I haven't noticed any differences with our table breaks, we've just started our 6.0 regression testing. We are using FOSI so I'll let you know if I see anything like that.
I may have some insight to your not-last-page-bottom-rule table issue (I think I used worse english than you to describe it). We use the following definitions in a custom.tmx file in the custom\inputs folder:
\TableSplitPreBrk=0 % -1=none, 0=solid, 1=dotted, 2=dashed
\TableSplitPostBrk=0 % -1=none, 0=solid, 1=dotted, 2=dashed
\TableRulePreBrk=1 % 0=none, 1=previous row rule, 2=first table rule
\TableRulePostBrk=1 % 0=none, 1=previous row rule, 2=first table rule
Everything after the % in each line is a comment (so I can remember which value means what). Those definitions preventour tables from having that "empty bottom" when they break across a page. This is workingin our 6.0 environment (worked in 5.3 and 5.4 for us as well).
Hope that helps.
In Reply to Paul Nagai:
I'm also losing the bottom rule on the not-last page of a table that
breaks. (I'm sure there is better English to describe that.)
I know some folks think that's proper (and I like the cue it provides) but
I'm pretty sure my authors, at least some of them, aren't going to like it.