memory effect in lithium batteries (LiFeP04 tested)
 

Not a deal breaker, nor do I see a large loss in capacity, yet, but there are going to be some apparent loss as state of charge gets a lot harder to compute. So next time someone says in absolute phrasing "lithium has no memory" you can say "You don't really know what you are talking about" :)

Memory effect now also found in lithium-ion batteries

Maybe I'm not totally understanding the article, but it doesn't seem like that big of a deal... I mean lithium batteries have been around 20 years now and we're just finding this out? It can't be a major problem.

We have seen them in full-recharge applications, but not so much in hybrids.

Not a big deal. "a few parts per thousand". This paper only states that the memory effect is there in a tiny amount if you look hard enough for it but in practical terms, it is still a non issue.
.
Partial charge/ discharge use is proven to prolong useful life of these cells. You have to stay away from the last 7% on either end.
.
NanoTech LiFePo4 batteries such as the A123 AMP20 pouch cells are great batteries if you don't need huge discharge rates with an easy reach to 3,000 80% cycles. It will be a long time before anything significantly better comes along.

Sounds good to me.
My evil plan for using LiFePO4 batteries as starting batteries and to run equipment with the engine off will work perfectly.