Richard Vines and I have just released a paper entitled “Interoperability and the Exchange of Humanly Usable Digital Content Across a Global Economy.” The paper has been published at both an Australian and a US web site. It discusses how to achieve interoperability in XML dialects, i.e. how to transform or translate XML-based digital content from one XML language to another, within one overarching transformation architecture. Some approaches to this problem reflect the theory that such translations can be completely automated, but Richard and I think, and argue in this paper, that this aspiration could well prove to be very problematic. In contrast, any transformation system – we think – must build within it a provision to apply Human Interpretative Intelligence (HII) to continuously reformulate the transformation rules and models used in such systems.
We also think, and argue, that it will be necessary to make a technical design choice about the means by which the translation process between different XML languages is achieved. By way of making our case, we examine three transformation systems that have three entirely different ontological approaches to address the same translation problem: the Common Ground Markup Language, (the CGML system), Contextual Ontology_X Architecture (the COAX system) and OntoMerge system.
Based on work by Kuhn, and Popper’s answer to his claims about incommensurability, we suggest that the primary criterion for comparing alternative translation/transformation systems for XML standards and schemas is the ease of formulating expanded frameworks/revised theories of translation that create commensurability between source and destination XML content and the merged ontologies. We break this criterion down into two sub-criteria: 1) the relative commensurability creation load in using each system, and (2) relative system provisions for and constraints on the use of human interpretative intelligence. On both of these sub-criteria, a detailed critical evaluation indicates that the CGML system is better than the other two, OntoMerge is next, and COAX ranks as the system with the highest commensurability creation load and the least relative provision for incorporating human interpretative intelligence.
The need for human intervention, and the need to make a technical design choice about the way in which content is translated – have important implications for the development of any global interoperability architecture, in the context of the current push for the semantic web initiative. We conclude by suggesting that premature standardization should be avoided, and that there is a need for further thought and careful additional comparison and evaluation before any system, including those reflecting the semantic web initiative, is adopted as the basis for any global XML interoperability architecture.
1 response so far ↓
1 frysystems // Dec 30, 2008 at 12:10 am
Excellent Paper – with a lot to think about.
At its heart, I doubt that the RDF format of a subject:predicate:object triple will be rich enough to fully support transformational ontologies, without taking up all the dependant attributes into the grammar of a solution.
And we are still left with the enormous task of checking and validating any modified transformational ontology. Any change should be treated, at best, as a “knowledge claim” and nothing more….