I'll cast another vote for 3 gallons as a reasonable minimum batch size. The right answer is to some extent going to be an individual one, as it involves trading off time, money and effort. I suppose it also depends a bit on just how experimental you want to be. Personally, I find that much of the effort of winemaking relates to the batch, not the number of gallons, so three 1-gallon batches would be nearly three times the work of one 3-gallon batch. Also, since it can easily take 6 to 12 months or more before you can really tell if a given batch is successful, it makes more sense to me to make a bigger batch. If it is successful, you have at least a case or so to enjoy; with a smaller batch size, you need to start the production process over again and wait another 6 to 12 months or more. Yes, there is the risk of "waste" if/when you decide that a particular batch just didn't work out. But (at least for me) that is a pretty rare event, and the cost of the ingredients is a minor one compared to the investment of time and effort.
As for life expectancy, I'm not sure there is any real rule of thumb. I've been told that three things contribute to longevity in the bottle: alcohol, acidity and tannins. To the extent you have medium to high levels of all three, the wine (under good storage conditions) should last longer. In the case of fruit wines, they generally don't have tannins; the alcohol is likely to be normal or a bit on the low side, and the acidity normal to high. In most cases, I would expect fruit wines to be at their best within a year or two, and to start to decline somewhere between 3 and 5 years. Individual wines are bound to vary, however.
Doug