Facebook, Pownce, MySpace, Twitter, Technorati, LastFM, WordPress…
two words: digital schizophrenia.
for god’s sake, people! can we please consolidate some of this crap?
there are some profile aggregators and reputation management things out there, but, really, they’re just solving part of the problem. i need something that follows along behind and cleans up after me.
it’s not like it’s THAT hard. it’s not like you have to be google and crawl 16 billion websites or something. all you need are the top few thousand. or even just the top few hundred social sites.
c’mon, mark, how hard is that?
i’ll even make is easy. here’s the simple, 3-step pathway to world domination (i’d do it myself, but, dammit, i’m already taking over the world with other projects):
STEP 1: Data Collection
like i said, all you have to do is crawl the top 100 social/web 2.0 sites — to get the bulk of your information. when people sign up, just ask them for their “handle,” name and email. then crawl away. (make sure you go grab all of their contacts too — very important in the next step.) between blog search, facebook, orkut, friendster, bebo, linkedin and myspace, you should be able to build a pretty complete profile.
besides, you can always let them go back in and add their own profile links.
STEP 2: Relationships
now that you have their data, wrap this around their identity with openid. now, you’ve got all of their links and contacts, you can start drawing the core relationship lines of a social network. both bob and joe link to pages about sarah therefore, she’s a shared contact.
now, you have bob’s openid and sarah’s openid related to each other. the power!
STEP 3: Reinforcement
there’s this cool philosophy/methodology/standard called microformats. publish your data in every conceivable format (html, web services, gadgets and widgets, rss feeds, everything) and build it with microformats in mind.
become the information standard. reward sites that use microformats with better search rankings. build the larger network through cross-linked quantity (sheer bulk of profile data a site has) as well as quality (how detailed their microformat information is).
sell it all to yahoo or google or microsoft. sit by the beach. have a cocktail. maybe even buy me one.
the key takeaway here is to crawl the web (and some highly targeted sites) with people in mind rather than content.
that’s really the difference in a content search engine and a people search engine. you probably don’t care about edge cases or brochureware or anonymity. you go to where the people are. you care more about large forum sites than something content-rich like wikipedia.
to be honest, there are some sites on the cusp of figuring this all out. technorati is really my personal favorite. it just has two downfalls. it’s terribly slow and all they care about is blogs and not other web content.
sigh… there’s a lot more data out there other than some blog trackback, guys.