Some time ago I realized, with mild panic, that our always-on, real-time communication channels weren’t going away. As I was gulping down the day’s feeds along with my morning coffee, it occurred to me that even if I wanted to, I couldn’t really opt out. My refresh twitch is so habitual now it’s almost hard to remember just how experimental things like the early days of Twitter felt.
Of course it once was, like all new things. The real-time web started as something we did because we could. Technological advancements like more efficient ways to retrieve large amounts of data, the cloud, and the little computers we now carry around in our pockets made it just a really sexy problem to solve. Successful experiments turned into trends, and those trends are now becoming unquestioned convention.
But is real time always the right choice? Do we even want everything we consume to move at this pace?
We’re just embarking on an instrumented era of logging all our personal data and making it available instantly, yet diminishing returns have already started to set in. We struggle not only to keep up with each other’s data trails, but more importantly, to know which crumbs in those trails are worth picking up, as well as how to find them again later—like when you want to relax on the sofa after a hectic week and you know there must have been a bunch of cool things to listen to or watch that flew by on Twitter, but gosh, where are they now?
Why is this? The two companies that pioneered real-time interfaces—that brought their users kicking and screaming from one redesign to the next as designers struggled to invent new ways of consuming; whose whales were failing left and right as technologists struggled to tame the data—what do they have in common? Facebook and Twitter are communication channels.
I have a hunch that when we invent new things, the first way we test our new technology is with talk. Our ability to communicate is simply one of the most basic use cases in the design problem of our lives. And not only is it essential and important and the rest of it, it’s fun. It makes us laugh. Why wouldn’t we?
This real-time barrage of voices works well for talk, because talk is fast, easy, effortless. We do it constantly. So what about things that take longer to make and consume: a song, a book, a film? Trying to squeeze these types of media up into the high-frequency end of the spectrum and expecting that we’ll enjoy them whizzing around our heads at the same speed as our daily chatter might create a missed opportunity to explore a whole other end to the spectrum of pace for personal data!
Finding the right pace#section3
At the music company Last.fm I worked alongside Matthew Ogle, helping to shape a service from a firehose of 800 scrobbles per second. When he and I left our respective product and design posts, we started talking more seriously about our remaining itches in online music. We realized we were missing a place where we could wholeheartedly express our musical taste in a way that’s more distinctive, more singular, more…tender.
We’d also seen enough attempts at real-time song sharing to know that while having songs tick by like stock prices may look cool, it’s not that valuable: How do I know which of those songs really matters to you?
This got us thinking about a service based around just a single piece of data at a time. What would a music service built only from “favorite songs” feel like? This raised the next question: What’s the right time-scale for a song? We took a guess that you might have a new favorite song about every week.
These two constraints—one song at a time, for a maximum of seven days—led the design of our new project, thisismyjam.com.
As builders of the next generation of web products, we need to consider the right pace for the personal data in question. Pace should inform how that information is presented, contextualized, and delivered to the user.
Lovers in a dangerous time#section4
IRL, this right pace—or “accepted lengths of time” for media things—is typically bound by the physical limitations of its particular medium, the people who make it, and how it is consumed. As different media industries matured and refined, we found the right pace for their outputs.
For example, when the American superhero comic was maturing as a medium, an artist could draw about a page per day. If you take a month and subtract eight days for weekends, that’s the standard length of a monthly single issue comic book—22 pages. It’s a somewhat arbitrary cycle, but it works with our other arbitrary monthly cycles, like getting paid. If comic books start coming to shops more frequently, and you have a limited budget, then all of a sudden your focus as a fan and a consumer is disrupted.
The single in the music industry is another example of pace. It’s hard to expect everyone to sit down and listen to a whole new album, but one song? That’s doable. It’s not surprising, either, that the first rule of marketing a single is to allow it enough time to become an earworm before the whole album is released. People need time to take it in.
Constraints breed creativity#section5
As media mature, their pace becomes not just a standard, but a helpful constraint that inspires further creativity.
Constraints have long inspired people who create, and the same is true for you and I. The limitations of the products we use every day inspire us: Twitter’s 140 characters; Instagram’s one photo at a time; or back in 2006, working within the range of code you could hack to make your MySpace page look unique.
Any network based around the concept of self-expression—the creation of personal data—will be more fun, easier to get started with, and more likely to create whole new genres of art if it includes limitations.
