Well, the plan certainly exists, and I have spent time looking into it. Its not a simple job though - its basically a complete rewrite, with extensive reference to the original. Its possibly a bit early to do this though - the Blender 2.5 api is not 100% locked down, though its close enough that any changes shouldn’t be too bad.
However, I haven’t committed fully to doing this - the reason is that the plan for Panda is to depreciate egg files for Collada files, and I am not sure I want to put in the effort of the port for a file format that will be going dodo at some point in the future.
I have no knowledge of the time scale of this plan, so don’t know what kind of ‘gap’ in service this would cause, but there is already dae2egg, so the collada pipeline already works, albeit with a conversion step. (Blender 2.5 already comes with a Collada exporter.) It might make more sense to work on that pipeline than create a new temporary one - my list of things to do includes sitting down and having a play, to see what the limits of the collada route are, as I don’t know how well it works right now. There is a strong likelihood that certain Chicken features will not be supported via this route - would need to see what they are and if they can be fixed.