In the past months, the JLU entry at Area 51 says that the number of questions per day "needs work", which during the last few months has dropped from 3.5 to 2.6 questions per day. There is probably little that can be done to increase the number of good questions, but I was just wondering if there is a limit on the beta period, and if it is really true that "If the site does not get used, it will be deleted" (Area 51 FAQ), in which case I would probably like to take some notes.
If you look at the stats on most of those sites, specifically the user stats, there wasn't much of a core community. Even after almost 300 days, there were barely any high-rep users with lots of answers (or questions), which is bad because it would suggest people aren't sticking around and it means the community doesn't have the reputation to help with moderating the site.
The criteria for graduation are subjective and not based entirely on the numbers alone. Some sites have really good stats on the surface, but have languished in beta for years because the quality of the content doesn't match the best content available elsewhere on the web.
How is Japanese doing specifically? It's a little hard to tell, but our traffic is growing (slow, but steady). The last public remarks the executive team made about the foreign language sites was they weren't doing bad enough to be shut down, but they also weren't doing great. Granted, that was over a year ago when all of the language sites had under 400 views a day.
I would suspect that our site is still kind of in the middle. We're not doing bad enough to be shut down. Some of our questions and answers are pretty good compared to the rest of the internet (though others are not). And I've seen a lot of new users posting lately - both questions and answers, so we are getting fresh blood. But, the growth certainly could be doing a lot better.