Re: [xml-dev] JADN

This sounds exciting! If one just considers this passage from the draft:

"Numerous data definition languages are in use. JADN is not intended to replace any of them, but serves as a Rosetta stone to facilitate translation among them."

And, related to this, if one considers that the draft starts with a thoughtful distinction between information model and data model. Isn't this exactly the direction to take?

I would like to make a couple of remarks.

(1) Target document tree structure

In many projects I have found it extremely useful to have a tree-structured representation of the target document structure - a machine readable document model which mirrors document structure (obtained by denormalization of the original model). This supports model queries otherwise very difficult to implement (e.g. "All data paths leading to 'foo' items?"), and, more importantly, it is the ideal substrate for transformation into many other artifacts, including tree-structured document content statistics, document-to-document transformators and documentation. The JADN spec might acknowledge the importance of such a representation and perhaps include a normative definition. This would encourage tool building very effectively. (More thoughts along that line and some implementation in [1]).

(2) SHACL

You could draw inspiration from SHACL [2], the new schema language for RDF - a beautiful piece of work, clean and rigorous.

(3) Extending Rosetta towards RDF (and graph)

JADN might even be translatable into SHACL, imagine - writing JADN schema which models trees and graphs at the same time! I believe it is possible. (More along that line in [3]).

Kind regards,

Hans-Jürgen Rennau

[1] https://xmllondon.com/2017/xmllondon-2017-proceedings.pdf , "Location trees enable XSD based tool development"

[2] https://www.w3.org/TR/shacl/ , "Shapes Constraint Language (SHACL), W3C Recommendation, 20 July 2017"

[3] http://archive.xmlprague.cz/2018/files/xmlprague-2018-proceedings.pdf , "Combining graph and tree: writing SHAX, obtaining SHACL, XSD and more"

Am Freitag, 12. April 2019, 05:15:00 MESZ hat Toby Considine <Toby.Considine@gmail.com> Folgendes geschrieben:

Within the OpenC2 community (OASIS) , there is a yearning for a schema language for JSON users. They are writing JSON Abstract Data Notation (JADN).

JADN is an information modeling language based on the CBOR data model. It has several purposes, including definition of data structures, validation of data instances, providing hints for user interfaces working with structured data, and facilitating protocol internationalization. JADN specifications consist of two parts: abstract type definitions that are independent of data format, and serialization rules that define how to represent type instances using specific data formats. A JADN schema is itself a structured information object that can be serialized and transferred between applications, documented in multiple formats such as property tables and text-based data definition languages, and translated into concrete schemas used to validate specific data formats.

In particular, JADN is data that can: 1) be displayed as tables in documents like the OpenC2 language spec, 2) be validated for correctness, ensuring that the tables are correct, and 3) validate data instances, ensuring that data examples correspond to the tables.

https://github.com/oasis-tcs/openc2-jadn/blob/working/jadn-v1.0-wd01.md

The standard has some interesting and influential backers, who hope that someday JADN will see use outside OpenC2, and that it will be understood by software tools.

I post this here because of the many fine arguments over XML vs JSON over the years, which have taught me a lot, and because that Subcommittee would be improved by the participation of some on this list.