"Kana" is the collective name for two the Japanese character sets Hiragana and Katakana. These two character sets are like the alphabet of the Japanese language (the linguistically correct term is "syllabary" rather than "alphabet"), and usually when you learn to read Japanese, you'll start with one of these two sets before moving onto Kanji.

When I'm presented with Hiragana, I can recognize the characters with near 100% accuracy. Katakana, I'm a bit weaker, maybe 80-90%. But when the two are mixed together, my recognition skills seems to drop significantly. I'll often confuse せ and サ, for example.

There's a bunch of Kana tests online, but all of them seem to focus either on Hiragana or Katakana one at a time, instead of mixing them together, which I figure must be why my skills were so weak in these mixed-situations. I never got any practice! So I wrote my own Kana test as a Java program. You can try it out at http://nebupookins.net/programs/KanaTest.php; you can use it either online just by going to that page, or you can install it on your machine by going to the page and clicking on the JNLP WebStart link.

I'm still developing it, so there'll be new features as time progresses. The program isn't going to teach you the Kanas. It assumes you already have some familiarity with them, and is there to test and reinforce that familiarity.

If you're just starting out, want to learn the kanas, and are sexually attracted to girls, you might want to try this page. The more Kanas you recognize, the less clothes the girls wear.