Re: Towards PROV-O Accounts

Thanks, Sandro.

I was careful to avoid the term "named graph" everywhere I could.
I used it only where there was some formal, established meaning (namely, the service description vocabulary).

I am using RDF Abstract Graph wherever possible. 
sd:NamedGraph pairs the _current_ RDF Abstract Graph with the (dcterms:identifier-ish) sd:name, whose value is used when SPARQL INSERTing.


On Jan 5, 2012, at 9:33 AM, Sandro Hawke wrote:

> On Thu, 2012-01-05 at 08:34 +0000, Graham Klyne wrote:
>> (It's kind of dated now; I use the term "formulae", from Notation3, to
>> mean 
>> roughly what we mean by named graphs.)
> And please note that the term "named graph" is deeply ambiguous, having
> completely different meanings (like night and day, or more like "night"
> and "gray") between the original paper than introduced the term and
> current SPARQL usage, and neither one is actually the intuitive meaning
> that an RDF graph is given a globally unambiguous URI name.   I avoid
> the term like the plague.
> (I'm trying to swap in this prov-o work, as much as I can, but haven't
> gotten very far yet.)
>    -- Sandro

Received on Thursday, 5 January 2012 15:07:27 UTC