I try not to write longer posts but I missed the boat on getting the following thoughts out of my head as they came to me. This one has been brewing for months and solidified at SXSW after listening to an Activity Stream panel and a Recommendation Engine panel back to back in the same room. The social web is at an interesting crossroads. Part growing up, part progress/innovation, part adoption by people beyond our insulated tech culture. With Twitter’s unusually quick growth and the issues emerging around Facebook’s re-design I am really excited to see further discussion and progress on two related topics. Ambient awareness and ambient findability.

Ambient Awareness: The product of all those status updates

The idea of ambient awareness caught a little buzz in the social web community through the New York Times article, “Im so, digital close to you.” In the midst of putting together this post I have seen it re-surface a few times as focus generally shifts towards activity streams. You could classify “ambient” as a flavor of the month buzzword, its definitely percolating, but I like it because it forces the question – “ambient awareness,” great, now what?

“Social scientists have a name for this sort of incessant online contact. They call it “ambient awareness.” It is, they say, very much like being physically near someone and picking up on his mood through the little things he does — body language, sighs, stray comments — out of the corner of your eye… But these new updates are something different. They’re far shorter, far more frequent and less carefully considered.” -NYT article

Findability: Can we get more out of activity streams?

Most people have seen someone struggle to walk a huge dog down the street. Well activity streams are walking us right now. Their difficulty to manage is shaping our behavior. Unless we elect to cap our follows/inputs at a certain level the signal to noise wall comes up pretty fast. Their are also people who are cool with the flood and sifting through an overwhelming amount of information. Thats fine for certain uses, but I would hope that this new aspect of the social web experience has more to offer. Echoing something my fellow Barbarian Rick said to me… This stuff is great for broadcasters right now, but not so great for readers who are seeking specific things. In a perfect world ambient findability would flourish along with ambient awareness.

“Ambient findability describes a world at the crossroads of ubiquitous computing and the Internet in which we can find anyone or anything from anywhere at anytime. It’s not necessarily a goal, and we’ll never quite arrive, but we’re sure as heck headed in the right direction.” -From an interview with Peter Morville on Boxes and Arrows

Ambient awareness is not the be all end all of activity streams. Its only one aspect of a significant shift in how we use the web. I find it hard to be ambiently aware of 100+ people every day and get anything out of it. What happens when the noise eclipses the signal? Ambient Findabiity by Peter Morville is one of my favorite books about the web and seeks to answer some of these question. Peter approaches findability on the social web from an interesting perspective. In 2005, when asked about the death of taxonomy according to Tim O’rielly, Peter Responded…

“People have been predicting the end of hierarchy since the beginning of hierarchy. But it’s not going away. In fact, I dedicate a whole chapter to explore the hyperbole that swirls around social software and the Semantic Web. I make the case for a “sociosemantic web” that relies on the pace-layering of ontologies, taxonomies, and folksonomies to learn and adapt as well as teach and remember.“

I found it interesting that the pace-layered approach Peter discusses came up at the SXSW09 Recommendation Engine panel but not the Activity Stream panel. We have a wide open faucet of activity streams across the social web inundating us with a barrage of status updates, photos, links, and other stuff. Unfortunately we are learning that an activity stream focused social web easily slips into information overload/fatigue with little effort. There needs to be a collective shift towards addressing findability within activity streams. Its not the flood of status updates (a common complaint) that “ruined” the new Facebook. Its the lack of tools to filter information of any kind down to what each individual wants. People are creating the status updates and other activity, and they are doing it more and more, its not going away. The tools need to catch up. Many aspects of what Peter Morville called the pace-layered sociosemantic approach are underway and will help solve these issues. For example the DiSo project is focused on issues of interoperability and meaning in activity streams. Its exciting to think about the potential user experience of these tools once some of the bigger technical issues on the table are resolved. Thats going to take time, but in the meantime there are some things designers and developers can focus on to ease the pain as we transition into a activity stream focused social web.

Filtering activity streams: The “all filter” is broken

What do you do when the noise eclipses the signal? Everyone has their own threshold for their email, news reader, twitter, etc. Imagine if one day you could not use your email filters, folders, labels, rules, smart playlists etc. Overload, saturation, fatigue, game-over man! Thats about where we are at with activity streams on the social web. After a certain amount of information flow the “all filter” just stops working for anything more than a casual glance. There have been some strides toward better filtering lately, but basic problems have not been addressed. The trajectory of filtering UI for activity streams is a traditional and very familiar vertical list of sources on the left and feed items on the right. Its easy to see why this works for applications that deal with a live or frequently updating feed of information. Not the most innovative UI, but its a safe and effective pattern. Google reader is my favorite example of this style of filtering UI on the web. Nambu the OS X Twitter client does a nice job of utilizing this UI style for twitter. And of course FriendFeed and Facebook among others. In addition to the all filter, what are some ways we can or should be able to filter activity streams?

