You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jena.apache.org by "A. Soroka" <aj...@virginia.edu> on 2016/01/31 20:49:23 UTC
(un)coincidental dataset descriptions in assemblers
I’ve got a question about a remark of Andy’s here: https://github.com/apache/jena/pull/123#issuecomment-177246334
"In the Fuseki case, we want shared datasets descriptions, that is, same name, to yield the same dataset."
Wouldn't we rather use actual URIs to refer to dataset descriptions to make them coincide? Is the use of names as inverse functional properties to do this a historical artifact of the way that assembler RDF evolved, or was there some reason to do it this way instead of using something more like "same URI = same thing”?
---
A. Soroka
The University of Virginia Library
Re: (un)coincidental dataset descriptions in assemblers
Posted by "A. Soroka" <aj...@virginia.edu>.
Ah, I misunderstood your use of the word “name” to mean an rdfs:label or something like that.
---
A. Soroka
The University of Virginia Library
> On Feb 1, 2016, at 7:26 AM, Andy Seaborne <an...@apache.org> wrote:
>
> On 31/01/16 19:49, A. Soroka wrote:
>> I’ve got a question about a remark of Andy’s here: https://github.com/apache/jena/pull/123#issuecomment-177246334
>>
>> "In the Fuseki case, we want shared datasets descriptions, that is, same name, to yield the same dataset."
>>
>> Wouldn't we rather use actual URIs to refer to dataset descriptions to make them coincide? Is the use of names as inverse functional properties to do this a historical artifact of the way that assembler RDF evolved, or was there some reason to do it this way instead of using something more like "same URI = same thing”?
>>
>> ---
>> A. Soroka
>> The University of Virginia Library
>>
>
> A name is something that identifies. IFPs are one way but where the "name" here is the subject of the description (which can be a blank node or URI).
>
>
> fuseki:dataset <#dataset> ;
> <#dataset> rdf:type ja:RDFDataset
>
>
>
> fuseki:dataset [ rdf:type ja:RDFDataset ;
> ...
> ] ;
>
> Andy
>
Re: (un)coincidental dataset descriptions in assemblers
Posted by Andy Seaborne <an...@apache.org>.
On 31/01/16 19:49, A. Soroka wrote:
> I’ve got a question about a remark of Andy’s here: https://github.com/apache/jena/pull/123#issuecomment-177246334
>
> "In the Fuseki case, we want shared datasets descriptions, that is, same name, to yield the same dataset."
>
> Wouldn't we rather use actual URIs to refer to dataset descriptions to make them coincide? Is the use of names as inverse functional properties to do this a historical artifact of the way that assembler RDF evolved, or was there some reason to do it this way instead of using something more like "same URI = same thing”?
>
> ---
> A. Soroka
> The University of Virginia Library
>
A name is something that identifies. IFPs are one way but where the
"name" here is the subject of the description (which can be a blank node
or URI).
fuseki:dataset <#dataset> ;
<#dataset> rdf:type ja:RDFDataset
fuseki:dataset [ rdf:type ja:RDFDataset ;
...
] ;
Andy