I'm sorry if I was unclear. I am not doubting the existence of an F cup. (I mean, really, that's kind of dumb. Do I live on the planet of flat-chested women or something? :-) What I am saying is that based on the old school bra sizing method I learned way back when, it would be very hard to get a size like 32F, because the number on the band was supposed to be your actual bust size at the widest part. 32 inches is a very small bust measurement for any adult woman, so that woman would have to be very tiny to have an F cup and still measure 32 inches around the bust. That would be about a 26 inch rib cage--not something you see on an adult woman. I grew up sewing a lot--sewing with my grandmother, sewing my own clothes, sewing barbie clothes, etc., and any time a size is given for anything on "top," it's the bust measurement at the widest part. Always the bust. Nothing is sized based on the measurement of your ribs.
Now what people are talking about here is a different way of calculating bra size, using the number as the measurement around the rib cage. Yet another method, according to the internet, is taking the rib cage measurement and arbitrarily adding 3 or 4 inches. And, finally, there's the Victoria's Secret method, where you use a chest measurement above the bust, under the armpit, as your band size. I went around the internet plugging in my actual measurements to see what my "bra size" would be, and I got sizes as disparate as 34C, 38B, and 40C. My conclusion is that bra size actually doesn't translate to a measurement you can get with a tape at all.
I do think vanity sizing comes into play here because even though dress sizes have changed, and today's size 8 or 10 is significantly bigger than it was twenty years ago, the average bra size is still supposedly an unchanging 36C. Using one's own breasts to track the drift of bra sizes, though, has obvious drawbacks. I myself have had a baby in the interim, so nothing is the same.