Re: Argumentation Schema

Good news. A discussion resulted in an internal bug report at Google with regard to their parser. Summarily, it appears that we do not need a new specification document detailing the merging of graphs or “semantic hybridization”.

From: Adam Sobieski<mailto:adamsobieski@hotmail.com>
Sent: ‎Saturday‎, ‎January‎ ‎21‎, ‎2017 ‎8‎:‎57‎ ‎PM
To: Jarno van Driel<mailto:jarnovandriel@gmail.com>
Cc: public-schemaorg@w3.org<mailto:public-schemaorg@w3.org>, public-argumentation@w3.org<mailto:public-argumentation@w3.org>, Richard Wallis<mailto:richard.wallis@dataliberate.com>

In addition to the major parsers and online test suites (e.g. http://linter.structured-data.org<http://linter.structured-data.org/> , https://search.google.com/structured-data/testing-tool), there are also JavaScript libraries. I noticed that one, Green Turtle (https://github.com/alexmilowski/green-turtle), includes tests for combinations of syntaxes or “semantic hybridization” scenarios.

It so happens that Green Turtle’s author, R. Alexander Miłowski, is a proponent of “semantic hybridization” (http://www.xmlprague.cz/sessions2015/#semhybrid). “JSON-LD and RDFa are being promoted for use on the Web to augment and annotate information. Yet, each format has its optimal use for encoding particular kinds of information.” A publication by R. Alexander Miłowski “describes a hybrid approach where JSON-LD and RDFa can be used together to provide optimal encoding while retaining connections to document locations.” The interested reader can refer to the conference proceedings (http://archive.xmlprague.cz/2015/files/xmlprague-2015-proceedings.pdf) for the publication.

I received some advice that the best courses of action for proponents of “semantic hybridization” include the initial steps of creating a Community Group at the W3C, having enough people showing interest, to write down a full-blown, short, specification, one which crosses all the "t"-s, and to convince various implementers, tool providers, et cetera, to provide the necessary parsers around such a specification. Such a specification would include RDFa, Microdata, Turtle and JSON-LD.

From: Jarno van Driel<mailto:jarnovandriel@gmail.com>
Sent: ‎Thursday‎, ‎January‎ ‎19‎, ‎2017 ‎7‎:‎39‎ ‎PM
To: Adam Sobieski<mailto:adamsobieski@hotmail.com>
Cc: public-schemaorg@w3.org<mailto:public-schemaorg@w3.org>, public-argumentation@w3.org<mailto:public-argumentation@w3.org>, Richard Wallis<mailto:richard.wallis@dataliberate.com>

I have 3 pages on which I've got 3 different multi syntax examples implemented and so far the Structured Data Testing Tool hasn't been able to parse them successfully, nor does Google's Search Console report they understand this correctly (quite the opposite actually, it clearly shows they don't process multi syntax markup)

Google representatives have spoken out in conversations that they currently don't support the coalescing of entities nor the de-referencing of URIs in case of cross syntax statements being made.

I'm still hoping to hear whether Google has the intention to start processing multi syntax statements, though I've been warned upfront it could take a while before I get an answer to that question.



2017-01-20 1:16 GMT+01:00 Adam Sobieski <adamsobieski@hotmail.com<mailto:adamsobieski@hotmail.com>>:
I added some RDFa + JSON-LD examples which I think show a natural usage of argumentation schemas (https://www.w3.org/community/argumentation/schemas/).

