BU's acceptance depth approach, I consider it to be a terrible approach to bigger blocks.
Usually 6 confirms is considered as statistically set in stone. Not with BU, all of a sudden a bigger block might fall through the gate causing a chain reorganisation and potentially the transaction might not have any confirmations at all due to the mining policy on the new longer chain.
It's overcomplicated approach introduces new economic risks, even if its bugs are fixed.
I am all for bigger blocks. However you can call BU Bitcoin Unavailable, Bitcoin Uncredible, Bitcoin Unneeded or the Bitcoin Undead for all I care.
imagine this though..
ALL nodes had a 1mb block consensus rule
but below that they had lets say a preference of .. well lets say the majority was viewed as preferring 0.25 in 2011 and where they would accept 0.25+ only after 6 confirms
it does not break the consensus rules. but does publicly show pools that the majority of node prefer only 0.25mb blocks.
pools wont want to go passed 0.25 because the majority of nodes wont like it.
(the nodes would accept the blocks, but wont treat them as set in stone unless 5 other blocks are ontop)
the pools would weigh upt the pro's and cons of going passed 0.25 much more wisely, rather than just doing it
EG pool A may want to make a block 0.26mb but pool B,C,D,E,F,G are building blocks that are only 0.249
so theres a chance pool A's block may get orphaned.
but all nodes would still be all on one chain due to all blocks being under 1mb.. the dynamics is just about orphan risk not chain splitting(altcoin creating)
now say its 2012 and nodes moved their preference of to a majority of 0.5mb..
EG pool B,C,D,E,F,G that were listening to community demand will happily make blocks over 0.25mb meaning so can pool A without any orphan risk
as you can see this is not about causing chain splits because all blocks under 1mb are acceptable. all its doing is keeping pools under control