Re: RDF/XML shorthand for RDF reification

On Jun 26, 2008, at 10:43 AM, Bijan Parsia wrote:

> On 26 Jun 2008, at 14:35, Alan Ruttenberg wrote:
> [snip]
>> Thanks for the example.
>> I agree that care would need to be taken. It seems that what needs  
>> to be avoided are collisions.
>
> That solves one problem by introducing another. Data URIs might help.

What's the idea for using Data URIs?

>
>> Allocating unique names is a fairly common practice these days - a  
>> combination of mac address, system time in milliseconds , and a  
>> reasonably sized random salt should be sufficient to ensure  
>> uniqueness.
>
> Uhm...so what happens to my annotations? Imports? If I serialize it  
> twice I end up with distinct subjects to my annotations. Or if  
> people downstream rely on those names which are "merely"  
> syntactic...bleah!

I think the axioms get named when the first annotation is added, and  
the names persist on saving. So hard to see how two copies would  
arise. But even if they were different after each serialization, In  
what situation would you get distinct subjects for the same  
annotations visible?

As far as other people using the name, they are advertised as not  
being stable (even if they would be).
However, a persistent name might be created for them based on an MD5  
of the concatenation of the ontology URI, versionURI, and rendering  
of the  axiom and annotation in some syntax.

I do think that there is a desire to name axioms  but I'm not sure  
this is the mechanism to accomplish that. I mentioned in another  
email using an annotation to give a name to an axiom, and I might  
propose that such a mechanism moves in to the language if we don't  
get RA.

-Alan