A while ago, one of my friends (given how pissy and obnoxious I've been lately, I won't shame my friends by linking them here) linked to an online Johari Window app.
This meme's been making the rounds for a while, but most such apps have been limited by a lack of anonymity and excessive segregation of negative and positive traits. Recently, another acquaintance of mine whipped out his mad web hacking skills and threw together realpersonality.com, which seems clearly superior to the standard Johari apps if you're into that sort of thing.
Now, at this point I am put in a slightly awkward position, because truthfully I'm not all that curious what people think about me --- on the one hand, I think I have a pretty good handle on it, and on the other hand, I'm self-directed enough that I don't care that much. But, it's odd for me to recommend this to people if I don't use it myself. So, go to town if you like.
Incidentally, it's illuminating to consider the relationship between these apps and the likes of Hot or Not. The comparison suggests a couple of followup directions:
Add ratings of physical characteristics. (Duh.)
Bootstrap a dating service on top of the personality evaluator. Possibly mix in some kind of social networking/reputation system to reduce the risk of collusion attacks. This seems challenging, because there's an inherent tension between reputation systems and anonymity.
Conversely, add anonymous peer review to an existing matchmaking service. This would help solve an obvious problem with online socialization sites: you can put a profile on Friendster or Match.com or whatever, but the only feedback occurs when somebody's interested enough to send you a message, which is pretty coarse-grained and opaque. You have no idea what impression you're making on readers who don't contact you. You can ask friends to review your profile, but their reactions may differ radically from the reactions of people who don't already know you.
Imagine a site where users could anonymously mark a profile with key adjectives, without actually contacting the person. One suspects that people would learn what works and doesn't work much more rapidly.
It's interesting to speculate about the effects this would have. Would all profiles recede rapidly towards a mean of inoffensive blandness? Would people stay with the service longer, because they'd be getting better dates? Would they get alienated by the negative feedback and leave sooner? Or would they more rapidly meet the partner of their dreams, and leave sooner for that reason? In any case, while they're using the service, it seems likely that people would visit the site more frequently, because they'd be getting more frequent feedback --- which would be good for an ad-supported site, but bad for one supported only by user fees.