Facebook team knew that the fundamental difference between their and other social networks is the “profile hopping” behavior as primary means of noticing new memes. MySpace widgets grew because people were seeking to copy what they observed on each others’ profiles. Facebook’s Feed makes profile hopping a lot less of a primary function. Meanwhile, success or failure of the platform meant (at least PR) success or failure of the company.

Front-loading various kinds of invitations was key to guaranteeing that the platform would turn developers’ and press’ heads.

The model you describe makes a lot of sense, but it is inherently riskier: what if nobody built great, compelling apps at the outset?

As the platform matures, creating a more balanced “game” (and better economic opportunities) is key to retaining developer attention.

better watch out — if you keep this up, people are going to expect you to write these kind of masterpieces all the time.

—

re: viral channels, i think there’s also a significant discussion around whether those channels are designed to amplify a) initial invitation requests, or b) demonstrated usage by other users.

while i know the discussion above is theoretical, i can’t help but jump into the lab to assess the Facebook platform. currently, my observation is that Facebook has designed its initial viral channels primarily around the invite process, and has in fact actively *prevented* amplification around the news feed / usage model of others, unless they already have the same app installed.

this strikes me as a bit odd.

one would think other social network users’ vetting of an app thru continued usage would be the BEST possible signal of “quality”, and should therefore be rewarded with signal amplification thru the social graph & news feed to other users… regardless of whether or not they have already installed the app.

in fact, an arguably better platform design choice might even be to PREVENT invites altogether, and to only allow viral channels by optimizing the news feed stories coming from apps being used by others, and amplifying those stories.

i can only surmise that Facebook may have had a concern that not enough viral activity would occur unless some amount of “spammy” invitation were allowed by developers. not sure i’m correct on this point (and if so whether i agree with the decision), but it does seem to indicate that as you suggest, Facebook is trying to work out sustainable game mechanics for its platform & developers.

in summary: God may not play Dice with the Universe, but Zuck may indeed play Craps with the Platform

- dave mcclure