Lifestreaming apps and designing time and flow

Twitter's success over the past year has given birth to a new category of social media applications. Lifestreaming apps, known also as flow applications, permit users to publish a steady stream of online activity. Readwriteweb.com has a primer and later published a roundup of 35 lifestreaming apps, some of which are already defunct. Where these apps aggregate comments, friends, content topics, and media types, they can also be categorized as aggregators of distributed conversation (see the Techcrunch article on Friendfeed for more.)

Lifestreaming applications pose an interesting challenge to designers. From the perspective of social interaction design, site organization, navigation, publishing rules, and content organization shape the user experience, and thus the social practices that emerge around lifestreaming. Twitter has set a convention based roughly on a hybrid of email inbox and chat: tweets flow from newest to oldest. Third party twitter apps hew to the convention for the most part, eschewing additional navigation or structure for the simplicity of the stream. Broadly speaking, lifestreaming applications serve as a news wire service or news scroller of personally-relevant announcements and messages. The content sourced for publication is selected by the user.

I suspect, however, that we're only at the beginning of the design cycle of these applications. Now that we have established the utility of a twitter (or friendfeed, facebook status, activity, or news feed) as both social and personal utility, focused around talk and speech rather than writing and publishing (e.g. blogs), we might anticipate diversity. This, I suspect, will come as it normally does in variations of the apps themselves, and in their application to social practices. We have spent much of this year on the tools, technologies, and companies providing lifestreaming applications but relatively little time exploring their user and social experiences.

Consider the user experience of time-based talk vs page-based talk. If most of the social web is organized around the publishing/print/web page model, which subordinates chronology to topicality, then lifestreaming tools do the opposite. They subordinate topicality (search, browse, drill down, categorization, relatedness by tags, taxonomy, etc etc) to the flow. Flow privileges the present, not the past, and not the enduring. Flow apps put the user in the flow (assuming that s/he is paying them attention), aggregating the multiple times/presents of one's friends into a common stream. They give the illusion of togetherness, as does any aggregation of content online, but in the now, in time, rather than in place, such as on a page/site. In fact this illusion works greatly to Twitter's benefit -- this sense that while each of us sees a unique timeline, we feel that we're on the same page (!). Most of us do not use Twitter for search, browse, or navigating content, but for a sense (foreground when we use it; or background when it's on standby) that we're "there." "Being there" is a matter of being in time.

If aggregating timelines is the design challenge addressed by lifestreaming apps, the current basket of sites and services leaves much room for innovation. By which I don't mean improvement. Social web design is iterative, to be sure. Not only are we all in beta, but each release of functionality or design updates engenders new user experiences. And as new user experiences accumulate and coalesce, new social practices take shape. The UI of social media is the social interface. Page-based social sites have been developing for years; lifestreaming apps are by contrast relatively new.

The techniques we use in designing page-based services haven't yet found their way to time-based apps. Scale, rank, featured, comparison, grouping and categorization, tagging, and more. More significantly, the value in time-based apps ought to be content over time. So, in this case, talk over time. Imagine twitter snapshots, timelines, and histories. Time-based apps will have rhythm and pacing where page-based apps don't. Time-based apps have moments, episodes, periods (of time).

News, for example, has its message content and then its urgency. The significance of news is as much a matter of its arrival as it is what is said. News is one of those strange kinds of message whose importance is announced on its envelope (Urgent!). Given that twitter now serves as a newswire, and is used so often for news (and not "What are you doing", which is rarely newsworthy), we could imagine interface solutions that explore the temporal dimensions of talk and speech over the content dimensions (which have been mined by search, browse, and other page-based navigation conventions).

To get more specific, and to explore these thoughts further, I will address these ideas further by looking at several lifestreaming apps in the days ahead.

Related reading
Stowe Boyd on Lifestreaming
Brian Solis on Lifestreaming
Mashable on Lifestreaming

Save This Page

Future of social web: system and practices

Jeremiah Owyang has posted his thoughts on what may come in the long-term for the social web, beginning with the increasing relevance of activities like friending: Why 'Friending' Will Be Obsolete. He writes that as the system learns about our behaviors, preferences, and relationships that it will be able to automate and supply information we normally have to declare explicitly today. I couldn't agree more.

Jeremiah summarizes his model like this:

"The System: The system is the combination of all websites combined, it's a massive data base of content, clicks, search terms, time on site, shared posts, wall posts, links, and tweets.

Teaching the System: Humans are constantly speaking in machine language, from use of hashtags in twitter, or boolean searches in Google, or even from the act of friending folks in your social network. All of these behaviors are humans teaching the system how to understand us, so it can better serve us.

The Intelligent Web: Software that is able to collect and make sense of all the data in the system and is able to deliver meaningful content back to people in context -- often without us saying or gesturing that we need it."


The web was built on links between documents -- objects -- and since it's inception has grown to accommodate not only many different object or media types, but their relevance, popularity, and other measures of use also. In fact links on the social web need not always point to the same thing. Social navigation in the form of a top-ten, for example, points to not only a changing set of top ten items, but updates itself as it is used, thus reflecting social use.

Behind Jeremiah's vision of the future is the system's interest in capturing and recontextualizing its own use. If the static web was merely a network of static connections, the social web is a dynamic network of changing connections. If we assume that social use will remain a priority for web builders and designers, applications and their businesses, then the relevance of information provided by the web will likely be qualified along two axes: the personal and the social, or the particular and the general. The next generation web, in systems speak, is a second-order observer system. Meta data supplies a second order observation of its own use: the web knows not only what it publishes but also how users interact with it.

Because the system is open, is dynamic, and is always in use, the new system is not a static collection but a dynamic and changing set of connections -- connections whose relevance to an individual user and to the audience in general change over time. The next generation system has time. The first generation system did not.

I see, or would like to imagine, a system whose links are no longer document links but are instead "views." Each view (link) of information might then take into account meta data along our two axes: one user-centric, the other social-centric. A user centric view would be informed by my past history and tacit (learned) and explicit (declared) preferences. My tastes and interests, in other words. The social-centric view would be informed by social usage, social ratings and votes, interests, trends, and so on. I might use sliders to set the view I want on a social site -- stuff for me or stuff socially organized.

There is another development coming for the next generation system, and that is the temporal organization of system (vs spatial organization). The topic comes up in discussions on lifesreaming and flow apps (which I'll discuss soon), and often takes the form of talk-based apps vs page-based apps. Twitter, for example, is not page based: it lacks navigation, topical organization, topical layout, and so on, choosing instead the temporal organization of content used by time-based apps like IM, chat, and email. As more of these apps innovate, become more visual, and go mobile, time-based interaction tools will mature. We'll have two modes of interacting with the system: from within the river of flow or from its shores: watching as it streams past.

Innovation of late may have produced many look-alikes. But it's when things begin to look alike that exploration begins anew at the margins.

Save This Page