XOM API for Processing XML with Java

[Nils_Kilden-Pedersen AT Countrywide.Com]

In the "What's wrong" slides, page 68 on Encoding, it is mentioned that 
the Serializer "supports all encodings available in the VM". It also 
states that "Modular design makes it fairly easy to add more by 
contribution." 
Does that mean, that I "fairly easy" can add support for an encoding not 
currently supported natively by the VM? And if so, how is that done?
Specifically, I need support for "ebcdic-us" encoding.

Thanks,
Nils