Thanks once again.
I'll try a roll-back, and to find all of the 0.20.2 packages. It seems to me that the segfault is on the SME box, as it appears inside of the ssh terminal session (which was indeed launched from the working ubuntu environment). I don't really know how to tell for sure, but that's what it seems to me.
Also, the libmyth*, well all the lib* sections of the howto didn't work for me -- I tracked them down manually in a webbrowser in atrpms and then pasted the link to each in a wget in ssh, and finally did a yum installlocal for those. It seems that the versions of all of these files on atrpms might be either (a.) incompatible with sme, or (b.) incompatible with each other. However, this is just a guess.
Lastly, I definitely ran into trouble with ivtv fighting with ivtv-firmware over something with init.mpg or init...mpg in it's name. I had to do a "force" install via rpm on those, but I'm not even far enough along to need ivtv yet, and so that shouldn't enter into yet.
I could certainly try from a centos live cd (which i have used elsewhere in virtualbox). I think I should definitely uninstall all, drop the db, accumulate all the right rpms and try a yumlocalinstall, then load the mc.sql anew. Could you or anyone else comment on matching versions of the various rpms needed, specifically the myth* and the lib* rpms?
Many thanks again,
--Matt