Re: vCard confusion and RDF insufficiency

Garret Wilson wrote:

> May I point out just a few things:
> 
>    * We're using RDF because we think it's a good way to model
>      information about the world.
>    * Names are very common things in the world, but they are also very,
>      very, very simple things. If you think modeling names is hard, try
>      modeling the concept of a "contract" in common law countries or
>      even in civil law countries. But contracts and the like are things
>      that we want to model in RDF. If you can't model a little old name
>      in RDF, you're screwed.

I'm a little worried about this problem too. FWIW, I'd feel more 
comfortable going with your suggestion in this case. Or Harry's latest 
suggestion may be fine as well; am agnostic on this, even as I note the 
general problem you point out.

More generally, I do think sometimes when you bump up against what can 
seem an insurmountable problem like this, you have to reassess.

For example: with the bibliographic work, contributors present similar 
difficulties in the sense that we don't know the cardinality, and they 
may be ordered, or not.

Rather than fight with rdf:Seq and lists (which I did for a long time!), 
then, we just model a Contribution class. It turns out to be pretty 
easy, amenable to queries, and so forth. It also is a lot more flexible.

Likewise, Ian Davis once suggested something for the resolution of the 
FOAF name discussions which in retrospect is quite wise: forget about 
ordered name parts, and think about names for different contexts:

<http://lists.usefulinc.com/pipermail/foaf-dev/2005-April/008092.html>

I don't think names are as simple as they appear.

...

> The problems were having with single/multiple cardinality and order; and 
> representing literals in a list; are nothing new---they are problems 
> inherent in RDF. They rear their heads in very, very, simple ontologies. 
> And they aren't going away. That worries me greatly.

But the problems are also there in programming languages (is an object 
attribute a string, an array, or an object reference?), RDBMSes (an 
unordered data model), and so forth, so they're not entirely unique to RDF.

So what's the problems in need of fixing?

1) Lists: the RDF/XML syntax and SPARQL need to be brought in line with 
the model. You cannot reasonably query lists with SPARQL, and Literals 
don't have the same syntactic sugar advantage in RDF/XML that resources do.

I like the idea of allow rdf:li in collections, BTW.

2) containers: I've heard people say they should never have been in the 
RDF specs to begin with? What's the endgame with these? Should they be 
deprecated? Should new vocabularies use them? If yes, when?

This doesn't help us now though.

Bruce

Received on Thursday, 26 July 2007 16:26:25 UTC