In my vocab list, I'm told that some are unavailable to study because "You have selected only kanji-related questions, and some terms a)do not have kanji or b)do not have kanji that you know." The second is fine and makes sense, I wouldn't want to learn them before the kanji. But the first makes me think, why not display those ones anyway?
Of course, for "Kanji → Kana" and "Kana → Kanji" questions, it doesn't make sense to test kanji-less vocab. But what if, for "Kanji → Meaning" and "Meaning → Kanji" questions, you allow these terms?
(I'm assuming I'm identifying the problem correctly, but I totally understand if I have the wrong idea here.)
Currently, the site does not presume that if a term does not have kanji in it and the kanji -> meaning is selected, that the user would be interested in kana -> meaning. (This may change in the future).
So, right now, the only way to make those available to study is to add a kana->meaning or meaning->kana vector.
That being said, this is the perfect solution for the everything index I'm building into the upcoming beta. You'll be able to easily rip these terms out, toss them into a katakana schedule, and add kana->meaning and meaning->kana vectors without overloading the terms with kanji in the original schedule.