W3C home > Mailing lists > Public > semantic-web@w3.org > May 2021

Re: A case for a datatype for identifying RDF graphs

From: Antoine Zimmermann <antoine.zimmermann@emse.fr>
Date: Tue, 11 May 2021 10:55:06 +0200
To: Martynas Jusevičius <martynas@atomgraph.com>
Cc: Semantic Web <semantic-web@w3.org>
Message-ID: <a5d35d04-55b4-fc40-c1d8-36ab6b4a1ce3@emse.fr>
Martynas,

Le 10/05/2021 à 22:51, Martynas Jusevičius a écrit :
> Doesn't this require having a canonical serialization?

No, not at all. The lexical space of the datatype can be any 
serialisation syntax, but the value space is the set of RDF graphs up to 
isomorphism. This means that (assuming the Turtle syntax and some fixed 
prefixes):

"_:b  a  owl:Thing, _:x ."^^rdf:turtle .

denotes the same thing as:

"_:x  rdf:type  owl:Thing, _:w ."^^rdf:turtle .

and as:

"[  a  owl:Thing, []] ."^^rdf:turtle .


If one wants to refer to a specific serialisation of a specific graph, 
then on ecan just use an xsd:string, or perhaps use the IRIs for mime 
types, e.g.:

"_:b  a 
owl:Thing"^^<https://www.iana.org/assignments/media-types/text/turtle>.

and if one wants to have the canonical form of a given graph in a given 
syntax, then one should probably use an external tool that computes the 
canonical form.

The point of my email is only to address a single use case: to identify 
a given RDF graph (set of RDF triples) from inside an RDF document in 
order to talk about it.


--AZ


  Even
> N-Triples/Quads can be ordered in different ways.
> 
> XML has a canonical form, but of course XML only has a single syntax.
> https://www.w3.org/TR/xml-c14n11/
> 
> On Fri, May 7, 2021 at 8:09 PM Antoine Zimmermann
> <antoine.zimmermann@emse.fr> wrote:
>>
>> To be fair, it has been attempted before:
>>
>> 1) Ivan Herman edited a draft that contained the idea (but apparently,
>> it did not really made it to completion)
>> 2) Henry just replied to my email saying that he had a similar (or even
>> the same) idea some time ago.
>>
>> I believe that this idea is in the head of several people, but for some
>> reason, no one has gone so far as to actually propose it formally,
>> implement it, advocate it, etc. to a larger community.
>>
>> --AZ
>>
>> Le 07/05/2021 à 17:56, David Booth a écrit :
>>> On 5/7/21 11:03 AM, Antoine Zimmermann wrote:
>>>> TL;DR: I propose a datatype that has values that are RDF graphs
>>>
>>> +1
>>>
>>> It's really kind of astounding that this has not been done already.
>>>
>>> David Booth
>>>
>>
>> --
>> Antoine Zimmermann
>> École des Mines de Saint-Étienne
>> 158 cours Fauriel
>> CS 62362
>> 42023 Saint-Étienne Cedex 2
>> France
>> Tél:+33(0)4 77 42 66 03
>> http://www.emse.fr/~zimmermann/
>>

-- 
Antoine Zimmermann
École des Mines de Saint-Étienne
158 cours Fauriel
CS 62362
42023 Saint-Étienne Cedex 2
France
Tél:+33(0)4 77 42 66 03
http://www.emse.fr/~zimmermann/
Received on Tuesday, 11 May 2021 08:56:11 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 11 May 2021 08:56:12 UTC