Don't Back Down

I read a post from Scoble about tenbyten. It's an entertaining concept, but in practice I don't quite find it useful. The idea is really sweet, and the interface is nice. But the problem I have is that, out of 100 pictures, I saw 7 of one, 15 of another, 6 of a third and... well, you get the idea. The words are different, but they are finding the same images. Neat, but needs work. I hope it gets better.

Somehow I made it from tenbyten to WORDCOUNT, which tops the uselessness of tenbyten even further by only providing words (no links from them or supporting info or anything). I mean it's marginally interesting to know that the is the most common word, but is it really surprising? Knowing that conquistador currently ranks last (at 86,800) is marginally more interesting, but is it really one of those things you'd like to see on a regular basis?

I then took the jump from WORDCOUNT to QUERYCOUNT, which takes the concept one step further. QUERYCOUNT shows simply how many times people have queried the WORDCOUNT interface. The winner is really not surprising, and I won't repeat it here.

What is a bit interesting is that the "by rank" option doesn't seem to work well at the end of the list. There are apparently 16,205 words in the list, and when you key in a number in that area (I tried 16,000 and 16,204 as well as 16,205), the flash gets all jumpy, and more importantly it shows words that aren't likely to be at the end of the list since they are at the beginning. A number in the middle, such as 1003, works without any apparent problem.