I became interested in online social networking systems because I love traveling, but hate staying in hotels. To me, they seem boring, wasteful, sterile, and isolating. Thankfully, although I travel a lot, I am only very rarely forced to stay in them. Instead, I use a hospitality exchange service called CouchSurfing.com (CS) to find accommodation with local residents in the areas I visit. This may sound a bit strange at first, but let me briefly explain how this works.

I registered an account with CS in 2004 and submitted a bunch of personal information including my age, sex, occupation, my Skype and chat IDs, links to my websites and photo galleries, etc. It isn't mandatory to provide so much info, but when thinking about opening your home to a stranger, it all seems very pertinent to the decision. If you really wanted to do a thorough background check on potential hosts/guests, think about all the information (that isn't covered above) which might prove useful. Still, many CS users' profiles even lack a facial photo. This represents a particularly egregious omission in the world of online social networking, particularly within a system like CS that is entirely predicated on face-to-face meetings in the “real world”. However, around 200,000 people worldwide have well developed profiles and access the system on a regular basis. So, whenever I'm planning a trip somewhere, I search the database for interesting and trusted people in that specific locality and ask if they might be able to host me at their place during a certain period of time. Though we are formally strangers to one another, we have an initial trust base from which to build on because reliable third parties within the system have attested to our status as good guests/hosts. Since this trust mechanism relies on accumulated reciprocity and reputation, I do my best to be a good guest wherever I go and an excellent host to travelers staying at my home.

CouchSurfing has so profoundly changed the way I travel (and live) that I felt compelled to explore further online networking technologies and their social impacts. By now, most of you have heard enough about MySpace and Facebook... you probably even remember Friendster. Exciting as they may be, I won't rehash their respective controversies and business fortunes. Instead, I want to emphasize that they should be viewed as much more than an anomaly of US youth culture. They represent a fast growing global phenomenon with extremely important social and economic effects. You may not have heard of the dominant online social networking systems in non-Western countries, like Mixi in Japan or CyWorld in South Korea. The latter has the largest national coverage of any system (over 90% of 18-35 year olds have accounts) and has recently attempted forays into the American market. Hundreds of online social networking communities with distinct user populations and sets of functionalities now exist, but these are overlapping more and more each day. While there will always be some holdouts, these networks will reach saturation among young people in the developed world sometime in the very near future.

However, there are looming problems with balkanization along linguistic, national, and interest lines. Most of my American friends are members of either Facebook, Myspace, or both. Most of my Indian friends are members of Orkut, a social networking service owned by Google that now draws most of its membership from India and Brazil. Other people I know primarily use networks that are oriented towards particular lifestyles and “real world” functionalities, like LinkedIn for business networking or LastFM for audiophiles. Like them, I choose to really only use CS because it fits my lifestyle best and offers me pragmatic value like savings on accommodation and meeting really interesting fellow travelers. These are services that the more generic “friend management” systems like MySpace cannot fulfill, but if I was more stationary and intent on keeping up with the hundreds of friends I have accumulated over my lifetime, they might very well better suit my needs. This doesn’t mean I don’t keep shallow profiles on these other services so that I can try them out and allow other people to find me through them. However, they don’t have much allure to me mostly because I don’t much “believe” in their vision. As I got to know the people and processes behind CS better by volunteering and researching their organization, I began to see them as a admirable social entrepreneurship venture. They are different from the others not just in bridging the real/virtual divide, but because of their organizational form. It certainly isn’t perfect, but they are trying to be member-driven, non-profit community that sees itself as providing this service as a way to improve the world.

I just cannot bring myself to embrace a bunch of different networks with their attendant bureaucracies and business models. Moreover, if I give in to the desire to have a digital presence everywhere, I subject myself to the process of registering an account, building a profile, and maintaining contacts in each new network I join. This would involve a lot of time, redundant effort, and crossover costs associated with learning different sites' designs. I’m sure many other “network loyalists” like me sometimes wish it were possible to simply export my data from their main networks to all the other networks, but since they all use such different data structures, it would take colossal redesign agreements to make more than any two compatible with each other. Contrary to popular belief, most application programming interfaces (APIs), like the one Facebook recently released won’t be designed to serve this purpose. This also means that new the aggregator applications like Ziki or SocialURL will never amount to much more than link dumps for an individuals' profiles on different networks. There is a large gap between this activity and an actual aggregation of digital identity, which I think will be closer to OpenID - a service that provides a more powerful means to manage one's identity online.

