As a non-OCT member I would like to strongly endorse this TIP.
Assuming it doesn't impinge on current development styles, making xotcl available can only help and attract developers (which everyone wants).
I personally would prefer to be able to use objects in OpenACS coding and would spend more time on OpenACS (as opposed to Perl and Ruby OO coding) if xotcl was part of the OpenACS core.
Another benefit of adopting xotcl is to pre-empt multiple (different) OO styles being used by OpenACS developers (like [incr TCL] for example).
I would be willing to invest time in any implementation required or regression testing.