Filter by Group – Twitter clients, Friendfeed, Facebook and others offer grouping features which allow us to create clusters of people to filter activity streams with. Groups are best used as a way to collect people that have something in common which may only be relevant to you personally, not necessarily for mapping real life relationships (co-workers, family etc). There is a lot of potential power in telling a system that a group of people are related, and even more power if you can tell it why. For example, I group people that are into the same obscure sub genres / cultures of music that I am interested in. You can imagine I do this for other topics, ideas, interests etc. I realize that this behavior may be on the power user side of the spectrum right now, but it has become essential to me effectively using Facebook and Twitter as a consumer/seeker of information. I imagine there could be some interesting outcomes of telling the system – “people in this group are related… now show me trends within the groups activity stream… and I will teach you (the system) which trends are related to why I have grouped these people together… then show me more of those relevant blips in the stream.” Unfortunately creating and maintaining groups on the social web is a absolute mess right now, but this is a topic of another (shorter) followup post.

Filter by object – Ambient awareness of WHAT? Relationship statuses, what someone had for lunch, peoples moods? If that floats your boat then cool. But what else are you into? What makes you tick? Videos of cats, tweeting about your feelings, web dev, collecting sneakers, obscure music, fixed gear bikes, knitting, politics, fencing, wine? Social objects are certainly floating around in activity streams. As more unique social objects manifest on the web in their digital forms, they become worthy filters for navigating the social stew. The more niche the more potentially interesting the results in my opinion. Integration of niche social-objects into activity streams hinges on what DiSo is working on, social network interoperability and the openness of social nets in general. In other words, the site defining those social objects need to be able to play nicely with other sites like Facebook and twitter in order for the objects to have any meaning within activity streams. TALL order there. For now, simple examples of filtering activity streams by objects can be found on FriendFeed or Facebook where you can filter down to just photos, or just status updates etc.

Filter by topic/context – This a nut that hasn’t been cracked yet. While these activity streams are full of a lot of disposable, fleeting chatter they are still often about something. There is a narrative, there is conversation, there are common threads, there is progression. Context is sorely lacking from activity streams. We have the magnifying glass soo locked to the “right now” that we cannot zoom out and see the greater arcs in the stream over longer periods of time. Something that is 5 minutes or older is easily wiped of it’s meaning on. Sure, its the real-time web, but the past and the future are of course still important, right Matt Jones? Hashtags on Twitter are a great example of how to hand-jam context and meaning into the system. It lacks grace, but it works, kind of. What if the system was smart enough to pick up on context/meaning and provide it as a set of additional filters? A question I am decades late to, but hopefully our fearless leaders are approaching an answer. There is some technical magic required that I know little about, but my brain goes back to the pace-layering “sociosemantic” web Morville put out there. Read/Write Web has some of recent semantic web happenings for further reading.

Filter by geography/location – Speaking of context, physical location is potentially a great way to glean context from information. Check out this chat with the DiSo boys and Brian Oberkirch if you have some time. Its interesting to me that Facebook doesn’t allow you to filter your activity stream by geographic network. You can filter your friends by geography, its just buried on the friends page. Perplexing. Mixing and matching location with other filters is definitely something I would love to be able to do. For example I am in NYC for a weekend. I have a group of people that are all into the same electronic music I am. I would like to filter that group by NYC and potentially an object like events.

Filter queries – There are always cases where a good old fashioned “advanced search” is sometimes necessary. Yes one of the goals of our smart systems of the future is to eliminate the need for these types of queries but we aren’t there yet and there will probably always be value in very precise detailed searches. Yahoo’s Sideline is an interesting approach to advanced twitter search. The Air application allows you to create groupings of multiple searches. These groups then produce a stream of tweets. You can include tons of queries in one group that produce very interesting results. The flow feels a lot like creating complex mail rules or smart playlists in iTunes. Its a really interesting way to follow topics on twitter without hash tags.

My gut: A shift in toolsets and mindset is coming

People may read this and say, “I just don’t use Twitter or Facebook like this, nor will I ever, I follow 50 people and it works fine.” Im cool with that, its a very reasonable approach and probably the majority approach. Im definitely approaching this stuff from a “design for yourself perspective,” for better or worse. I recognize that. But what keeps me chomping at the bit for more effective tools is the imbalance of power between the social-media broadcaster and the sociac-media consumer. Old-school broadcasters have fallen all over themselves to effectively utilize and monetize this new way of communicating. There are a ton of tools aimed at broadcasters using social media. We haven’t necessarily seen the same robust tools for the people soaking up more than they are putting out. On second thought, soaking and consuming are not the best words to represent who I am talking about. I am really talking about the finders/seekers. That is after all what the majority of people do on the web. They find stuff. I want tools for the pro-finder on the social web. Google Reader / RSS has been the undisputed hub of my information consumption for a few years. Activity streams are on track to match syndication/aggrigation, but before they can, we need better tools for being seekers of information on the social web. We have arrived arrived at an ambient model of communication. Now lets work on better finding what we want in the social stew.