At the MIT Enterprise Forum tomorrow in Boston, I’ll be moderating a panel with three social software entrepeneurs on the topic of getting to critical mass. I want to ask the panelists about overcoming the friction involved in joining and learning to use their services.
Years ago at BYTE Magazine my friend Ben Smith, who was a Unix greybeard even then (now he’s a Unix whitebeard), made a memorable comment that’s always stuck with me. We were in the midst of evaluating a batch of LAN email products. “One of these days,” Ben said in, I think, 1991, “everyone’s going to look up from their little islands of LAN email and see this giant mothership hovering overhead called the Internet.”
Increasingly I’ve begun to feel the same way about the various social networks. How many networks can one person join? How many different identities can one person sanely manage? How many different tagging or photo-uploading or friending protocols can one person deal with?
Recently Gary McGraw echoed Ben Smith’s 1991 observation. “People keep asking me to join the LinkedIn network,” he said, “but I’m already part of a network, it’s called the Internet.”
Now of course LinkedIn offers protocols and features that the open Net doesn’t, at least not yet, and the same is true for all the specialized overlays that we call social networks. But there’s a ton of duplication in those layered protocols and features. If we can’t factor out a bunch of the duplication, I think social network fatigue becomes the major hurdle standing in the way of reaching critical mass.
I’m sure everyone will agree that sign-in protocols should be extracted and made common. What else can and should be refactored? What can’t and shouldn’t?
I look upon LinkedIn as a professional network as opposed to a “social” one. That said, the point is true. One can’t impose networks onto people. The same actually holds true for IM clients, etc. Hopefully this will be the next step in the evolution of social networks
The internet should become the Semantic Web. Imagine that you have your own profile and lots of data and resources (also other profiles) that can be connected to your profile. These connections are really flexible because built on RDF (http://www.w3.org/RDF/). This semantic network can be rendered via onlologies (e.g. in OWL (http://www.w3.org/TR/owl-ref/)) to select only valueble links. There is a great advantage of ontologies that you can have different ontologies to render your semantic network in different ways to extract different values.
There is a number of projects that pushing this idea:
Protege (http://protege.stanford.edu/)
SemanticLinks (http://semanticlinks.blogspot.com/)
TopBraid (http://www.topbraidcomposer.com/)
How about a modular and extensible profile?
Like, one module for reputation, another for likes/dislikes, and yet another keeping track of your purchases online? You decide what’s public and private. The social network should be distributed. In the real world, we don’t have to meetup at a certain physical place to see our friends or make new ones, so why force this limitation online?
This module should allow us to search all publically available profiles and make connections as appropriate.
Imagine the power of a system that allowed me to send a question out to the network, and that question could be routed to a set of people that were most likely to provide the *best* answer based on the system’s knowledge of their interests/abilities.
Now, **that** would be amazing.
Jon –
I’ll give you an instructive counterexample as to why I think that social networks, or more properly networked applications where groups are first class objects in the system, are here to stay and to grow.
I belong to a little web-based community of people who use pedometers to keep track of their walking practice called Walker Tracker. It lets you track steps, compare with your friends, and manage a little step blog of what your day’s walking is like.
As it happens, this gives me an opportunity to make a whole new class of contacts who would be completely inappropriate to cultivate and keep track of by any other methods. The solo efforts of writing down steps every night are replaced with a social effort of trying to keep up with a friend in London who walks everywhere or reading along on some friends who have left Ann Arbor but who stay in touch.
With some growth, this site is already “too big”, in that I don’t know everyone and don’t show up on the high score list like I did last summer. So the next logical step is to create some kind of group infrastructure, so that people can self-identify in groups of 2 to 12 to 150 and compare progress and share notes within those groups.
LinkedIn’s biggest weakness is that its group infrastructure is really awful. Administering a group is harder than it should be, and creating a new one is a manual and trouble-prone process.
“So the next logical step is to create some kind of group infrastructure, so that people can self-identify in groups of 2 to 12 to 150 and compare progress and share notes within those groups.”
Here’s my claim, though. Those groups are not naturally affiliated to the Walker Tracker service. They are instead affiliated by things like geography, type of activity (exercise walking, nature hikes), technology preference (mapping systems, GPS toys), and so on. These affiliations amount to what the aspect-oriented programming culture calls “cross-cutting concerns.”
If the group-formation capability is siloed within Walker Tracker, or a mapping service, or a service for nature enthusiasts, or for GPS owners, then you can’t mix and match, you can only choose a silo and then try to leverage whatever group formation capability is available inside it.
“I think that social networks, or more properly networked applications where groups are first class objects in the system”
Agreed. The question being: to what system do these groups-as-first-class-objects belong? My answer: the Internet, not group-formation silos within it.
Recently Gary McGraw echoed Ben Smith’s 1991 observation. “People keep asking me to join the LinkedIn network,” he said, “but I’m already part of a network, it’s called the Internet.”
Russel Beattie made that observation too – http://www.russellbeattie.com/notebook/1008411.html.
“I think in general, people who would want to use this service are pretty contactable without using this system, no? At least to me they are. I mean, I had my email and web site in the bottom of my profile as did many others.”
