Thursday, October 16, 2014

The trouble with social


Despite the global hubbub the other week about the new ad-free social network Ello, I don’t think I’m going out on a limb here if I suggest that the trouble with social media is not the ads. It’s not even the non-stop data collection grinding away in the background. The problem is all the noise. And you’re not just imagining it. The average number of Facebook friends has almost tripled in the past five years, bringing us to 338 for 2014.


*Note that these numbers are estimates, as averages vary from one region to the next; e.g. averages in the U.S. are higher than worldwide averages. These figures are generally representative of overall averages.

A template for a website, or a delicious dinner?
During this same period web design has gone from the bento box style grid to templates that are based on scrolling newsfeeds, like Facebook, or the parallel vertical columns of dashboards, like Tweetdeck. In other words, a zillion things happening all at once.

Unless you have a lifestyle or job that allows you to spend hours every day scrolling through the ever-growing feed you’re just not going to see it all. The inherent quality of a newsfeed is that it is always moving, and is ephemeral. If you miss it, that’s generally it; though I realize that the Facebook feed re-inserts items when comments are added, which drives most people crazy. But still, the point remains: the feed is a taskmaster, and really, ain’t nobody got time for that.

I don’t know about you, but I’m getting to the point of Facebook fatigue. So does this mean I'll quit? No, I won’t. But, it does mean I am actively looking elsewhere for ways to unearth the myriad things I’m interested in online. I know they’re there...somewhere, I just don't always find the best paths to them. Serendipity is great if you've got all the time in the world. But again, ain't nobody got time for that.

That’s why this quote from one of the founders of a new site called Milq jumped out at me:

The Internet should make it easier to find what you’re not looking for

- Milq co-founder Don MacKinnon

Yes, yes, and yes. So I started googling the company, and read the handful of articles I could find online. Then I played around  on the site itself. It's a place where playlists of ideas are collaboratively created. A place where you don’t have to know exactly what you’re looking for in order to find the embarrassment of riches the Internet has to offer in the way of arts, entertainment, and cultural content. That means everything from movies to design to comedy, fashion, books, music, even sports, organized thematically using something that Milq calls 'beads'. It‘s not a social network, but more of a 'taste network', to use the phrase of one of the company's founders.

For example, there’s a bead called ‘hitting the bottle', devoted to scenes of over-imbibing.




You go into the bead and there you’ll find a community-curated playlist of clips on this theme. We're talking Scotty in a drinking game on an episode of Star Trek, exchange student Long Duk Dong in Sixteen Candles, scenes from Withnail and I, and the inevitable clips from the Thin Man series of movies.  




And if you want to move laterally to other collections of thematically related clips, a series of related beads is offered in the margin, that looks like this:




Okay, this is working for me, I thought to myself. This is what I wish YouTube did for me, but does not. So I did a bit more rooting around and I discovered that Milq has two offices: one in New York, and one in Toronto. Well then, let’s get down to business. And so I did. I got in touch with Milq’s co-CEO Jordan Jacobs and within 48 hours I was in his office, overlooking the sprawling U of T campus, smack in the middle of downtown Toronto. What follows are highlights from that conversation.

If I like something it’s not because of an algorithm, it’s because of a person

JJ: The mid 90s dream of the Internet was everything all the time. We didn’t quite know how we were going to get there, but just the idea that we could potentially access everything on our screens was mind blowing. But, twenty years later we can see how feed-based information and sharing via Twitter and Facebook is not the answer. They’re not designed for discovery, they’re designed for self-expression, and those are two very different things. 

What we’re doing is designing for things that people don’t yet know they’ll love. And we’re not doing it just with algorithms that make recommendations. There’s a real human at the source of every bead, and a real human at each point along the way as the bead attracts comments and additional links. We use something called Machine Learning, which is a way for the system to teach itself as it goes, and U of T (pointing out his window) has one of the best programs in the world for it.

The structure of the Internet has killed browsing

JJ: With anything that’s subjective – like movies, music, books, art – it’s almost impossible to find what you’re looking for online.  At best you’ll find what you already know you like. Or you’ll get overwhelmed by things like Spotify. Or you’ll just go back to the same handful of things you’ve always liked. When we were just starting out we would spend hours and hours together on Google Hangouts every day, asking ourselves ‘how do you do a content-centered network’? If there’s, say, a hip hop expert in Mozambique who knows something that nobody else does how do you bring what he knows to the surface? How do you architect something like that? And we came to the conclusion that you do it with collaboratively created playlists. And you add tons of metadata and you create influence graphs for each person who posts, comments, or clicks. And then on top of it you add that layer of machine learning, where the system learns from itself as it goes.

Achieving high quality on a massive scale

JJ: I remember when (our CTO) Tomi, who was one of the creators of Yahoo Answers, told us about how Yahoo spent something like $10 million on an advertising campaign for Yahoo answers in Times Square, after the site had been up and running. They wanted to make it bigger and better and thought that a big ad campaign was the way to do that.  And what happened? The user base went up. Okay, good. But -- the quality of the content went down. Way down. 

