At first sight, dye release seems a good idea, but would the dye stream from great depth rach the surface. Would the dye diffuse far and fast enough to be of any use? If it came from a part, such as a composite fin, that was likely to separate in any breakup and reach the surface, then it could make location of the debris field a bit easier.
I have no knowledge of flight data recorder technology, but I would be surprised if every parameter going into the FDR had its own wire and connection. So providing the same data to a slave recorder would not require a heavy wiring loom, probably no more than a single wire pair, or even a single optical fibre would be needed. A slave CVR, ideally with capacity to record the whole of a flight, could be implemented the same way.
A recent post suggested buoyant FDR that would use GPS to establish where it reached the surface and would broadcast a signal to assist in its being located. The original idea was that triangulation from a number of receivers could give a rough idea of its location. But if it has a GPS receiver, why not add a little more electronics and broadcast its position in the location signal. I believe such a system has been used in recent years for tracking animals in the wild.
The problem of getting a locator to the surface from great depths has also been mentioned, along with the problem of finding materials that would have the necessary strength and buoyancy. It is probably not ncessary to look for something exotic. How about a plastic bag with, at the open end, a block of something that would react with water and produce a large volume of gas. The most difficult thing would probably be ensuring that the gas went into the balloon.
And finally, although AF447 has the highest recent profile, there have been other accidents in recent years where the location, on land or under water, has been difficult or impossible to find. Some of the ideas promoted on this thread might have been useful in those cases.