As easily as technology allows us to erase constraints, it gives us the power to create new ones.
Like many, I believed that attention data was one of the most valuable types of data to collect when building an online service—because of its honesty. However, at Last.fm I learned that attention data is only valuable in aggregate. The “cold start” (when your profile is empty and recommendations are useless until you start scrobbling) was one of the biggest design challenges I dealt with. When the unit of data is so small, and created so passively, you must reach a tipping point before those single units add up enough that you can extract some value out of them for the user.
A unit of data like your current favorite song may not be as precise, but it’s a unit that carries a ton of human meaning. Asking someone what song they’ve been into lately is almost always a good conversation starter, and a lot can be inferred and asked about based on it. A favorite song is instantly valuable, and a handful of them can go a long way.
I’ve been calling this notable data. Knowing what song was the soundtrack to that summer, or why you’ll always want to wiggle to that guilty pleasure from the ’90s, or which track you want played at your funeral, is a piece of personal data so weighty that if done right, it can create network value almost instantly. And its value is twofold: it’s not just that it’s more special; it’s scarce. How many favorite songs can one person have?
In the physical world, the scarcity of something, like a Stradivarius or Michael Jackson’s glove, is one of the driving factors of its value. Online, scarcity is almost a forgotten word. But maybe we just need to explore it in a different way?
Notable data starts to get really enticing when that single piece of data is crafted—contextualized with other pieces of data to make it even more valuable. It takes time, so it’s bound to happen less often. A Foursquare check-in with a tip has more value than just a check-in; an Instagram photo that’s been run through three different apps to get that perfect effect has more value than one that uses the standard filters, which still has more value than just a crappy camera phone pic.
Pace and value: an inverse relationship?#section7
If scarcity breeds value, where can we find it online? Our time. In the virtual world, where we can make endless copies of data and “limited editions” don’t exist, the one thing that prevents us from doing even more than we already are is the limits of our brains (and our sanity).
When we pull down to refresh and find a little gem of digital craft—not just an automated personal stat or an off-the-cuff remark, but something that took time to make—it’s delightful. It’s valuable. But if these valuable, scarce things are slow, you may ask, won’t they always be niche? Like slow food and artisan coffee, a rounding error in a world of McDonald’s and Starbucks? Am I just some kind of internet hippie who thinks we all need to take a deep breath and slow down? No, there’s actually real value in this model. Let’s do the math, using This Is My Jam as an example:
- Let’s assume you have a favorite radio station. Let’s also assume the most new music you can enjoyably consume in a day is an hour, tops. (Listening to all new music is exhausting; there’s a reason most radio stations play stuff you’ve heard before).
- The average length of a pop song is 3.5 minutes, so you need about 17 songs for an hour of new music.
- 17 songs per day x 7 days per week = 119 new songs per week.
- If 1 This Is My Jam user = 1 new song per week, you only need to follow 119 people for an hour of new, handpicked songs to listen to every single day. (My guess is there are also a lot of people who would be quite satisfied with only an hour of new music per week—which would require following 17 just people.)
When you compare that to the number of people you follow on Twitter or are friends with on Facebook, it’s probably not as many, or maybe just teetering in the same range. Yet even at this slow pace, it satisfies the use case of discovering new songs, because it’s a network that’s built around the right pace for music.
Go forth and explore the spectrum#section8
In 1967, when describing the community of the future (our present), Marshall McLuhan predicted “electric circuitry has overthrown the regime of ‘time’ and ‘space’ and pours upon us instantly and continuously the concerns of all other men.” He was right; this is the real-time state we’re currently living in.
I believe it’s time to envision another community of the future—one slightly less dystopian than all information and media pouring down on our heads, whether it be night or day, whether it makes sense for that content to travel at high frequencies or not.
As the people who build this next vision of the future, we must consider pace.
If it helps, use analog metaphors to dream up limitations that help create that right pace. Experiment with speed. Try letting this drive the design principles of your work: If it should be at the fast end of the spectrum, how does this dictate how the data should be presented and delivered to the user? Likewise for the slow end of the spectrum: What’s the best context for your product? Twitchy pull-to-refresh data works well in your pocket, but what about for the best films your friends have watched this month?
If there is an exciting bit to the slow end of the spectrum, one that plays with scarcity and value, what do we have to lose by investigating it? I mean, the real-time world will always be there when we want it.