<html>
  <head>
    <script type="application/ld+json">
[{"@id": "2", "http://schema.org/supports": { "@id": "1" }},
 {"@id": "3", "http://schema.org/supports": { "@id": "1" }},
 {"@id": "4", "http://schema.org/supports": { "@id": "1" }}]
   </script>
  </head>
  <body vocab="http://schema.org/">
    <span about="1" typeof="Statement" property="text">Statement 1.</span>
    <span about="2" typeof="Statement" property="text">Statement 2.</span>
    <span about="3" typeof="Statement" property="text">Statement 3.</span>
    <span about="4" typeof="Statement" property="text">Statement 4.</span>
  </body>
</html>

The RDFa + JSON-LD works with <http://linter.structured-data.org/> http://linter.structured-data.org/<http://linter.structured-data.org/but> but not with https://search.google.com/structured-data/testing-tool?url .​​ I’m working on asking the RDFa and JSON-LD teams about the status of the specification of the combination and about potentially getting the combination specified.

From: Adam Sobieski<mailto:adamsobieski@hotmail.com>
Sent: ‎Tuesday‎, ‎January‎ ‎17‎, ‎2017 ‎1‎:‎47‎ ‎PM
To: Richard Wallis<mailto:richard.wallis@dataliberate.com>
Cc: public-schemaorg@w3.org<mailto:public-schemaorg@w3.org>, public-argumentation@w3.org<mailto:public-argumentation@w3.org>

There are two topics:


  1.
argument maps
  2.
agreement/disagreement quotations with rationale

Your understanding of the use case is correct. I’m updating the examples (https://www.w3.org/community/argumentation/schemas/); I’ll utilize technical and content examples from the Wikipedia argument map article (https://en.wikipedia.org/wiki/Argument_map) and, as needed, other Wikipedia article examples of argumentation.

It could be that expected utilizations with regard to argument maps are more often JSON-LD because:

(1) “JSON-LD is a concrete RDF syntax” (https://www.w3.org/2013/dwbp/wiki/RDF_AND_JSON-LD_UseCases). (2) “JSON-LD, unlike microdata and RDFa, is completely separated from the presentation layer” (https://www.quora.com/What-is-the-difference-between-Microdata-RDFa-JSON-LD). (3) In JSON-LD, we can utilize @id to refer to items in multiple triples of arbitrary graphs and without regard to where the elements appear in a sequence of sentences or a document tree.

Some argument maps of documents can be represented by putting RDFa @about on each statement or quotation and serializing a graph with JSON-LD.

See: http://linter.structured-data.org/
See: https://search.google.com/structured-data/testing-tool?url


<html>
  <head>
    <script type="application/ld+json">
{
  "@context": "http://schema.org",
  "@type": "CreativeWork",
  "@id": "1",
  "citation": {
  "@type": "CreativeWork",
  "@id": "2" }
}
   </script>
  </head>
  <body vocab="http://schema.org/">
    <span about="1" typeof="CreativeWork">
      <span property="text">Statement 1.</span>
    </span>
    <span about="2" typeof="CreativeWork">
      <span property="text">Statement 2.</span>
    </span>
  </body>
</html>

or

<html>
  <head>
    <script type="application/ld+json">
{
  "@context": "http://schema.org",
  "@id": "1",
  "citation": {
  "@id": "2" }
}
   </script>
  </head>
  <body vocab="http://schema.org/">
    <span about="1" typeof="CreativeWork">
      <span property="text">Statement 1.</span>
    </span>
    <span about="2" typeof="CreativeWork">
      <span property="text">Statement 2.</span>
    </span>
  </body>
</html>

From: Richard Wallis<mailto:richard.wallis@dataliberate.com>
Sent: ‎Tuesday‎, ‎January‎ ‎17‎, ‎2017 ‎4‎:‎57‎ ‎AM
To: Adam Sobieski<mailto:adamsobieski@hotmail.com>
Cc: public-schemaorg@w3.org<mailto:public-schemaorg@w3.org>, public-argumentation@w3.org<mailto:public-argumentation@w3.org>

Hi Adam,

Thanks for the markup examples.  What I am having difficulty in ‘seeing’ is what a real world example would look like.

If my understanding of the use case is correct I expected to see an article, or extract from one, that was attributed to an author (plus other metadata: date, URL, etc) plus supporting or dissenting statement(s) referencing it, each with their own author, date, etc.  If my understanding is also correct the supporting/desenting statements are very likely to be published on different website(s) to the original article.

