They don't even need to upgrade. Remember that mIRC supports Unicode from 6.17 onwards, so it's likely the client they are using already CAN use Unicode to communicate. You simply need to convince them to enable it. That should be easier than telling them to upgrade.
Backwards compatibility would be nice, but unfortunately it would defeat the purpose of moving to Unicode. As I mentioned, 6.x "supported" Unicode, but it only did so good of a job. Many of the issues related to Unicode breaking down had to do with the exact legacy code you're talking about. This is why Khaled chose to streamline mIRC and remove a large amount of complexity in the data handling and display routines. This will allow mIRC to work better with Unicode, which is where we're all headed anyway. Basically, all that legacy code was making it impossible for mIRC to move forward and properly support Unicode. An mIRC 7.x with Unicode and the legacy encoding support would just be an mIRC 6.x, because that's exactly where we were before we started this process.
That said, it should be possible in the near future (upcoming betas or next release at worst) to script this kind of backwards compatibility encoding support with the new /raw -n functionality that was added in 7.03. It might be a little more complex because it has to be scripted, but eventually I'm sure someone will write such a script, once all the loose ends are tied up.
