Lehrblogger

A week ago Google launched its much-anticipated social network, Google+. I think they executed remarkably well, given the size of the company (24,200 employees), the stakes involved (see the ever-hyperbolic TechCrunch), and their history with social (Buzz, Wave). It was important that they launch something good, and I think what they launched is good. New features include a cross-service toolbar, Hangouts, Huddles, and Sparks, but here I’m going to talk about Circles. If you’re not on Google+ yet, the below video gives a good sense of the feature and its interface, but you should sign up and try it for yourself.

Wanderli.st, the ITP thesis project I presented last May, grew out of similar ideas about social contexts. It was an application that would let us socialize within online contexts that are like our offline contexts, and a tool for managing and synchronizing relationships across social websites. I’m no longer working on it for a variety of reasons, but the most important of them is this: Regardless of the interfaces and features of Lists on Facebook or Circles on Google+, I don’t think people actually want to sort their contacts.

Since we are so good at deciding what is appropriate to say to a given group, it seems backwards for our applications to make us define those groups before we even know what we’re going to say. In real life, the thing we want to share and the group with whom we want to share can influence each other, so our software should work the same way. There are several issues with manual sorting of contacts:

1. I know at least several hundred people, and it’s a lot of work to go through them one at a time and categorize them into Circles.
2. When I begin that task, I can’t anticipate which Circles I will need, and which will be useful for sharing. Will I want one for each place I’ve lived, school I’ve attended, job I’ve had, and topic I’m interested in? If I start with too few, I’ll have to go through my list of contacts again when I remember the Circle(s) I forgot, which is daunting. If I start with too many, the whole process will take much longer, since I have to decide for each of those hundreds of people if they belong in each of a dozen or more Circles.
3. Relationships are not always symmetric, and I don’t want to publicize how I group my contacts. As a result, Google+ Circles are private, and each user must undertake this task for him/herself.
4. Relationships change, and it’s even even more work to continually maintain my Circles so that they mirror my current real world relationships.

Others, such as Foursquare’s Harry Heymann, have expressed similar sentiments:

It doesn’t matter how slick your UI is. No one wants to manually group their friends into groups.

Even Mark Zuckerberg said today that that many users don’t manually build their social graphs:

A lot of our users just accept a lot of friend requests and don’t do any of the work of wiring up their network themselves.

Facebook offers a comparable but relatively unused feature, Lists, that lets users organize their friends, but Circles has a superior user interface that makes the categorization work much more enjoyable. Former Facebook employee Yishan Wong, however, makes a slightly different critique Google+ and Circles:

Besides such features being unwieldy to operate, one’s “friend circles” tend to be fluid around the edges and highly context-dependent, and real humans rely often on the judgment of the listener to realize when something that is said publicly is any of their business, or if they should exercise discretion in knowing whether to get involved or just “butt out.”

Google’s ideas around Circles can be partially attributed to former employee Paul Adams, who gave a compelling presentation about social contexts last year. He left Google for Facebook several months later, but wrote a new blog post that asks “two big questions”:

1. Our offline relationships are very complex. Should we try and replicate the attributes and structure of those relationships online, or will online communication need to be different?
2. If we do try and replicate the attributes of our relationships, will people take the time and effort to build and curate relationships online, or will they fall back to offline interactions to deal with the nuances?

I now think the answer to first question is “No” and the answer to the second question is “Neither.” Offline relationships are too complex to be modeled online, but I also don’t think those models are important to online social interaction. It’s worth noting how simple my social interactions feel offline – I can see all of the people within ear shot, so I know who can hear me and who might overhear, and this allows me to adjust the things I say accordingly. Furthermore, creating these contexts is straightforward – if I want to talk about something with a specific group of people, I’ll organize a time when we can all talk face-to-face. Offline I only need to keep track of my relationships with individuals, and I can adjust my group behavior based on the individuals present.

