social networks

A bunch of folks have created technology to let me network with my friends and their friends.

I have tried solutions from LinkedIn * (generic), Plaxo (update), orkut (friends), flickr.com (photos), GoodContacts (corporate), Friendster (dating), Six Degrees (dating), Classmates (social), Peoplestream (recruiting), meetup (meetings, not sausages), eliyon (sales), tribe.net, Ryze and others. These all have very different goals, but share the concept of being able to contact friends of your friends etc.

Here is a longer list from Cynthia Typaldos.

Most social network companies have and will continue to fail for these reasons:

  1. They are too easy to use to send spam. For example, I recently got an email from the secretary of a top venture capital guy I know, asking to update my info in her address book. The template email which was generated made it sound like she was a friend of mine. But I found her email offensive, as I actually had no relationship with her, and she did not take the time to state what her claimed relationship to me actually was.
  2. There is too much hype about six degrees of separation. (And btw the popular understanding of this phenomenom is actually incorrect.) Who cares about the friends of the friends of my friends? I sure don't. I have a tough enough time keeping track of my friends, let alone the friends of their friends.
  3. Most of these solutions keep a copy of my personal data on their central servers. Sure, they make all kinds of claims about security and policies, but I chose not to trust startups with my personal data. (Even "well established" companies such as Palm can end up turning into desparate scumbags who change their privacy policies and start selling their lists.)
  4. There is not enough value-add for any of these solutions to justify their existance.

The ultimate people networking solution should have the following features:

  1. True peer to peer. I don't want my contact book to ever live on a shared server. Period.
  2. Use progressive intercepted IM instead of email for updates or search requests from people already on my contact list, as email is too UI heavy and not real-time. (See Knock-knock below.)
  3. Throttle controls to make bulk email difficult. E.g., put up a stickup "Do you really want to send an update to all 428 people in your contact book?" E.g., only let them contact a maximum of 10 people at first, and require a certain response rate (e.g., at least 5 of the 10) before letting them contact more than those 10.
  4. Impossible to automatically contact friends of friends.
  5. Impossible to import friends of friends, thus converting them into your "friends".
  6. Blog RSS integration, so everytime someone emails me looking for a job, or looking to hire, I can one-click add a pointer to their request to my own website. Then having a distributed one-level (not two-levels) XML search of possible matches across my friends' websites / contact books would be much more efficient than going to some anonymous glob such as monster.com.
  7. Free and open-source of course. :)

See also my page on contact books.

paulenglish.com - articles - startups - nonprofits - press 31-Dec-2005