As to microdata/RDFa/JSON-LD, in all cases the data structures encoded in these formats would be identical, it is only how they would be encoded into the html that would differ.  The choice of which serialisation to use would be down to individual website developers.  This is why examples on [most] Schema.org documentation pages are supplied in all three formats.

~Richard.


Richard Wallis
Founder, Data Liberate
http://dataliberate.com

Linkedin: http://www.linkedin.com/in/richardwallis

Twitter: @rjw

On 16 January 2017 at 21:05, Adam Sobieski <adamsobieski@hotmail.com<mailto:adamsobieski@hotmail.com>> wrote:
I put some rough draft examples up at:

https://www.w3.org/community/argumentation/schemas/


My thoughts on the discussion question were that more intricate structures could be in JSON-LD <script> elements. With microdata/RDFa, the structures go atop the markup, atop the natural language; elements are utilized once in microdata/RDFa.

From: Richard Wallis<mailto:richard.wallis@dataliberate.com>
Sent: ‎Monday‎, ‎January‎ ‎16‎, ‎2017 ‎11‎:‎48‎ ‎AM
To: Adam Sobieski<mailto:adamsobieski@hotmail.com>
Cc: public-schemaorg@w3.org<mailto:public-schemaorg@w3.org>, public-argumentation@w3.org<mailto:public-argumentation@w3.org>

I am finding it difficult to see how these options would work without having some marked up example use cases to look at.

I am also a little confused by the discussion question about which microdata/RDFa and JSON-LD scenarios we should be looking at.   In Schema.org (in the vast majority of cases) the encoding syntax should not be relevant - the vocabulary should work the same for all three syntaxes.

~Richard.

Richard Wallis
Founder, Data Liberate
http://dataliberate.com

Linkedin: http://www.linkedin.com/in/richardwallis

Twitter: @rjw

On 16 January 2017 at 16:33, Adam Sobieski <adamsobieski@hotmail.com<mailto:adamsobieski@hotmail.com>> wrote:
Schema.org Community Group,
Argumentation Community Group,

Thank you for your feedback and comments so far. I’ve refactored the schemas.

https://www.w3.org/community/argumentation/schemas/


I’m exploring two approaches to modeling argument maps. A first approach is to model the relationships between statements or quotations.

Relationship — Extends Intangible<https://schema.org/Intangible>. A relationship between a subject and an object.
subject: Text<http://schema.org/Text> or Quotation<https://www.w3.org/community/argumentation/schemas/> or Relationship<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList> or URI
object: Text<http://schema.org/Text> or Quotation<https://www.w3.org/community/argumentation/schemas/> or Relationship<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList> or URI

A second approach is to model statements which extend CreativeWork and which can be interrelated.

Statement — Extends CreativeWork<http://schema.org/CreativeWork>. A statement.
supports: Statement<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList>
supportedBy: Statement<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList>
opposes: Statement<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList>
opposedby: Statement<https://www.w3.org/community/argumentation/schemas/> or ItemList<https://schema.org/ItemList>

I’ll explore how the approaches work in Microdata, RDFa and JSON-LD.

Regardless of approach 1 or 2, a topic of argumentation schemas is to convenience the expression of agreement and disagreement and to support the expression of rationale for so doing.

AgreeQuotation — Extends Quotation<https://www.w3.org/community/argumentation/schemas/>. A quotation which is agreed with.
rationale: Text<http://schema.org/Text> or ItemList<https://schema.org/ItemList>

DisagreeQuotation — Extends Quotation<https://www.w3.org/community/argumentation/schemas/>. A quotation which is disagreed with.
rationale: Text<http://schema.org/Text> or ItemList<https://schema.org/ItemList>


Best regards,
Adam Sobieski

Received on Wednesday, 25 January 2017 03:24:34 UTC