Commentators have proposed that 2007 may be a weed out year for the ever expanding ecosystem of social network sites. Although consolidation of the online social networking landscape may indeed be in the cards, I think traditional market mechanisms of buyouts or die offs will play a very small role. One network will not be so successful that it eclipses all its competitors. Instead of a winner-take-all outcome, I predict the emergence of an entirely different framework. There is a paradigm shift looming, one I often refer to as “open source social networking”, or a “SocioNexus.” It will not look or behave anything like existing social networking sites. In fact, I don't think it will really be a “site” with discrete functionalities at all, but a set of distributed mechanisms individuals can use to create, store, and control access to their personal data. In a few years, most people won't be logging into bounded social networking sites with different passwords and limited personas. They will still be using a multitude of social networking applications like today, but the data used to drive them will be pulled from their own central resource of personal information that they are responsible for managing and control access to instead of from a bunch of services’ databases. Why?

It partly has to do with the nature of the technology that they will utilize. Bigger social networks can deliver more complex information services than smaller ones, but growing to truly massive sizes requires a high level of data standardization. Ever since the inception of the Internet, there has been a push to standardize its protocols. While this may seem at odds with the observed tendency towards innovation and diversification of online content, greater consistency of information remains a core goal that continues to exert ideological and institutional pressure. The World Wide Web Consortium (W3C), the leading advocate of Internet standards (and the folks that brought us the long-standing common html format for web pages), develops and promotes consistent protocols for presenting information on the Internet. This is key to the emergence of the next generation of social networking applications because it will allow complex data to be interpreted by computers without human help. The ability of machines to “read”, or make sense of human directives and reference information without the aid of abstract programming languages is referred to as “Web 3.0” or “the semantic web.” The family of technologies that can make this happen (RDF, OWL, SPARQL) are now in the final stages of development and approval by the W3C. However, even without invoking these advanced technologies, it should be simple enough to state that different services cannot understand each other’s data unless they standardize the vocabulary they use use to describe it. On the Internet, a rose by any other name means next to nothing.

Still, it partly also has to do with the nature of the information involved. Businesses generally have an interest in breeding exclusive and non-congruent information frameworks, which essentially is a deal breaker for any kind of open source social networking project? Moreover, they would have to brave huge liability risks in the event of leakage or theft of personal data. Because the intensive personal data stored within it will likely be very sensitive, it is highly unlikely that users will trust it to any larger entity without some kind of tight oversight. The level of opposition to large corporate holding of sensitive information varies from place to place, but has been shown to be especially strong in Europe, where the German government has been quite hard on Google’s search history features during past months. Governments themselves are certainly not immune from these problems either, though they have been faster to undertake more ambitious efforts in the name of international justice and health issues. Still, these collection efforts still worry most people when they hear of them, both because they don't like the idea of the state knowing so much about them and because they think (with good justification considering recent stories of massive data leak) that they cannot fully protect their data from falling into the wrong hands. Thus, not only will the framework and the data likely be non-proprietary, but non-governmental as well.

Without a doubt though, businesses and governments will still be especially interested in using it. However, it is my hope that they will have to buy the right to access it. The idea of individuals controlling their personal information is very different from how companies that provide online services today own, control, and sell their users' demographic and behavioral tracking data at will and without remuneration. No one wants their personal information forcibly or covertly collected, but most of us actually assent to this whenever we register on a new site. I believe there is a viable alternative to a future in which we are unknowingly tracked and manipulated whenever we surf the web or pass within range of a radio frequency tag. Instead of our attention, attitudes, and behaviors being tracked by some third party and sold to interested businesses, we can do most of that ourselves; to the extent that we wish to participate in such activities. Currently, the vast majority of overt data gathering efforts, from phone surveys to customer feedback forms to national censuses, yield no real compensation to participants. Moreover, they’re incredibly boring. I've written and administered enough surveys in my time to realize that people generally hate these processes. It doesn’t have to be this way - online data gathering efforts could be a lot more entertaining and useful if we put more thought into the methods and presentation. So much is now possible just by using Flash technologies, from simple little games to complex multimedia adventures.

