Re: Semantifying RDF graphs

Dear Antoine:

Thanks for your promising contribution!

I think we should join our efforts in making April Fools' Pranks more understandable to machines and extend them so that information is given well-defined meaning, better enabling computers and people to laugh in cooperation. Also, we should be radically change the way April Fools' Pranks work today, resulting in true data amusement as well as improved funniness.

IMO, there are several design alternatives that fit the Architecture of the WWW:

1. Defining a URN for the type April Fools Prank

So basically taking a declarative approach:

<https://www.emse.fr/~zimmermann/Ontologies/rdfgraph/> rdf:type <urn:example:april_fools_prank>.

Some people think that it is funny that the type URI will never be dereferenceable.

2. Publishing a SHACL Shape for April Fools Pranks

ex:AprilsFoolsPrankShape a sh:NodeShape ;
	sh:targetClass schema:CreativeWork ;
	sh:property [
		sh:path schema:datePublished ;
		sh:hasValue "--04-01"^xsd:gMonthDay
	] .

This would allow the automatic detection of creative works that are likely April Fools Pranks. One may have to add a funniness score property, since not all creative works released on April 1 can be assumed to be funny.

Also, I am unsure whether SHACL implementations can handle xsd:gMonthDay properly.

3. Funny RDF BNodes

I tried to find out whether two April Fools' Pranks represented as RDF Bnodes are actually the same April Fools' Prank.

First I thought that was funny in itself. But even after applying OntoClean meta-properties of rigidity and identity I was unable to solve this. Is being an April Fools' Prank a rigid property? E.g., do they remain funny on April 2? 

I felt asleep from exhaustion and in my dreams, multiple blank nodes representing Chris Welty and Nicola Guarino started an extremely sharp argument on each others' identity criteria. I awoke with my sheets soaked wet, trying to keep the +U unity meta-property for my mind.


;-)

A happy first of April to everybody, and thanks for a smile in these challenging times!

Cheers
Martin




> On 1. Apr 2021, at 10:02, Antoine Zimmermann <antoine.zimmermann@emse.fr> wrote:
> 
> Dear SemWebbers,
> 
> 
> It is "2021-04-01T10:02:42.42+02:00"^^xsd:dateTime and it's sunny outside, so it is a good time to submit to you my progress on semantifying RDF graphs.
> 
> RDF graphs are syntax. We want to produce *knowledge* graphs from them. I set up a tool that semantify RDF graphs: the RDF2RDF converter!
> 
> https://www.emse.fr/~zimmermann/RDF2RDF/
> 
> See an example here:
> 
> https://www.emse.fr/~zimmermann/RDF2RDF/do/?uri=http%3A%2F%2Fdbpedia.org%2Fresource%2F%C3%89cole_nationale_sup%C3%A9rieure_des_Mines_de_Saint-%C3%89tienne
> 
> It is using the rdfgraph ontology (https://www.emse.fr/~zimmermann/Ontologies/rdfgraph/) to describe RDF graphs in RDF.
> 
> You can use the result of the conversion as input, giving a more meta flavour to it.
>

Received on Thursday, 1 April 2021 10:07:41 UTC