The xenon isotopes that result from fission have short half-lives: 9.2 hours for Xe-135 and 5.2 days for Xe-133. So if they show up, as they have at Fukushima reactor 2, that means that fission has been taking place.
The Japan Times has a report that includes all the important information and puts it into perspective. The amount of both isotopes of xenon detected is one hundred thousandth of a becquerel per cubic centimeter in gas samples. That means that there is one becquerel - one count per second - in a hundred thousand cubic centimeters of gas, or in a hundred liters of gas, a small closetful.
When you add in that "there have been no drastic changes in the reactor's temperature and pressure" reported by Japan's nuclear safety agency, what this adds up to is that there probably is a pile of broken and melted fuel pieces at the bottom of the reactor vessel. Somehow, a very limited part of that pile got to criticality. Tepco has added boric acid to the water to absorb neutrons and prevent further chain reactions. But the criticality was very limited if it didn't make a difference in temperature or pressure.
I've seen a few fevered tweets and Facebook posts on this. There's an element of "I told you so" in some of them. I argued against the earlier reports of "re-criticality" and see no reason to change those opinions. Obviously there has been a small criticality at Reactor #2. But I thought that the reason people were worried about "re-criticality" had to do with the "China Syndrome" idea that the core would sustain its criticality to melt through the bottom of the reactor vessel and on to, um, Nebraska. That's not what's happening.
Update (November 3, 2011): More information has become available, and it appears that the xenon is coming from spontaneous fission of curium isotopes produced by normal reactor processes.
Cross-posted from Nuclear Diner.