RE: Reification - whats best practice?

Err I am not sure were the data problem is

if using a table or data object that each triple is stored in
Adding a Reification field, say for a "who said" field  is only a
reference to a witness/author data object.

The reference itself does not explode the data size, and the witness
objects should be reasonably finite.


Am i missing somthing?
Lisa

> -----Original Message-----
> From: www-rdf-interest-request@w3.org 
> [mailto:www-rdf-interest-request@w3.org] On Behalf Of Eric Jain
> Sent: Thursday, August 26, 2004 11:39 AM
> To: Leo Sauermann
> Cc: 'RDF interesting groupe'
> Subject: Re: Reification - whats best practice?
> 
> 
> 
> Leo Sauermann wrote:
> > My email was about RDF/XML as storage space. No statement table.
> 
> Sorry then, I may have misunderstood you. The largest inefficiencies 
> that I have encountered so far are with tools that need to expand a 
> single statement into four triples in order to be able to 
> state anything 
> about the statement. As I pointed out, this issue can and has been 
> solved by some tools.
> 
> But your problem, if I understand you correctly now, is that 
> you need to 
> provide information on large sets of statements. This is my 
> problem as 
> well :-) However, I consider it a separate issue.
> 
> One partial solution is to put statements with common metadata into a 
> single file, and then say:
> 
>    <rdf:Description rdf:about="">
>      <dc:date>2004-06-08</dc:date>
>      <cc:license 
> rdf:resource="http://creativecommons.org/licenses/by-nd/2.0/"/>
>    </rdf:Description>
> 
> Obviously this solution is not flexible enough for all cases.
> 
> 
> > BTW Real life:
> > 99% of all this Reification will be used to state things like:
> > - When was this triple added to the store? (date)
> > - By Whom ? (chown leo)
> > - who has access rights (chmod 777) to it?
> 
> This is only handled in a more compact way by contexts if this 
> information applies to fixed sets of statements. Correct me 
> if I am wrong.
> 
> If you are familiar with scientific data, you may know that 
> you need to 
> provide references to literature to back just about anything you say. 
> This is just one use case where provenance information is given at a 
> very fine-grained level, and with a lot of overlapping.
>