Hi,

i think the pull mechanism can be very helpful for real world scenarios. In this case i'd like to add UML as a third source besides XML and JSON, and i am very interested in this subject.

However, it does not help to much regarding the original question (" Is the set of languages expressible using XML a superset of the set of languages expressible using JSON") because all depends on the pull mechanism.

We know that every UML model can be transformed in XML (called XMI, defines by OMG), and i am pretty sure that there is a way to transform XMI into JSON. Is that helpful for the original question?

Frank Steimke

Am 25.01.22 um 08:27 schrieb Hans-Juergen Rennau:

Hi Liam,

when you speak of winged trains, I've no clue what you mean.

When you speak of lossiness, I think we have very different points of view, to know - push versus pull. To speak of loss of information when mapping document node to triples makes only sense if you regard the document as an entity pushing information, the completeness of which you may reason about. But my point of view is a pull one: the mapping is driven by a semantic model into which to pull information from the document, with perfect freedom concerning what information from the document contents to use, and how to use it. To give an extreme example, the document may be mapped to a single triple, the subject of which is the document URI, the predicate the URI xyz:foo-count, and the object the integer number of <foo> elements found in that document.

Finally, when you speak of characters: I do not care about serializations, my thinking starts with an XDM document node, which, by the way, could have been obtained by parsing an XML, JSON or HTML document, or parsing a CSV file, by in memory construction from relational database results, Elasticsearch contents, etc.

Kind regards,

Hans-Jürgen

Am Montag, 24. Januar 2022, 19:00:42 MEZ hat Liam R. E. Quin <liam@f...> Folgendes geschrieben:

On Mon, 2022-01-24 at 15:59 +0000, Hans-Juergen Rennau wrote:
>
> If we accept the point of view that a set of RDF triples (R) is an
> unequivocal statement of semantics,

Um... If we accept the view that trains have wings and fly through the
water...

>  the semantics of an XML document - as well as of a JSON document -
> is implied by the specification of a mapping M of a given document
> node D to a set of triples:
>    D + M => R
> Such mapping should be specified using a new mapping language,
> consuming XDM document nodes and emitting RDF triples.

That's lossy, if you care (e.g. whether attributes were specified with
single or double quotes is significant in some systems).

Note also that an XDM instance is not guaranteed to be unique for a
given sequence of XML characters, and in general won't be - it's the
result of one particular processing chain operating on that sequence of
characters.

>  To define it would be a matter of diligence, more than anything
> else. (Given the availability of XPath.) It is a pity that the W3C
> did not take that path.

Henry Thompson at least, in the XML Core WG, explored it, but it wasn't
a very productive avenue, partly for political reasons, and partly
because he started with the XML Information Set, which puts individual
characters into their own items, so you don't get strings.


--
Liam Quin, https://www.delightfulcomputing.com/
Available for XML/Document/Information Architecture/XSLT/
XSL/XQuery/Web/Text Processing/A11Y training, work & consulting.
Barefoot Web-slave, antique illustrations:  http://www.fromoldbooks.org


_______________________________________________________________________

XML-DEV is a publicly archived, unmoderated list hosted by OASIS
to support XML implementation and development. To minimize
spam in the archives, you must subscribe before posting.

[Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
Or unsubscribe: xml-dev-unsubscribe@l...
subscribe: xml-dev-subscribe@l...
List archive: http://lists.xml.org/archives/xml-dev/
List Guidelines: http://www.oasis-open.org/maillists/guidelines.php