Q&A from the WEBINAR: “What is New in Recent Releases (6.1 and 6.2) of TopBraid EDG?”

Knowledge Graphs, Q&A, Blog

We fielded several questions as part of our recent webinar (recording and slides available here): What is new in recent releases (6.1 and 6.2) of TopBraid EDG? 

Questions from the webinar included:

Q1: I have just upgraded to 6.1.1 base version ontologies, taxonomies and crosswalks. I do not see a datagraph option. Is this a separate package?

Yes, data graphs require “Custom Defined Asset Collection Types” add-on.

Q2: Can the tabbed interface enable switching between multiple models? This is useful in Composer, but I have not found a way to access multiple models in DEG.
EDG.

You can switch between different asset collections by using the “hamburger” menu to the left of the collection’s name. This will open a selected collection in a new browser tab.

Q3: It would be helpful if you mentioned the version number and package options required for demonstrated functionality. I do not see many of the options that you are demonstrating. If you organized the demo by basic packages, and which packages are required for each functionality that you are showing.

We considered organizing the demo by a package, but found it difficult to do since many of the features apply across packages. It is possible that you are not seeing some of the features because they became available in 6.2 or because they are only available when using new, GraphQL forms. To enable GraphQL forms in 6.1, go to Server Administration -> Server Configuration Parameters. In 6.2, they are enabled by default. If you download the files used in the demo, you will see for each item what release it is included in and what collection types it applies to.

Q4: Is the datagraph module free?

It requires “Custom Defined Asset Collection Types” add-on which is available for one fixed price fee irrespective of the number of users.

Q5: Irene mentioned the ID of a wikidata state. In general, would this be the URI of an external instance?

This would be a local ID (slug) of the class of external resources. Class is needed in order to provide a list of properties that user may want to copy for each instance.

Q6: Perhaps a useful sample Glossary would be a glossary of all the terms you use in EDG, such as lineage, facets, and all the asset collection terms. In other words, this would be a nice place to look for your definitions for all these terms.

Interesting idea. We will consider it.

In the meantime, there is a page in the documentation that explains some of EDG terminology http://wiki.topquadrant.com/display/master/Terminology.

Q7: Do you currently support or are you planning to support asset types that are based on the HL7 Fast Healthcare Interoperability Resources (FHIR) standard?

If FHIR is available as an ontology, then you can load it in EDG and then use Data Graphs to create instances. 
 
If the ontology you are using does not provide SHACL declarations, you will need to use Transform -> Convert OWL Axioms to SHACL Constraints to automatically create them. For an explanation of the OWL-to-SHACL mapping, see TQ Blog entry: From OWL to SHACL in an Automated Way. For even more exact and up-to-date details, see the rdfs:comment items in the  owl2shacl.ttl workspace file (path: <EDG-WORKSPACE>\projects\TopBraid\SHACL\).
 
You will also need to declare classes for which you want to define instances as either public or protected classes of the ontology. This can be done just once for a root class. This is information is entered in the GraphQL Schema section of a class form. Please take a look at the demo example we are providing with this webinar. Select Release Concept class in the ontology. Also, for additional technical information see https://www.topquadrant.com/graphql/shacl-graphql.html.

Q8: CRUD on Graphs?

Yes, this is available through GraphQL and/or SPARQL.

Q9: Are the governance role configured views enforced?  Or are they just default, and someone in the role can still access other views?

They are default for now, and we are interested in getting more feedback on how they work for users.