(I'm intrigued by your mention of BTC or similar as a possible native asset. I presume you don't mean an IOU that happens to be spelt with the letters B,T,C? 
Do you mean some sort of crossover protocol, where burning of blockchain BTC causes something called "[non-IOU] BTC" to appear in Ripple shortly afterwards? [There'd be no crossing back of course. Not unless the Bitcoin developers, ah, got the Ripple bug. ] Or do you mean something else entirely? Apologies for sometimes being slow to understand things. - Again, whatever your exact meaning, this would be for someone else to try: I'm going to try a clean slate.)
No I indeed meant using native BTC as native currency on Ripple - XRP have the same features as BTC, they are just a bit faster to irreversably confirm, that's all. It might be a bit more complicated, as validators then would need an SPV or maybe even a full BTC node to verify transactions, but fundamentally I don't see many issues with that. Transactions of course will take longer to clear (different validators might implement different rules e.g. one just waits a bit for double spends and then confirms, while others are waiting for X amount of blocks deep), but there would be no need to destroy BTC, as the anti-spam mechanism is anyways enforced by the BTC network. Do you mean some sort of crossover protocol, where burning of blockchain BTC causes something called "[non-IOU] BTC" to appear in Ripple shortly afterwards? [There'd be no crossing back of course. Not unless the Bitcoin developers, ah, got the Ripple bug. ] Or do you mean something else entirely? Apologies for sometimes being slow to understand things. - Again, whatever your exact meaning, this would be for someone else to try: I'm going to try a clean slate.)Compiling Ripple is relatively easy, the only issue is currently with using CentOS and other RHEL clones since they use relatively old compilers an versions of boost.