Preliminary synopsis *

* More to follow once we have hands-on experience

Information representations in any computer system are designed by the maker of the software. He/she did that in his/her own way, focussed on performance, maintainablity, etc. But one thing is sure: it is "proprietary", so non-standard.

That is fine, and ISO 15926 doesn't intend to change that.

But we "glue" a "System Façade" to such system, in which we do store the information in an ISO-standard format. This format is defined in Part 7 of ISO 15926.

For each data element in the Data Dictionary of the User System, that we want to publish, we define a specialization of an ISO-standard template class, with most Properties "fixed".

Instances of the missing Property values are then to be taken from the applicable tuple(s). Often this mapping isn't that straightforward, and then certain (often: engineering) rules must be applied.  

NOTE - For this the upcoming RIF (Rule Interchange Format) of XCL (XML notation for Common Logic, part of ISO 24707) might be used (not sure yet).

The result is put in a Transfer File, and sent to the applicable System Façade.

In cases where a User System would need data that are stored in any Façade in the applicable CPF (Confederation of Participating Façades), then a SPARQL query can be executed. The query result then has to be mapped to make it useful for the User System.

One of the activities in the ADI Project is to work on a mapping tool. More about this when this becomes available.