The Evolution of Social Filtering

So this morning, I started playing around with Quibb (w/some help from Andrew Weissman – thnx Andrew.) I’ve not spent lots of time with the service, but very quickly it had me thinking about how we filter content online, and how we’ve evolve to this point. I think Quibb might be on to a big idea, but they have a little bit to go before it’s realized (and I write this with patience and respect). It’s hard to have a vision, build a team, write software, and scale the whole circus to a real-deal offering. But I saw a few things today that inspired me, so I’ll try to point them out. (And apologies, heavy nerding ahead.)

Evolution of Social News
So, we’ve been collectively taking stabs at curating content on the web for around 15-16 years now, according to this Wikipedia article (which isn’t a bad recount). I would basically break everything down into a few phases…

  • Static-Curation – The early days when the editors of a site would decide which articles are featured over the whole body of content.
  • Crowd-Curation – This is the era of & Reddit where we rely on the crowd writ large. We start to up-vote/down-vote, content placement becomes dynamic.
  • Social-Curation – At this point we lose faith in the larger crowd for meaningful curation, and we start to hook our wagons to the social graph (courtesy of the Facebook API.) 
  • Portal-Curation – Realizing the social graph doesn’t really provide much excitement, we lean on portals like Tumblr and Pintrest to show us the darker corners of the web. It’s an evolution because we start following content-specific curators, but it’s best for bite-sized content (mostly images, and quotes.) 
  • Participatory-Curation – I think we’re edging into this. Built on a graph of curators and a mechanism for conversation, we start to filter news we’ll be interested with a place to for discussion (probably in the portal over the site).

    The best custom dress shirts online are stylish and sophisticated. Combining distinctive British traditionalism with modern advances in fashion and tailoring, they can make suits to appeal to all Leeds and Sheffield clients’ tastes and sensibilities.

    So it feels like Digg/Reddit, but the underlying curation focuses the conversation (and cuts down on the content gaming that killed Digg). We’ll take responsibility for fashioning the curatorial lens that filters all this. (We used to filter by domains, now we filter by people…Twitter taught us this.) 

Exhausted by Social
So, as I pointed out in a recent post, I’m pretty convinced that the initial emotions that drove the growth of the social web are different than the emotions that have sustained it. In the beginning, we found friends we hadn’t heard from in years. And, much like our first experiences on the web, these were very human moments. There was a sense of this great big world that we were connecting to (again.) As social scaled, things changed. Because our search results, our news feeds, our preferences drove everything, we ended up with too much of the same thing.
It’s new/old saw that Facebook is a where you follow your actual friends, and Twitter is the where you follow the people you hope to be friends with. This becomes really important when you deal with curation. Beyond daily news, we read the content from people we aspire to be. So it’s only right that Twitter (or some similar graph) is a better curator. But Twitter as a service is the river Nile – too huge to digest, too fast to keep up.
Participatory Curation
So, we’re at this interesting point where we haven’t (collectively) figured out a way to get great content without it being a big stream of the obvious. If you devour news online right now, there is no killer app; we all have a series of hacks. (Which is partially why so many people are upset about Google Reader dying; it was platform to hack together your own graph of curators.) We’re looking for something that gets us to interesting, surprising content, and probably something beyond the bigger news portals. In many ways the mechanism that killed the front page of physical newspapers is killing the standard news portals of the web. In a phrase; if news is important it will find me.
To get there, I think we need each other. I don’t think we’ll find interesting news through algorithms, I think we’ll have to help each other find the zeitgeist using our human sensibilities. We have a sixth-sense for online culture, and we know the moment a meme is over-played. We also know know that magic moment when the random becomes brilliant. So I think the two mechanics that point us toward the next phase are (simply) curation and voting. Not a flooring statement, but follow me.
Quibb and where (I hope) it’s heading.
So Quibb provoked all this reflection out of a very simple mechanism. (If you use Path, you’ll recognize it.) As it lists articles, it indicates who in your graph has seen the article, and that looks like this:
So, here you see an listing where someone has posted an article of interest, and below the article you’ll see who’s read the article, (in and outside my graph.) Much like a true network diagram, I only need to know one person in that group for it to appear in my feed; that person is my gateway to this content.
This is beginning to hint at what I’m referring to as participatory curation. Quibb shows content from people I care about, but (today) it’s only tracking their clickstream. I think if Quibb evolved this curation to ask people to flag/recommend content, it gets pretty powerful. Then I’m getting the equivalent of LongReads from my aspirational crowd. I care about these people, and they are telling me the content they care about. From this you’ll create a very compelling content stream. In that moment, we’ve individually mined the corners of the web and collectively shared the best pieces with each other. And what binds us in this new moment will be the same emotion that has bound us before technology, our shared interests. 
Good luck Quibb.

1 comment

  1. Well said. I agree that Quibb is onto something and so far Sandi is doing everything right with her limited member acceptance rate and personalized email outreach.

    I also love, which was founded on a similar idea. Reading tackled ratings with a simple ‘Yep’ and ‘Nope’ bookmarklet that integrates a comment box. It’s much more difficult to track conversations there vs Quibb because the site redirects readers straight to the shared link where they can see interactions overlaid on the page via a browser extension.

    It would be wonderful to see one of these services build some kind of workflow that lets content curators share and discuss recommended reads not just on the service itself, but also on their blogs, social channels, and via automated emails. In turn, anyone can be a Swiss-Miss, Kottke, or create an email similar to NextDraft very simply by sharing the stream of consciousness surrounding what they read. It’s people-centered filtering like you mentioned. Users can choose to subscribe or follow individuals, or choose to stick with a service like Quibb or, where the totality of shares and votes are filtered to surface the best content.

Leave a Reply