With email, my online conversations can work in a similar way. If I have something to say, I’ll think of precisely the people I want to say it to, and compose/address my message accordingly. Each person who receives it can decide if they feel comfortable responding to the initial group, or to some other group. Furthermore, email threads do not span the entirety of a group’s communication, so it’s easy to add or remove someone for a different conversation about a different topic, just as we can do in face-to-face conversations. With email, the group does not persist longer than the conversation. Facebook’s recently revised Messages and Groups features address some of these issues of social context, but those groups are still uncomfortably permanent, and the single-threaded conversation history feels unnatural.

Email, notably, has no explicit representation of a relationship at all. Anyone can email anyone else, yet we’ve reached a functional equilibrium through a combination of social conventions, email address secrecy, and filters. Despite this lack of explicit data, email has rich implicit data about our relationships, and in 2009 Google launched a new feature in Gmail Labs with little fanfare: “Suggest more recipients”. Wired.com wrote about Google+ shortly after launch, and hinted to the future of this data:

It’s conceivable that Google might indeed provide plenty of nonbinding suggestions for who you might want it your Circles. “We’ve got this whole system already in place that hasn’t been used that much where we keep track of every time you e-mail someone or chat to them or things like that,” says Smarr. “Then we compute affinity scores. So we’re able to do suggestions not only about who you should add to a circle, or even what circles you could create out of whole cloth.”

Rather than use this data to make static Circles that will inevitably become irrelevant for future conversation, Google should let the list of individuals in each previous conversation serve as a suggestion for future conversations. If Gmail is able to make guesses about who should be included in a conversation based on who else has already been included, it could also leverage the content that I intend to share to make dynamic suggestions. It can help me remember who I might want to carbon-copy on a message before I send it, and it can do this without overburdening me with the overgeneralized Circles of my past (1). Once the spatial boundaries of that conversation have been defined, the discussion can continue until no one else has anything left to say or until a subgroup wants to split off and have a side conversation, much like a social interaction in real life. The fundamental design of email has shown more promise than the categorization-based alternatives (2).

We want some of the things we say on the Internet to be public and accessible to anyone who is interested (3). For everything else, explicit persistent groupings of the people I know are tedious to maintain and unnatural to use. Each discussion is different, so we need discussion tools that support robust privacy control on a per-message basis.

Thanks to Jorge Just, Abel Allison, and Jorge Ortiz for their help with this post!
Edit: And thanks to Ninakix for some great comments on her Posterous.

1. There are many ways to improve this recipient suggestion interface, and profile photo thumbnails would be a good place to start. It could also suggest some Circle-like groups, such as my family, and even let me upload my own photo to make those groups easier to identify. It should not, however, present me with a list of all of my groups, because then that is something to manage – I only need to see the groups when I am addressing a message.
2. It is important not to let our thinking get bogged down by the current limitations of our inbox interfaces. What if, when you searched for a person in Gmail, you got a grid of attached photos in addition to a list of conversations? What if Gmail was as “real time” as Gchat or Facebook? What if Gmail didn’t make you feel like you needed to read every message? What if Gmail searches were, dare I say it, fast? Some of these changes would break how we currently use our inboxes, so perhaps a separate tool that was modeled after email would be better, but that’s a detail. Other changes, such as streamlined Rapportive-style contact information for the people in a conversation, are already beginning to be built-in.
3. I have some ideas on this, but that’s a separate blog post. See Pinterest and Subjot in the interim :)

People love to play games. They’re fun, of course, but they also teach us valuable real-world skills. Chess teaches pattern-matching and strategy, Dungeons & Dragons teaches exploration and collaboration, Olympic sports teach physical coordination, and others such as soccer and Halo teach many types of skills simultaneously. In all cases, games provide safe environments for learning as well as clear criteria for success. It’s better to learn to run fast when you’re on a field because you want to win a game, and not when you’re running from a sabre-toothed tiger because you don’t want to get eaten. Games create artificial environments and structure incentives in ways that make us better equipped to prosper in reality.

As technology allows us to measure things that were previously unknowable, we will design new games that improve our ability to live in this increasingly complex world.

