This site isn't very useful unless a lot of people are using it. [1] Furthermore, the only people with any real motivation to input data are those that want to game the system. How about this as an alternative version 1: a facebook or linkedin app that lets you request a recommendation for a caterer, mechanic, whatever, from your friends. That's useful immediately, and you can aggregate responses.

Also, the people giving me the thumbs-up were beggin for a hurtin.

[1] The current version of your site doesn't care about location, which makes it look less empty but be less useful.

The first question that comes to mind when I see the front page is, "Why would I use this instead of LinkedIn recommendations? Especially since I don't want to convince everyone I know to switch from LinkedIn, etc.".

It seems like your answer is that trust is weighted by how much you trust someone.

However, that isn't clear without some digging. Certainly, that point is more important than the "thumbs-up" people, yet the "thumbs-up" people are more visible (and annoying).

A couple other questions -

- I'd still need to be sold on getting people I know to sign up - maybe there's some way to piggyback off of existing networks?

- Just because I trust someone in one area doesn't mean I trust them in others. For instance, someone may be a fantastic lawyer, but just because they recommend a developer does not mean I think they're competent to rate developers.

- The tagline, "Your credibility, online" strikes me as a little much. For one thing, the idea of trusting my credibility to a single site gives me the heebie-jeebies. Fortunately, that's not the case, credibility cannot exist in a single database, and thinking about that drives home that the tagline may be overstating it a bit.

- I really dig that the registration form is minimal, but it makes me wonder how to distinguish multiple people with the same name.

Great suggestions. Thanks.

Are trust ratings just raw sums, or can they be based on how much people you trust have rated someone?

For instance, I don't care about how many sockpuppet accounts have endorsed someone, I only care about how they're rated by people I know or trust (or, to a lesser extent, how they're rated by "friends-of-a-friend", though there should be a more than linear decay between each degree of separation.

Also, having publicly (or even privately!) viewable trust ratings can skew the ratings. I might have a true opinion of someone that's better or worse than what I'm willing to publicly announce (we can't all be Zed Shaw ;-). Viewable ratings also create the problem that I might feel pressured to rate someone well so that they won't retaliate against me: let's call this the ebay effect ;-).

There also remains the problem of proving that a given account actually belongs to the person named.

You need to have a posted, credible policy that you won't tamper with or sell ratings, or reveal private information (such as how I really think of so-and-so).

A system like this could be great, but you have to take into account every possible way it could be gamed or cracked, and build in solid defenses against all of them from the very start.

You got it right, the idea is to make these trust ratings based on how much people you trust have rated someone. In the alpha version it does not go that far yet, that's the idea though.

I agree that there has to be a good way to prevent how the system can be gamed. We are building those into our design, just haven't had the time to totally add it into the front yet.

Good suggestions. Thanks.

I would suggest adding another radio button (e.g. "Company"), inactive if necessary, so there isn't just one People button. Looks a little weird with just one button.

The system looks useful. Not sure if I'd want people to contact me after searching for programmers. You could make sure that the example tag search ("E.g.: Chinese food caterer") returns an entry or that it fits with your targeted audience ("E.g.: PHP programmer").

Seems similar to the original intent of rapleaf.com, which is a platform for building reputation. They are actually changing their business model because their product didn't gain any traction. You might want to check that site out, and see if you can differentiate.

Thanks. Rapleaf was based on rating buyers and sellers and based on email reputation. It did not provide much insight into the rating.

EndorseYou is more like Yelp for people in our mind if we were to come up with an analogy.