Many thanks for your answers.
Erik, David
Yes, I do have a two way translator for the languages in question. In fact, the script based version has to be translated to its xml form for execution as the engine understands xml only. However, I am not sure how do I exhaustively check all possible combinations of features and translate them two-ways, that would be too much. About induction, I am not sure that paradigm of proof we can use for such a scenario. Are there any existing such proofs that you know of?
Dank
Actually I need to show both and funny thing is that we have a proof that the xml based version is Turing complete and we have a complete EBNF grammar for the script-based version of the same language. I am thinking, it would be nice to do both ways and see where it goes.
By proof, indeed I mean a mathematical proof rigorous enough to be published in a scientific journal. :)
Thanks again for the replies and please post if you have any more insights. These answers have definitely got me started.
cheers
Ketan