In the few short years since I originally floated this proposal, there has not been a groundswell of enthusiasm for the notion of batterychips. But sometimes these things take time to catch on. The biggest objection seems to come from people who use only rechargeable batteries; their claim is that because they never swap out their batteries, they would never get new software, and that this raises the per-copy price on the software included with the battery.
I'm not sure that this is a valid issue. First of all, today we see a broad mixture of people who use replaceable batteries and rechargeable ones. It is a price/convenience trade-off for most of them, and the market adjusts the prices so that neither side goes out of business. Similarly, I think the price of the software piggybacked onto each kind of battery will be priced according to the expected life of the battery. So the same software piggybacked onto a rechargeable battery will be more expensive on a per-unit basis than the same software piggybacked onto a non-rechargeable battery, because it has a longer life. It's just a pricing issue, and the market will decide.
The key notion, from my point of view, is getting software into the consumable category and out of the capital expense category. Once you do that, usage-based pricing becomes logical, and in the end I believe that is what serves the market best.
See you at the batterychip store!