Fwd: getting language tags out of the fundamental model (ISSUE-12)

Oops, my hand slipped on send.  Forwarding...

Regards,
Dave




Begin forwarded message:

> From: David Wood <david@3roundstones.com>
> Date: May 31, 2011 12:17:57 EDT
> To: Andy Seaborne <andy.seaborne@epimorphics.com>
> Subject: Re: getting language tags out of the fundamental model (ISSUE-12)
> 
> On May 31, 2011, at 12:04, Andy Seaborne wrote:
>> 
>> On 31/05/11 15:17, Sandro Hawke wrote:
>>> 
>>> I'm happy with the rdf:string-{Lang} datatype design, but if that seems
>>> inelegant to you....
>>> 
>>> On Fri, 2011-05-27 at 12:32 -0500, Pat Hayes wrote:
>>>> Now we are proposing to bury one of them inside a URI to get rid of
>>>> it. I would vastly prefer that we simply accepted that some literals
>>>> have more than one string, and adapt our notion of literal typing to
>>>> accommodate to that fact, rather than trying to disguise it or pretend
>>>> its not true, and so become obliged to swallow some clearly artificial
>>>> notion (such as a language tag being a kind of datatype) just to
>>>> preserve what is in any case a purely arbitrary model of literal
>>>> typing.
>>> 
>>> In that vein, I think the *clean* thing to do with language tagged
>>> literals is to get them out of the fundamental model.  RDF can model
>>> anything, so it can certainly model strings with language tags.
>>> Anything else is an optimization, I think, put in place for folks who
>>> think language tagged strings are so common they need special support.
>>> Then the question is what they really need (conceptual simplicity for
>>> humans, nice syntax, efficient machine processing, ...?), and what does
>>> the least damage to anything else....
>>> 
>>> In other words, we could say "foo"@bar is syntactic sugar for something
>>> like [ a rdf:LinguisticExpression; rdf:language "bar"; rdf:value "foo"].
>>> I know that doesn't address everything, but it has pretty much the same
>>> problems everything else does being modeled in RDF.  :-)
>>> 
>>>    -- Sandro
>> 
>> Modelling everything at a very fine grained level moves the burden on to the application.
>> 
>> c.f. RDF containers and collections.
> 
> 
> Conditionally, yes.  It would only arise when language tags are used.  Most strings do not use language tags.
> 
> The question is, IMO, whether the benefit of fixing the equivalences between RDF strings is worth the pain to be experienced by users of language tags in this context.  *Personally* I would rather query the above pattern than need to guess whether a string is a plain literal or a language tagged string or an xsd:string.
> 
> Regards,
> Dave
> 
> 
> 
>> 
>> 	Andy
>> 
>