Jesse Schell gave an excellent talk at the 2010 DICE Summit:
(I’ll quote/paraphrase the applicable parts of both talks, so feel free to keep reading and watch them later.)

Schell observes that many of the unexpectedly wildly popular games from the past couple of years (such as Farmville and Guitar Hero) “are all busting through to reality.” He predicts that increasingly inexpensive data sensors will become ubiquitous, and will record where we go as well as the things we buy, read, eat, drink and talk about. This data will enable corporations and governments to reward our behaviors with game-like ‘points’, when really those points are just a way to trick us into paying more attention to advertisements, and we will consent to this because we will be able to redeem those points for discounts and tax incentives. Schell concludes:

The sensors that we’re going to have on us and all around us and everywhere are going to be tracking and watching what we’re doing forever, [...] and you get to thinking about how, wow, is it possible maybe that, since all this stuff is being watched and measured and judged, that maybe, I should change my behavior a little bit and be a little better than I would have been? And so it could be that these systems are just all crass commercialization and it’s terrible, but it’s possible that they’ll inspire us to be better people if the game systems are designed right.

Jane McGonigal recently gave a compelling talk at TED that approaches similar ideas about how gaming can save the world:

McGonigal observes that games, and especially immersive massively multiplayer online role-playing games such as World of Warcraft, always offer quests that are perfectly tailored so as to be both challenging and possible. She offers four useful descriptive terms for these activities:

1. urgent optimism – gamers tackle obstacles without hesitation, and they expect to succeed
2. social fabric – games require trust and collaboration, and as a result the players develop strong relationships
3. blissful productivity – gamers work hard when playing because they enjoy it
4. epic meaning – game narratives make it easy for the players to see the big picture

These are part of a larger argument that gamers can save the world if they play games that are designed to have positive effects outside of the magic circles of the games. McGonigal cites an example from Herodotus in which the ancient Lydians survived a famine by distracting themselves from hunger with dice games, and she makes an argument that we can similarly use contemporary games to solve contemporary problems. That example breaks down, however, because we don’t need games that distract us from our problems like the Lydians did – we need games that enable/encourage us to face our problems and overcome them.

She goes on to describe an example of a game she worked on at the Institute For The Future called World Without Oil, which was “an online game in which you tried to survive an oil shortage.” The game provides online content to the players that presents a fictional oil crisis as real, and the game is intended to get people thinking about that problem and how they might solve it. But just as the first example is missing the direct applicability of the game to the real world, this one is missing the application of the data from the real world to the game, and both directions of influence are important.

When the ubiquitous sensors described by Schell are combined with McGonigal’s vision of games designed explicitly to save the world, the content surrounding the games (that presents real-world crises as ‘quests’) will no longer be fictional, and can instead be based on real-world data. The games will provide frameworks for understanding and leveraging all of this new data about the world. They will motivate us to act for the greater good through both monetary rewards such as tax incentives and social rewards that play to our instinctive desire for the esteem of our peers. Some games might make the model of the real world immersive, so that we as players can ignore distractions and concentrate; others might be similar to the digital tree that grows inside the dashboard of the Ford Fusion Hybrid, and will provide subtle yet constant feedback for our behavior.

We live in a world in which ‘all models are wrong but some models are useful,’ and as that world becomes increasingly interconnected and complex, the games will provide us with the data collection tools and data processing shortcuts that enable us to act intelligently. In this future we will design game-like incentives that teach and encourage us to make wise long-term decisions, so that we can outrun that tiger and save this planet. Which is important, because we only get one shot at each.

—

Note: the remainder of this post contains spoilers of the novel Ender’s Game by Orson Scott Card.

—

In the book, Card’s characters play war games that they do not know are actually quite real. The protagonist Ender, who is just a child, excels at the games because he thinks they are games. He uses ruthless tactics to win, unaware that he is actually committing those atrocities in the real world. Ender, unburdened by the extreme pressure resulting from real-world consequences, believes that he is merely playing a game and is thus able to save humanity from an alien threat.