Still, some argue that no one ever really wants to take a survey, no matter the format. I ask them to take a look at some MySpace pages and marvel at young people's enthusiasm for publicly displaying their survey results via the kitschy backgrounds designers provide them. I argue that if surveys (or anything else) are made into a game, people (especially young people) will play them. This isn’t the same as the “educational” software that many critics rail against ineffectual, because they are inherently social. It seems sometimes surveys, like interviews, help us learn more about ourselves. People desire interesting presentation mediums to help share these insights with others. Marketers fielding such surveys have an interest in allowing them to accumulate this data (especially in addition to their personal information) because it will help them to deliver better targeted advertisements and promotions in the future. This benefits users too because when ads are developed for and delivered to the particular groups of people most likely to purchase a product, they are usually far less annoying than those that are delivered indiscriminately. Moreover, users should receive monetary compensation for their attention to these online ads because they assume much more potential value with these additional demographic and behavioral data attached. Reshaping the advertising landscape in this way stands to really empower the individuals and reign in the increasingly ubiquitous power of those who deliver media to assail our attention everywhere in our world.

These data gathering and aggregation improvements will undoubtedly yield huge amounts of data on individuals around the world. I cannot wait for the day when I can search the majority of humanity for a certain kind of person, skill, experience, good, service, or a combination of them. Some folks within CS like to talk about a future in which people no longer just look for places to stay when traveling, but learning experiences and personal growth. This may seem murky, but consider it in the context of resource sharing in general. Using someone's spare bedroom instead of getting a hotel room reduces the demand for hotel rooms in a city. Being able to borrow someone's bicycle in the city makes it less likely that you will need to rent a car in order to get around. Learning some yoga from an amateur enthusiast instead of a certified instructor may seem less professional, but it is certainly cheaper and perhaps less intimidating. Remember, this will all happen under the umbrella of well developed trust mechanisms within a social networking system like CS. Better information about the world around us allows for the discovery of possibilities that before were not even perceptible. It also allows us to make more informed and efficient decisions in our day to day existence. Think about how much time young Americans spend driving around in cars fairly aimlessly just looking for things to do. It will allow people to navigate the world in new ways and alter the way they are able to envision and shape their lives. When we combine data about individuals with the resources we already have concerning knowledge and commerce, the possibilities explode.

For social scientists like me, the scope and depth of its data will be unprecedented. It will quickly outstrip that of national censuses. This is because our constant utilization of it and contribution to it will yield a constant stream of data (lifestreaming) that is incredibly longitudinal and lends itself to all kinds of causal analyses researchers can only dream of now. It will become the most important source of data on individuals and social trends. Behaviors made possible by this new technology will entirely alter the nature of the debate on privacy and spawn changes in the legal system. However, the effects will certainly be bidirectional - network processes will also be guided by new laws and social norms created through these changes. I hope the framework under which necessary changes are made will be shaped by the open source emphasis on remaining non-profit, independent, self-maintaining, and self-evolving. Still, the path from here to there is still quite unclear. Despite the huge amounts of press coverage over the last year dedicated to publicizing different aspects of the most popular online social networking systems, there is still a lack of meaningful discussion about the future direction of this medium. Most of what does exist is found on the blogs of people who study this phenomena or run businesses involving them, which are relatively obscure and difficult to find if you don't know where to start looking. The point I'm trying to make herein is these developments are imminent and that we should start focusing more attention on the possible implications, both good and bad. There will certainly be downsides and dangers to this technology as well. We should not allow the technical aspects of these issues to overshadow its potential social consequences.

