Re: [VCARDDAV] How to encode X-Properties in XML?

[Peter Saint-Andre <[e-mail]>]

On 3/30/10 3:47 PM, Joe Hildebrand wrote:
> On 3/30/10 2:12 PM, "Peter Saint-Andre" <[e-mail]> wrote:
> 
>>>> - On input, the content of the XML: property MUST be interpreted as a
>>>> single XML element.
>>
>> Qualified by what namespace? My sense of the discussion so far is that
>> the single XML element contained in the "XML:" property MUST have some
>> namespace.
> 
> I'm just asking "or else what?"  

Just as we Jabberites defined the "urn:xmpp:example" namespace, I
suggest that we define an example namespace in the urn:ietf:* tree and
throw these offenders in there ("urn:ietf:params:xml:ns:example").

> What are the consequences of being
> non-conformant, particularly when it can cause loss of information on
> round-trip.

That sounds like an appropriate consequence to me. :)

>>> No.  I'm suggesting that we put these extensions in a new namespace,
>>
>> "DAV:" perhaps? ;-)
> 
> Not funny in several directions at once.  Nicely done.
> 
> (Note to those who don't get the important one of the jokes, DAV: is a
> prefix, not a namespace.)
> 
>>> preferably a new namespace per spec that adds 1+ properties.
>>
>> Why are we defining this bucket in the first place? It seems to me that
>> we want to be encouraging people to use the extensibility they have with
>> XML, not redefine the "urn:ietf:params:xml:ns:vcard-4.0" namespace by
>> creating new elements in that namespace outside the schema we've defined.
> 
> Agree that we should be using XML for extensibility, and agree that we don't
> want those elements to go into urn:ietf:params:xml:ns:vcard-4.0.  For
> example, draft-daboo-vcard-service-type might use
> urn:ietf:params:xml:ns:vcard:service-type.  If that I-D defined multiple
> properties, they could all use that namespace URI.  The
> ...vcard:service-type namespace then becomes the bucket.

Rightio.

Peter


-- 
Peter Saint-Andre
https://stpeter.im/