Paul Madsen puts it in a nutshell, far more neatly than I could. There's a difference between interoperability - though that is a laudable goal in itself and to be applauded - and a metasystem.
A meta-language is a language for describing the characteristics of languages in general. It doesn't translate from one language to another.
A meta-system provides a means of describing the characteristics of other systems and how they relate to one another. If the systems are 'players in the game', a meta-system 'stands aside from the game'.
The importance of being able to make this distinction is made very clear by Mark Wahl's commentary here. For instance, he notes that issues in the Open Source Identity Selector interoperability exercise arose out of "the lack of schema management for claims, and a lack of semantic definition for claims handling".
Arguably, a meta-system would proceed from definitions of elements such as that to the implementation of specific technology instances, and not the other way round.
As I say, I think interoperability (particularly if based on open standards) is a laudable aim, but I also think it needs to be clearly distinguished from a metasystem.
PS - Just for the heck of it... I have a similar aversion to the word "methodology", which ought to mean "the study of method/s" but is used as if it means simply "method".
![[RSS Newsfeed]](/themes/sotto/images/valid-rss.png)
![[This is a Roller site]](/themes/sotto/images/roller-logo.gif){kind=logo}