Now, what of the fate of the networks to which we now belong? Does this mean that all the different social networking sites existing today will simply fold at the arrival of the SocioNexus? Will the manifestation of a more consistent information architecture for social networking be all it takes to supersede earlier tribal models? Not necessarily. Some of the services they offer are complex, useful, and may not be easily duplicated under this framework, at least at first. We should expect users of the old networks will cling to them so long as their needs are still best served there. However, it does likely mean that they will increasingly cease to be “online communities” in the sense that users see them as a bounded and protected social space shared only among “members”. This seems sad, but it won't destroy most users’ worlds, since they were never really uniquely digital social spaces. Individuals mostly just modeled their “real world” social networks digitally. People with huge social networks online are often looking to project their popularity or connectivity, but really only maintain a small fraction of those links on a regular basis. Even in a “real world” service like CS where users have relatively fewer total friends in the community, most represent brief ties established with geographically distant hosts and guests that are rarely well maintained, digitally or otherwise.

The SocioNexus will be neither a service nor a community per se, but a medium for the development of both. It will allow people to use a multitude of ever-changing online services more easily and effectively by structuring their personal data in a much more consistent fashion. It will allow them to build communities not bounded by these services and their attendant corporate brand loyalty strategies, most often aimed at keeping them sequestered in specific domains for as long as possible. The critique commonly leveled against this idea is that internet is too overwhelmingly expansive, that we need these services to subdivide social spaces and make navigation simpler. I argue that the new nature of data online makes it possible for a very big network look, feel, and behave like a smaller network. However, scalability only works in this direction – small networks simply cannot offer more with less. Information and the ability to search it generally becomes far more robust (accurate and efficient) the larger it becomes, but only it utilizes standardized and accessible querying protocols. If a centralized data storage mechanism for humanity did exist, how would individuals manage and protect their stored data - determining who can access what and when. Would everyone need their own unique domain or subdomain name with common extensions like .bio denoting space exclusively reserved for information about real, live human individuals? Could all their data just be behind the scenes on traditional web pages (using html) using microformats and the W3C technologies I referenced earlier? Under such a framework, what kind of expertise would an individual need to construct and protect their data? Could federated (distributed) searches of decentralized web pages ever hope to match the efficiency of the more centralized indexes of a big social networking site? I do not know. I am not a technologist, thus I cannot claim any credentials to make such predictions. However, I am quite confident that this is the direction we are moving.

Still, within all these technically murky areas, the very critical area of verifying identity and establishing trust mechanisms is now perhaps one of the clearest. I think it can best help us see the possibilities for this kind of technology in the very near future. Recent advances in Internet identity technology have given us the OpenID protocol and its sister project, ClaimID. They function together to allow individuals to assert their unique identity online and make ownership claims on particular web content to establish credibility. OpenID doesn't just allow us to use the same login and password protocol for different sites. It also provides a decentralized means for individuals to be able to verify each other's identities. That means they won't need to be reliant on larger services to do this for them. Also, it may allow individuals to share their data across websites, if both accept the OpenID protocol, which a fast growing number of sites do. So, if I want to import my friends' chat IDs from CouchSurfing to another social networking service, I may finally be able to execute an automated mechanism to do this task. So, just as we can share data between websites, we should be able to share it with other individuals as well. The depth of the data we could potentially share will allow trust to be fostered outside the big social networking systems. Still, if on Facebook I know that someone is a “friend” of someone else we know, that is not really a good enough reason to trust that third party very far in the real world. However, if we know more detailed information about the nature and duration of their relationship, trust can be more easily cemented. This is easier to accomplish with decentralized mechanisms that allow us to look at personal data across networks. Anyhow, I encourage you to look into these services today, because their marketing budgets aren’t big enough to receive mention in too many places outside hard core tech enthusiast sites.

To summarize, the basic idea is for people to be able to submit their data online in many ways, have it stored in consistent and secure formats that that can be accessed and compared with a colossal amount of other people's data, yet allow many different display and application options. One will “own” their identity by determining what data they submit and who can access which parts of it. So, one might want to list their email address, but only allow other certain kinds of individuals within their trust circle to see it. Moreover, potential employers visiting your site would get access to very different pieces of information under very different presentation formats than would potential dating partners. Again, if it indeed does become the most important central repository for personal information on the web, it will also become the foundation for the next generation of integrated social networking tools. Programs will be written to parse all this human data and display its patterns in meaningful ways that help us better understand social, economic, and political patterns relevant to our lives. These calculations will take into account a huge number of human data factors found within particular population parameters. They will help weigh us weigh different choices and make intelligent recommendations based on what we have learned about our own behavior as well as that of others. This will vastly enhance our ability to navigate the world when we shop, date, travel, work, volunteer, communicate, and generally share our resources and knowledge about the world. The richness of the information when uniformity, intense data collection, and massive population coverage coalesce will yield possibilities that we cannot even currently imagine.

