I'd also be interested in the longevity of Eneloops which don't get cycled very much. The background for this is the question if it pays off to use them in devices like wall clocks which typically run at least a year on a cell (or in this case, on a charge) or if it would be better to use Alkalines in this case.
The way my current calculation runs is that it's currently possible to get no-name Alkalines from about 20 cents per cell here in Austria. They may not have the highest possible capacity, but I'm confident that in such low-drain devices you would still get at least 2 Ah out of each of them. Whereas the cheapest standard Eneloops I've seen cost about 3 Euros each, so they would have to replace 15 Alkalines... which in a wall clock which runs for 2 years, for instance, would take 30 years. The question is if the Eneloops would really last that long or if they would get killed by decomposition of chemicals or such during that time.
This probably isn't easy to determine. One calculation of mine says that the ageing of cells shouldn't be quicker than their self-discharge. Now due to official specifications, the current gen Eneloops keep 75% of their initial charge after 5 years, and 70% after 10 years if I remember right... which means that from the 75% point on, they go down by 1% per year. If they continue to lose 1% per year after that, they should last for 70 more years for a total of 80 years.
However, another calculation of mine calculates the degradation differently... it assumes one year of degradation to be equal to some amount of cycles whose exact value is yet unknown, but I've read tests of older and newer Eneloop cells, and from them I'd estimate that value to be around 30-35 "typical consumer" cycles per year, out of which there are about 500, so that would give a shelf life of about 16 years. The reason why we don't see any degradation after 10 years then would be that Eneloops tend to keep their capacity pretty much constant through the first half of their life, only falling off later. The "typical consumer use" would be running them dead every time and then recharging them with a quick charger which fully charges them and doesn't let off until they are already pretty hot.
So the question is which of them could be true... is it 80 years or rather 16 years of shelf life?
Another question I'd have is about Eneloops going dormant when in low use... which means that the internal resistance rises more and more. This could maybe also apply to a wall clock if the batteries are being charged less than once a year because each charge lasts longer than that... do they get to a point where they need to be cycled pretty quickly multiple times in order to wake up again? This would also lessen their advantage over alkalines since then it's more of a hassle to use them in low-drain applications.