Re: Welcome to the RDF*/SPARQL* mailing list

Introduction: Gregg Kellogg. I’ve also been working with RDF for several years, particularly as editor of a number of specs [1]. I also maintain the Ruby RDF/Linked Data set of libraries [2] as well as the Structured Data Linter [3] and my own RDF Distiller [4], all based on those libraries.

While I think it’s useful to be able to make statements about statements as in RDF*, I am concerned about the N-Triples/Quads equivalent formats, which would seem to call for some form of a statement identifier to be able to stick to the one triple/quad per line format.

I think there’s a lot of value in the Notation-3 Formulae for being able to make statements about groups of statements (or, trivially, one statement) due to the ability to have a formula as the subject or object of a triple. In my own implementation, I fake this using named graphs named by blank nodes, which also appear as the subject or object of a triple, but the RDF 1.1 semantics of this are unspecified, and blank nodes come with their own baggage (existential quantifier, or internal label of a specific node in a graph?). I’d like to tackle these issues more thoroughly.

Gregg Kellogg
gregg@greggkellogg.net

[1] JSON-LD, JSON-LD API, JSON-LD Framing, CSV on the Web, Microdata to RDF
[2] https://github.com/ruby-rdf/
[3] http://linter.structured-data.org
[4] http://rdf.greggkellogg.net/distiller?command=serialize

> On Aug 29, 2019, at 11:14 AM, David Booth <david@dbooth.org> wrote:
> 
> Introduction: David Booth.  I've been involved with RDF for several years, with a focus on healthcare and life sciences.  I have also been facilitating discussion and collecting ideas for making RDF  easier to use by average developers: https://github.com/w3c/EasierRDF
> 
> I think RDF*/SPARQL* has potential for addressing a long-known gap in RDF.
> 
> My main concerns:
> 
> - It must be easy to make statements about an entire graph -- a set of triples -- rather than one triple at a time.  At present RDF* does not allow this, but my understanding is that it could be extended to do so. IMO this is critically important.
> 
> - It should be harmonized with other existing mechanisms, such as named graphs and N3's ability to talk about graphs.
> 
> Thanks!
> David Booth
> 
> On 8/9/19 9:09 AM, Andy Seaborne wrote:
>> Intro: Andy Seaborne
>> I have the pleasure of working with Richard at TopQuadrant.
>> Previous, I have worked on SPARQL and RDF (for the latter, a certain amount around syntax alignment of Turtle).
>> For RDF* I'm interested in hearing what people are using it for and how it interacts with the idea that an RDF triple is "true" regards of the rest of the graph. Making statements about triples, whether probabilities or other conditions on the triple (including (bi)temporal usage) changes that from an application POV. Maybe we are moving to a more syntactic view of RDF?
>> On the SPARQL* side, I'm interested hearing about implementation experience, for example in the balance in implementation for RDF* as it relates to data when RDF* is not present.
>> I don't think there are any fundamental problems - it is a matter of making choices and explaining them, some of which may be "best practice" around usage and modelling.
>>     Andy
>> On 06/07/2019 11:11, Olaf Hartig wrote:
>>> Hi all,
>>> 
>>> I just wanted to welcome everyone who has already subscribed to this list!
>>> 
>>> The main purpose of this list is to have a place where folks can ask and
>>> discuss questions related to the RDF*/SPARQL* approach and how to use this
>>> approach.
>>> 
>>> If you are here to help or support the work on the submission document about
>>> RDF*/SPARQL*, I will send another email later to share details about how we
>>> are going to do this. Given that the vacation period here has started,
>>> probably not much is going to happen anyways during the next weeks.
>>> 
>>> In the meantime, maybe, everyone who has joined the list may introduce
>>> themselves; who are you and what's your interest in RDF*/SPARQL*?
>>> 
>>> For myself, I simply list the documents I wrote about about RDF*/SPARQL* with
>>> a brief summary of what each of them provides:
>>> 
>>> 1/ As an introductory document I created the short position statement that I
>>> wrote for the W3C workshop in March 2019 in Berlin. This position statement
>>> provides a brief motivation and an overview of the approach, a summary of its
>>> properties, and pointers to other publications and material.
>>> http://blog.liu.se/olafhartig/2019/01/10/position-statement-rdf-star-and-sparql-star/ 
>>> 
>>> 2/ "Foundations of an Alternative Approach to Reification in RDF"
>>> http://arxiv.org/pdf/1406.3399
>>> This is the initial document that I wrote together with Bryan Thompson. The
>>> main contributions of this document are a formal definition of the RDF* data
>>> model, a mapping from RDF* to pure RDF with the RDF reification vocabulary, a
>>> definition of the Turtle* grammar, a formal definition of SPARQL*, and a precise
>>> specification of how SPARQL* (as a query language) extends the W3C
>>> recommendation of the SPARQL 1.1 Query Language.
>>> 
>>> 3/ "SPARQL* Update"
>>> http://blog.liu.se/olafhartig/documents/sparql-update/
>>> This document specifies SPARQL* Update, which is an RDF*-specific extension of
>>> SPARQL Update.
>>> 
>>> 4/ "Foundations of RDF* and SPARQL* - An Alternative Approach to Statement-
>>> Level Metadata in RDF"
>>> http://olafhartig.de/files/Hartig_AMW2017_RDFStar.pdf
>>> This document is a research paper that looks at redundancy in RDF* graphs, and
>>> shows that the reification-based mappings from RDF* to RDF and from SPARQL* to
>>> SPARQL possess two desirable properties: they are information preserving and
>>> query result preserving.
>>> 
>>> 5/ "Reconciliation of RDF* and Property Graphs"
>>> http://arxiv.org/pdf/1409.3288
>>> This document describes the relationship between RDF* and the Property Graph
>>> data model. More technically, this document provides two formal mappings from
>>> RDF* graphs to Property Graphs, and one such mapping in the other direction
>>> (PG to RDF*).
>>> 
>>> Additionally, together with some of my students, we have developed an RDF*/
>>> SPARQL* extension of the Java RDF framework Apache Jena to provide a
>>> collection of tools and Java libraries to process RDF* data and SPARQL*
>>> queries. See: https://github.com/RDFstar/RDFstarTools
>>> Moreover, we have some preliminary extension of these tools that connect RDF*
>>> and Property Graphs: https://github.com/RDFstar/RDFstarPGConnectionTools
>>> 
>>> 
>>> Olaf
>>> 
>>> -----
>>> Dr. Olaf Hartig, Docent
>>>   Associate Professor (Universitetslektor, docent)
>>>   Division for Database and Information Techniques (ADIT)
>>>   Department of Computer and Information Science (IDA)
>>>   Linköping University
>>> 
>>>   Website: http://olafhartig.de
>>>   Office: Building B, Ground floor, Room 2B:478
>>> 
>>>   Postal address:
>>>        Dept. of Computer and Information Science (IDA)
>>>        Linköpings universitet
>>>        SE-581 83 Linköping
>>>        SWEDEN
>>> 
>

Received on Thursday, 29 August 2019 21:39:20 UTC