntax doc suggesting that parser writers be aware.
>
> On the other hand, I personally agree that is a change to M&S and perhaps
> we can find some other way to deal with it.

I hope so (eg. see DanC's suggestion). I don't like re-opening issues but
I realise I'd not until today really thought through the consequences of
this particular (early) decision.

cheers,

Dan


-- 
mailto:danbri@w3.org
http://www.w3.org/People/DanBri/

Received on Tuesday, 22 October 2002 17:47:17 UTC