The Educated Guess is a game in which players try to ascertain the biographical details of a person unknown to them. It's basic appeal involves challenging the almost universal human assumption that we can tell a whole lot about a person just with the help of a few key visual, verbal, and social cues. We test this by providing players with a few initial clues about an individual (a photo, video, perhaps even their name) and then asking players to make educated guesses about other aspects of that person based on those revelations.

For instance: a contestant sees a photo of a long haired, athletic, teenage boy wearing a glasses and a baseball cap sideways. If the contestant chooses the question category of "Hobbies" he is presented with multiple choices including A) Skateboarding, B) Chess, C) Baseball, D) Guitar. Long hair (rock n' roll?) might seem to play against the baseball cap (if it indeed features an MLB logo) and for skateboarding, but the glasses may in some way shout "chess club!" as well. Perhaps more than one of the choices is valid. The contestant must use the other non-visual information he has managed to reveal throughout the game to aid his choice. This might include the subject's parents' backgrounds, his place of residence, or the consumer products he prefers. As trial and error discoveries made throughout the game reveal further dimensions of the subject's biography, more accurate predictions becomes possible. The educated guess becomes more educated with each piece of knowledge gained.

However, this is all mediated by the contestant's own cultural background, areas of expertise, and experiences in society. The game reveals how the background of the players influences the assumptions they make and the paths of logic they follow. Imagine the playing field as a large asymmetrical map shaped by different facets of the subject's identity (education, family, occupation, etc) and a design template they have chosen to present this info through. Since each one of these important areas of life usually influences others in a systematic fashion, players who are able to quickly assemble a few key clues in each may be able to really open up the board in fits and starts. In this way, the game may look a bit like minesweeper or sudoku but with a much more multimedia feel. The winner of the game will have to utilize a little competitive psychology and a lot of sociological background knowledge in order to accumulate the most correct information the identity puzzle in the least amount of time.

Perhaps the most intriguing thing about the framework game is how in depth and variable the subjects and clues about them can be. The possibilities are so expansive that we are designing a parallel version of the game online based in a social networking model. The ability to play the game against real people, (potentially even acquaintances) will familiarize audiences with the format and help generate interest in the television show before it debuts. In order to earn the "credits" necessary to play the game, users must answer a few questions about themselves, thus enriching their profile and making themselves much more interesting subjects. This will also provide a valuable testbed for experimenting with many variants of the game with different rules, scoring methods, and player configurations. The latter could include individual play with arcade-like score rankings, one-on-one real time competition, or multi-player alternating formats.

A potential twist could involve using a subject known to the player(s) as friends or acquaintances as a "how well do you know me?" or "get to know you better" mechanism. Such possibilities might provide very interesting synergy with a live television version and supply a very effective viral marketing avenue. When appropriate, a host may take time during play to question contestants' reasoning and further explore (for the benefit of the audience) how stereotypes, biography, and media exposure guided their decision making. Moreover, the online game will be exceptionally scalable and affordable because so much of the content will be user generated and play itself can be automated by some fairly simple algorithms. Finally, it has the potential to generate significant advertising revenue because the personal information voluntarily submitted by users will allow for highly targeted advertising, for which providers must pay more. It may very well have the potential to spawn a colossal social network to rival today's leading systems both in form, function, and fun.

The background to this idea is that I've spent a lot of time trying to imagine how the next generation of social networking technologies will look and behave. Here is really long essay I'm working on (as part of my dissertation) to address key issues surrounding this topic... if you would care to read it some other time. I start by imagining a future where I could learn a whole heck of a lot of information about an individual just by checking out their "profile" (if they let me) and moved on quickly from there. However, I never bothered to address the first hard question: how to get users to supply extensive information about themselves. Now, as someone who's spent a lot of time writing and administering surveys, I can say they aren't generally well done - meaning they aren't fun and don't reward respondents for their efforts. I write about this problem more in depth in this blog entry.