Of course the games of our future need not be so ethically questionable, but the point – that games can simplify the world to enhance our focus and remove our hesitation if we are less sure that they are actually real – is still important.

—

Sup dawg, we heard you like games, so we wrapped a game around your game so you can save the world while you save the world.

Paul Adams, a member of the user experience team at Google and the user research lead for social, recently gave the below presentation at the Voices That Matter: Web Design Conference:

It’s worth reading through the entire thing, but there were a few groups of slides I found particularly clear/insightful/interesting (you can jump to a particular slide from the bottom toolbar) –

• social contexts: 15, 58, 71, 83, 181, 212
• evolution of the web: 19
• status updates: 145, 179
• memory and information: 150, 152
• influence: 158, 159, 171, 172
• privacy: 193, 198, 199, 204

There are lots of other useful ideas in there, but there’s one in particular on which I want to expand. Adams discusses the categorization of our relationships into strong ties and weak ties, saying that, “Strong ties are the people you care about most. Your best friends. Your family. People often refer to strong ties as their “circle of trust.’ [...] Weak ties are people you know, but don’t care much about. Your friends’ friends. Some people you met recently. Typically, we communicate with weak ties infrequently.” Adams then goes on to define a new type of relationship online, the temporary tie, for “people that you have no recognized relationship with, but that you temporarily interact with,” such as strangers in public online social spaces.

He also discusses the cognitive limitations of the human brain that make us unable to stay up-to-date with more than 150 weak ties at a time (see Dunbar’s number). Given that we now have social tools for keeping track of many more people than that – Facebook ‘friendship’ seems to be for “everyone I know and don’t actively dislike”* – I wanted one additional term to help me think about the portion of my 859 Facebook friends with whom I wasn’t keeping up at all and had some sort of tie that was weaker than a weak tie.

Latent ties seems to work nicely here, for those people with whom I’m not at all in touch but also have not forgotten, and who could potentially become a bigger part of my life and replace one of my weak ties. This is a new type of tie – it used to be possible to have no way to contact someone I once knew but hadn’t heard from in years, and these new tools will prevent this from ever again being the case. I think it’s especially important to design for these latent relationships on Facebook/other websites where there are social stigmas around friending and unfriending that make it difficult for the user to keep her ‘friends list’ as an accurate representation of only her current strong and/or weak ties.

* Who was it that first said this? Please let me know if you have a source for that quote.

Several weeks ago I presented my project Where Do You Go at the NY Quantified Self Meetup Group‘s seventh Show & Tell at ITP. Evan Creem recorded and edited together videos of the presentations, and mine is below:

If you’re interested in self-quantification in general, The NYT Magazine recently ran a good article titled The Data-Driven Life. About a month ago I received my Fitbit, one of the devices mentioned in the article, and you can see the public data I have collected so far here. I’ve been using it primarily to get a sense of how much I actually walk and how little I actually sleep – two things about which it’s slightly too tedious to make daily notes but which still might be interesting to examine in the aggregate.

A couple weeks ago I presented Wanderli.st during Thesis Week at NYU’s Interactive Telecommunications Program. The list of all of the presentations is here, and ITP has a copy of the video hosted here, but I’ve also embedded it below. Some of the slides are difficult to read in the video, so they are embedded as well. If you’re in a hurry and think you can read faster than I was talking (hah!), the notes on which the talk was based are below each slide.

I’ve been thinking a lot the best way to continue this work now that I am free from academia, and about how Wanderli.st fits with other proposals such as Diaspora (which has gotten incredible support). I’ll continue to publish updates here, and let me know if you have any ideas you’d like to discuss!

Note: I had originally wanted to synchronize the PDF of the slides with the video, but I couldn’t find a good tool to help me with this – Omnisio has disabled the ability to create new presentations since they were acquired by Google, and the Zentation player was simply too ugly (despite their much prettier main website). Let me know if you can recommend something!