For Milq we don’t need a huge user base in order for this to work. We can handle a huge user base, and I hope we get a huge user base, but it’s not the most important thing.


Initially we self-funded and then we made the rounds and talked to the VCs and many of them thought we were crazy. Others, particularly in Silicon Valley and New York, got it, but then we stepped back and re-thought things this way. We wanted to build a strong advisor and support system so we went the route of taking investment from a variety of smart, experienced, and connected people we knew from our backgrounds in media and tech. And that’s when it worked. People like Tom Freston, former CEO of Viacom and the guy who founded MTV and is now an advisor to Vicecame on board.

And now we’ve got broadcasters coming to us, and people with these huge storehouses of content that they don’t know how to get out there. They’re saying to us: ‘we have a huge problem’.

What it looks like in action

Film Festivals

JJ: Every year film festivals are tasked with having to completely rebuild their websites, starting from scratch. What’s playing where, at what times, what are reviewers saying, which films are most popular, who’s getting interviewed, etc., for hundreds of films.  What we’re able to do is offer something that builds around themes, about the way people think and talk about movies…not according to names of directors or their countries of origin. Here’s an example of what we did with TIFF (Toronto International Film Festival) and another example of what we did with the Tribeca Film Festival. 

Bands on Tour

JJ: Kings of Leon used the site for a “what song should we cover”, put out to fans on their last tour. Instead of a tweet or Facebook post they got a dynamic playlist, with links to Soundcloud, YouTube, Vimeo, etc. And anyone can jump in at any time.

Beads for Books

JJ: We just announced a partnership with Simon & Schuster, and a few authors have already started beads, like Walter Isaacson, the Steve Jobs biographer. He just started a Great Digital Innovations bead, and Anna Todd, who has over a billion reads of her fan fiction on the band One Direction on Wattpad, has started a bead too. And, of course, anyone can post audio or video related to their favorite books and start their own beads.

What you wish watercooler talk was like

JJ: Graydon Carter, the editor of Vanity Fair, started this bead: Metaphors for Failure. You’ll find things like the original ad for the Edsel, the SNL sketch on the launch of healthcare.gov, newsreel footage of the Bay of Pigs invasion, and the launch of New Coke.




And now back to me, your unembedded blogger. Moving from one metaphor to another, there are a lot of beads in the world that could benefit from being brought together on a string. As Milq’s Jordan Jacobs put it: “in a way we’re kind of building the great library of Alexandria online.” And while things still aren’t perfectly intuitive at Milq -- there’s a bit of a learning curve when you go to the site and try to figure out how to best make it work for you -- bear in mind that the site has only been out of beta since September, so these are still very early days. (I’m guessing the first few months at the library of Alexandria were a bit rough going too.)


The Great Library of Alexandria.
How they rolled pre-wifi and Wikipedia


The larger point for me is that all the gems that live online, the things that we may not even know we’re interested in, to echo the comment of Milq co-founder Don MacKinnon highlighted earlier in this post, are pulled out of their hiding places online and brought to the surface. And this, I believe, is the next phase of where things are going online.

We know that the wisdom of the crowd works pretty darn well online, because we already have the prerequisites, like extreme heterogeneity of opinion, non-economic actors, community policing, and the ability for anyone to chime in with a viewpoint. Just look at Wikipedia. Written by no one in particular, with labor supplied voluntarily and free of charged, editorial oversight by the community only, and results as, if not more accurate than Encyclopedia Britannica. Should it work? Absolutely not. Does it work? Indisputably yes.

And then there's Google, whose stated mission is to organize the world's information. I can't imagine life without it now, but I have to say it works a charm for things like finding general information, or news, or reviews of electronics or suggestions for headache remedies, but is decidedly less helpful when it comes to things involving value judgements or tastes. Just look at Google-owned YouTube, a modern day miracle in terms of the bounty it holds, to be sure, but if you sit down in front of your screen and just go to YouTube.com you're not that likely to get matched to the content most interesting to you. And it's not that YouTube hasn't tried to fill this gap. They have. Sadly, we only have to look to YouTube's original channel experiment of 2011 to 2013 to see that investing a reported $300 million on about 200 individual YouTube channels didn't go very far in terms of solving the problem.

I would say this is because they were still thinking largely in terms of television tropes, of favoring traditional media stars, and of organizing programming by channels and subscriptions to those channels.

And really, isn't a channel:

a) just another form of more or less linear programming
b) something that lives within a walled garden environment

And aren't these exactly the kinds of things the online environment was supposed to liberate us from?

Maybe that Internet we dreamed about in the mid 1990s, the one that would make anything and everything we could imagine available to us, in a way that is both efficient and a little bit transcendent, is now a few steps closer. But we're all going to have to kick in, because coming up with a captivating list of metaphors for failure is not something that can or should be left to just one person.