One could play against other online competitors, in real time or sequential play with score rankings like an arcade game. There already precedent for this idea. Facebook runs a mini version that tests how well you know your friends based on knowledge of their profile information. NBC is currently trying (apparently failing) to run a dumbed down TV version called "Identity". The other is that my bigger goal is to use this as a tool to build a new kind of social networking platform around it. If the game allows for the collection of an unprecedented amount of user data, then I seek to allow users to use/control it in compelling new ways. I call it an "open source" social network - meaning that anyone could develop applications for it through an API. So, the game is a vehicle not just to generate a network with much more extensive data, but may also provide explode the functional possibilities of social networking. In my opinion, this really needed right now, because current applications are simply not meeting the immense potential of the idea.

If I truly believe that people having access to more data can make the world a better place, then I need to focus more on the process of collecting it. I love writing surveys because I get to think hard about how to best extract accurate information from people that can help me better understand them. However, the problem is, as I see it, is that all the survey formats I have come across are tedious at best. Certainly, there are much more indirect way of collecting data that should be mentioned. The best I've seen lately, while not a survey in the traditional sense, are the ESP game and Peekaboom. Getting people to offer information in the context of competition and entertainment certainly seems to be the right way to go. I think this is generally the only reason that people submit to data collection at all outside of institutionally mandated or monetary remuneration frameworks. Whenever people tell me that no one wants to take surveys because they are so boring, I bring up the example of MySpace. I have seen so many profiles littered with dumb surveys help takers claiming to classif personalities, intimate compatibility, or other fun nonsense indicators of humanity. I think the youth of today is actually more likely to willingly take surveys than preceding generations. That the medium for response has moved online has only made them quicker, easier, and more highly customizable.

However, it is still difficult for me to conceive of a game that would help one collect the kind of information that I want from sites like SocioNexus and SocialProblems. The former is perhaps an easier challenge because of the huge diversity of data it seeks to collect. Also, I should point out at this point that getting data on someone's attitudes is much easier than establishing a reliable measure of their behavior patterns. With Social Problems, all I need is a bit of contextual demographic data and their attitudes about the a particular piece of media they have been exposed to. This doesn't take long, but it still isn't really fun for users. I myself love when people ask me rate, discuss, and suggest improvements to movies that I have seen, but I may have a freakish disposition in this area. Moreover, because it strives to be low end and anonymous, users don't get any kinds of points or recognition for their participation. They don't even get to see the patterns in their own data, which people like me would find extremely interesting.

Now, there is something to be said for social networking sites that can compel users to endure yet another tedious sign up process. They don't offer entertainment or financial rewards, but merely the right to use their service. This not only guarantees them success in high completion rates, but accuracy as well. People care about how other will perceive them via their answers, so they spend a lot of time thinking about how they will most likely affect other's perceptions. I guess there is a slippery slope to fall down though. If one is concerned primarily with how their answers make them look as opposed to how accurate they are across different scales of assessment (we'll avoid the epistemoloical dispute herein), then the answers may not be of much objective value to the people who are actually collecting them. However, the worst surveys don't even try to take any of these advantages, despite their potential risks. The data doesn't go anywhere meaningful to the participants, their time and effort aren't monetarily rewarded. Moreover, they are boring and do not speak to issues that they care about.

More time and money should be invested in the creative process of making surveys more fun. I know marketers are already working hard on this, but they just aren't producing results. Does this then a unsolvable problem? Hardly. Integrate some flash components into a survey and it becomes a game, puzzle, or piece of art. I remember way back when seeing some kind of personality survey that gave you results in cool colors or shapes ties to particular response patterns. This is just the tip of what can be achieved if we get serious about integrating multimedia into data collection. Imagine how many more men would participate in a survey that featured a sultry female questioner. Check out a site called ask Ms. Once, this key part of the project is further developed, we can focus more attention on getting the data stored efficiently and in a way that allows users to utilize their power to make the world a more easily navigable place.