I’ve been working on serendipity for some time now. In a series of essays, I’ll share some perspectives, patterns, and practices. I hope this work will lead to the fourth “p”: protocol. My quest is driven by the belief that the digital, workspace and urban environments can be designed to increase the likelihood of unexpected, beneficial discoveries and encounters. These affordances and rules are not going to only stimulate unusual combinations to happen but can influence the behaviors of agents to provoke serendipitous events and their attunement to notice them.
Personally, I don’t recall any examples of serendipitous events. I am probably too skeptical and dismiss most (all?) coincidences as biases on my part, which makes it harder to remember any specific examples.
Maven is a new kind of social network–a serendipity network–that doesn’t work like the usual popularity-contest style of network you’re used to experiencing everywhere else. We don’t have likes (and therefore don’t count them) and you don’t follow other people’s accounts. Instead, you follow interests, and your feed is a reflection of the interests you follow.
Long explanation about why Maven uses 'minimal criteria' instead of maximization to compose the feed:
(…) So what then does quality really mean when you’re talking about rating all the content in the world? What should we be looking for? That’s where an understanding of open-ended systems comes into play.
The difficulty of judging subjective interestingness has long held the attention of the field of open-endedness (Open-endedness: The last grand challenge you’ve never heard of – O’Reilly ). Many ideas have been proposed and one of the simplest of those is simply to stop trying to assess and optimize interestingness. Instead, the idea is to check every candidate against a threshold, called “the minimal criterion,” and then to treat those above the threshold differently from those below it.
Notice how this idea is radically different from maximization (e.g. maximizing likes or maximizing follows). In a maximization paradigm, you are always incentivized to get more or climb higher, so nothing is ever enough. That’s one reason for all the pathologies we see on social networks. In contrast, with a threshold, we don’t even try to judge what’s better than what above a certain threshold, which actually makes a lot more sense.
After all, once you cross a minimal level of quality, does it really matter how high a score something gets? Because it’s subjective, a post with a thousand likes could be less interesting to you than one with five hundred! At some point, the numbers just stop mattering for anything other than the most generic sense of appeal.
By taking this point to heart, the minimal criterion is able to generate and expose far more diversity while still maintaining quality. The vast majority of what you see is above threshold, but beyond there you get a much broader exposure to potential serendipity.
For these reasons, Maven maintains quality through minimal criteria, which are based on engagement statistics (not likes and follows), and above which all posts are treated equally.
Bibliomancy is a form of divination in which you ask a question, or just state a context in which the result will be framed, and then riffle the pages of a book and drop your finger at random. You can do it with passages, but I like to do it with single words.
Incorporating ‘epistemic luck’ could be effective in creating serendipity-conducive environments and protocols, especially in dealing with scepticism. In other words, it might promote tolerance for doubts about the validity or value of serendipitous discoveries.