Re: A proposal for ISSUE-104 (built-in vocabulary)

Bijan Parsia wrote:
> On 11 Jun 2008, at 14:30, Ivan Herman wrote:
> 
>> Bijan Parsia wrote:
> [snip]
>> Hm. As an RDF/XML user, probably yes. As a turtle user, well, it would 
>> be suboptimal...
> 
> Turtle can compile things behind the scenes as it wishes.
> 

I am not sure I understand. Do you mean that all turtle parsers (in 
Jena, RDFLib, Sesame, you-name-it) should be changed to generate the owl 
terms? I do not think this is realistic...

>>> (I'll note that we're in this pickle due to removing punning on 
>>> object and data properties :()
>>
>> Yep.... :-(
>>
>> Actually, Boris' mail did not refer to rdf:Seq/Bag/Alt and the 
>> corresponding rdf:_n properties. There might be users out there 
>> preferring those to Lists, in which case the same issue arises for 
>> rdf:_n. (But I am not sure at all whether we should are bout those)
> 
> Container properties are not particularly easy to accommodate. Standard 
> techniques call for an infinite number of predicates which is probably 
> not a good idea to build in to the language (at least, it could have 
> unknown ramifications). Since the general best practice for several 
> years now has been to denigrate contatiners, I think it's a barrel of 
> trouble we can safely leave unopened.
> 

I agree.

I.


> Cheers,
> Bijan.
> 

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf