Re: Validating RDF Data Book

For your interest, I independently did some work on applying Augmented Transition Networks (ATNs) to RDF shape rules. ATNs were developed by AI researchers back in the 1970’s as graphical rules for natural language parsing. At the time, the researchers wanted to make the parsing rules easier for non-experts to understand as a graphical representation of recursive finite state models.

     https://en.wikipedia.org/wiki/Augmented_transition_network <https://en.wikipedia.org/wiki/Augmented_transition_network>
     https://courses.cit.cornell.edu/ling7710/readings/bates.pdf <https://courses.cit.cornell.edu/ling7710/readings/bates.pdf>

Using ATNs for RDF lends itself to graphical tools for browsing and editing shape rules.  I chose to use RDF to express the rules along with a small JavaScript library that maps them to the Graphviz dot format for asynchronous rendering with a Web page Service Worker.  You can play with this at the following:

     https://www.w3.org/WoT/demos/shrl/test.html <https://www.w3.org/WoT/demos/shrl/test.html>

The drop down menu can be used to change the presentation of rules from Turtle to a diagram. There is a delay while the very large JavaScript library for Graphviz is loaded by the service worker. You can use your browser’s view-source menu action to see the handful of JavaScript libraries used in the demo.

I have no plans to standardise this, but wanted to draw attention to the potential for graphical representations as a means to reduce the effort and learning curve involved in existing approaches to RDF shape rules.  I am hoping that someone is inspired to create graphical editing tools for SHACL and ShEx.

p.s. I am working on an extension of RDF based upon Cognitive Science to better reflect characteristics of human memory and reasoning, e.g. spreading activation and reinforcement learning of rules, and I am hoping to re-use some of the above idea for graphical representations of declarative and procedural memory. This is about evolving the Semantic Web into the Cognitive Web, and replacing the process of hand generation of RDF vocabularies by machine learning against curated collections of use cases expressed in natural language.

Dave Raggett <dsr@w3.org> http://www.w3.org/People/Raggett
W3C champion for the Web of things & W3C Data Activity Lead