I have been trying to figure out Marc Canter's People Aggregator for some time. Why? Marc fascinates me. Loud. Obnoxious. Arrogant. Passionate. You can't keep your eyes off the guy - even if he pisses you off.

Anyway the new redesign helped and I think I am starting to get where he's going, but I think he is a bit off the mark with the present implementation, and I am not talking user interface.

So, since I am a truly fascinated by Marc, I thought I would pass on my thoughts on how he can improve his People Aggregator.

1. Your new home page while better message wise is a barrier for guests and people like me checking out your site. I want to know what's going on without logging on first. Permit full guest access entry from the home page.

The rest are big structural changes, but I think you could do them with medium effort considering you already have a core framework, from what I see, to support the architecture I am suggesting.

2. Forget the idea of "Community in a Box". There are a ton of them out there and companies' already offering corporate versions for Intranets. Also, your present generalized approach offers no compelling reason to adopt the People Aggregator. Each sale is going to be a hard sale requiring lots of your time, and your passion. Both are not scaleable.

3. Forget "Build your own Network" or "be your own oligarch". This is related to the above. You cannot generalize a community. Features are specific to a community of interest. Culture, and the relationships between users/content and users/users vary widely depending on the community of interest. You are generalizing, which has very limited value since generalization by definition is the antithesis of a community. Broadband Mechanics or Donat Group can build requested communities reusing existing code as a pay per service businesses. Donat Group has built 5 community sites in the last year. Each is highly specialized and differ greatly from one another even though they were all social networking and community sites. We obviously could reuse a significant amount of code in each site.

4. Stick with what you are passionate about: Open standards, open ID, and open data formats. You should be servicing other existing communities. Not only is there a business there, your product name, People Aggregator, would actually begin to make sense.

5. Be the conduit to other communities using a simple API. If you register with People Aggregator you can sign up to "All these other existing communities of interest" with a single click. That's important. Imagine, signing up to Bebo, Project Opus, Flickr, and Blogger with a single click. You can do this! I (projectopus.com) don't want to worry about SXIP, OPEN ID, etc... I'm happy for you to worry about it. Become a trusted identity HOME site. I won't balk at this; it doesn't stop people from signing up directly at projectopus.com. I'm not tying my ship to yours. You are offering me a set of members I potentially wouldn't have had.

6. Become the goto point for "what's happening" on the other communities (or at least those who have signed up to the aggregator). Using your API, aggregate and filter content from the other communities. This will facilitate cross pollination and connection among communities creating more value for your service, and enhancing the value of our communities signing up with you. Technorati and Bloglines do a poor job at this, because RSS alone is not sufficient.

7. Stay open: you yourself know we are not going to sign up or stay with your service if we sniff a power play. Dare the competition to compete.

8. Focus on aggregating communities surrounding your other passion: Digital Lifestyle. Pictures, Video, Audio. Let others using your open APIs develop "peep aggs" for for Alt. Lifestyle, or IT, or Investing, or.. or ... or...

9. Charge other communities a onetime fee for signing up to your service. Do not require them to be reciprocal with members. They will eventually, once their users start bitching.

10. Offer communities signed up to People Aggregator a range of fee-based services. I can think of a dozen off the top of my head.

11. Offer the actual members of People Aggregator a number of fee based services. I'm sure you have more ideas here than I do, but there are tons!

Anyway. The advice is free. Probably all it's worth.

PS
You should still be eating your own dog food. If you won't no one will, and you will never know why.

A slick pop song by Christina from the City of Champions