jtivat:
Well if this is the reason you gave this is more of a convenience to still make a second dive with a half full pony no real safety advantage just don't do the second dive.
I mean this with utter sincerity, and I mean no disrespect, but I can't quite figure out exactly what you were trying to say here. As close as I can tell, it sounds like you were saying something like:
"Making a second dive is simply a convenience. If a half-full pony is no real safety advantage, just don't do the second dive."
If that is indeed what you meant (if I am mistaken, feel free to rephrase), it would actually validate my position. I consider having a full pony a safety advantage. Wanting to preserve that safety advantage for a second dive in the event of what I perceive to be a certain common, non-compounding failure modes is a valid reason for carrying two back-gas-connected second stages.
"You can just skip the second dive" is hardly an effective counterpoint. It is as absurd a position as if someone claimed you never need to clean your regs, as if they fail, you can just skip the second dive. You can always choose to skip a dive, but having procedures to lessen the chance that a gear failure will scrub a dive is a good thing, is it not?
jtivat:
Maybe I missed it but how is that second octo going to help besides cluttering up your rig and adding a spot for loss of gas or getting hung up on something.
I have two second stages on my back gas and one on my pony. If I or my buddy suffer a failure of our back gas, we share the other's back gas using the alternate second stage. We would not use the pony unless there was an additional unforeseen problem precluding a standard back-gas-sharing ascent.
By executing a standard air-sharing ascent and not using the pony, if the problem precipitating the air share can be solved on the surface interval (such as would certainly be the case if it were a yoke face O-ring -- you wouldn't use the same tank the second dive), the second dive can be made with the pony in pristine condition, completely filled and ready. It would then be available in case some unforeseen problem precludes a standard back-gas-sharing ascent on the following dive(s).
Basically, when diving with a known dependable buddy, I consider the pony as something only to be used in the event of a cascading failure. Any single failure could be handled by standard practices, and the pony would remain as "backup to the backup". In the event that the pony is called upon due to a cascading failure, diving is over until the multiple failures can be addressed and the pony can be brought back to operational capability. It is a more conservative way to dive, but I am likely somewhat more risk-averse than some (or perhaps, I simply think too far).