>It was my understanding that since the DOM interfaces are specified >as IDL, it was intended that implementations use the appropriate >CORBA language binding for implementing the DOM API.
The DOM Working Group used IDL informally, as an API abstraction language. There was no intent that the IDL published in the DOM spec be taken literally as a binding in its own right. While there's nothing wrong with the idea of an official IDL/CORBA binding for the DOM (and in fact, I believe the OMG was working on one), there's also absolutely nothing wrong with language-specific bindings... and in fact there isn't _necessarily_ anything wrong with having multiple bindings for a single language, if the users of that langauge agree that there are conflicting requirements which a single binding can't meet. At this time there is no officially blessed C++ binding for the DOM; the DOM WG doesn't have the cycles available to develop one, and the C++ community has not invested the energy to reach consensus on one and submit it as a proposal/Note. There are some major issues to be settled regarding memory management and such. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