When he said that – a couple of years ago – I kind of agreed, but since then my LinkedIn network’s grown by virtue of me doing nothing more than saying yes to someone every month or so – and the practical upshot has been real opportunities that would otherwise not have come about. For all the theoretical reasons why the open internet makes LinkedIn redundant, the reality is that there are many LinkedIn members around who wouldn’t remotely consider keeping a homepage, blog, or myspace page.
“the reality is that there are many LinkedIn members around who wouldn’t remotely consider keeping a homepage, blog, or myspace page.”
That’s true. I have a few friends who are invisible to the public web but who are extensively and powerfully connected on LinkedIn. It would have been difficult, probably impossible, to jumpstart that kind of dynamic on the open Net, in a federated way.
Will that class of service evolve into one that runs on the open Net in a federated way, though?
Flashback to 2004, Jason Kottke going “-ster”-crazy … and my 2005 observation that decides to look at social networks from the point of view of the owner/operator of the network, instead of the user.
What happened to that FOAF stuff, anyway?
Jon –
To use my Walker Tracker example again, just as a reference point:
Some of the group forming on the system is clearly from people who already know each other from some other world. There’s a little pedometer-tracking cluster at the U of Michigan School of Information, and those folks seem to connect on most any system you’d care to look at (Facebook, LinkedIn, UM’s InCircle, Vox, etc) just because they already know each other.
I don’t think that group forming on “the internet” is enough. These sites offer some affordances specific to the task at hand, and aren’t just a static declaration of affinity. Consider the symbiotic relationship between LinkedIn and recruiters, for instance; if you’re looking for a job, there’s enough of a pool of people who make their living from connecting applicants to positions to ease your burden. Walker Tracker is not just about being friends, it’s about counting steps and doing contests and comparing notes on your gadgets. The group-forming software also brings in tools specific to the needs of that class of group, whether it be chess clubs or World of Warcraft guilds.
The social aspect of it is more than a little important. Every group-forming network has its alpha connector, the person who starts it who brings in the lead users. Think of Reid Hoffman at LinkedIn, or Ross Mayfield at Socialtext, or Ben at Walker Tracker, or Craig at Craig’s List. People join, not just because the software helps them keep track of their walking, but because they are attracted to the sort of people who are attracted to the sort of people who find the place. From long experience on online system it’s not the technology that makes the difference (think of all the groups that formed on hideous systems like Compuserve). It’s the ability for people to gather. No reason for that to be homogenized network-wide.
“There’s a little pedometer-tracking cluster at the U of Michigan School of Information, and those folks seem to connect on most any system you’d care to look at (Facebook, LinkedIn, UM’s InCircle, Vox, etc) just because they already know each other.”
Exactly. And because they probably choose the same screen names everywhere it’s possible for them to do that. But it’s still necessary to reconnect in each of those contexts, using different mechanisms.
“It’s the ability for people to gather. No reason for that to be homogenized network-wide.”
I’m not envisioning homogenization at all. Group formation is an engine of diversity and, as you say, the impulse will overcome many obstacles.
But why couldn’t group affiliation be a portable construct so that the group could use a variety of services without having to reform itself in terms of each service?
Although I agree with the overall premise, the problem is more complex. Social networks are an insulation from the larger mass of people, and especially of people you do not want to be connected to in any way. When you become connected to others in LinkedIn, MySpace, or any other social network, you have exposure to far more information about each other. Further, these systems usually expose your other connections to those people. So, context matters a lot. If you belong to a knitting network (I picked a safe one ;-), you do not really need to have all your LinkedIn business contacts see those knitting people, nor do they want to be seen.
What you would need is one grand login, and “tagged” circles of networks that are related and connected, but with no requirement to cross connect. Of course, once you do that, it is not much different than what exists today other than perhaps some convenience. As happens over and over with the internet, this just says that there is room for an aggregation and personalization site which allows you to link all of your social networks onto one page with one login (such as multi-search sites and news aggregators have).
“What you would need is one grand login, and “tagged” circles of networks that are related and connected, but with no requirement to cross connect. Of course, once you do that, it is not much different than what exists today other than perhaps some convenience. As happens over and over with the internet, this just says that there is room for an aggregation and personalization site which allows you to link all of your social networks onto one page with one login (such as multi-search sites and news aggregators have).”
What about something like Netvibes? I could imagine an online dashboard where I manage all my social networks. A user would log in to their dashboard thereby logging in to all their communities. The user could also make his or her dashboards public if they so choose (another way to connect).
Yes, Netvibes is the kind of thing I am thinking of as well. A couple extensions are likely what Jon had in mind: a ping mechanism to make it easy to see what has changed/updated on any of your networks (vs. manually looking); a group model that allows groups of friends to sign-up together; a find-my-friends method which makes all the connections for you (rather than manually doing so) when you sign-up to a new network.
“So the next logical step is to create some kind of group infrastructure, so that people can self-identify in groups of 2 to 12 to 150 and compare progress and share notes within those groups.”
Are social networks some kind of self-dividing protozoa? I mean, once that guy’s site for pedomometer users will eventually get really big and some users start getting annoyed because it doesn’t cater to retired diabetic pedometer users enough, then somebody will start a social network for that niche. Rinse, wash, repeatn
“Are social networks some kind of self-dividing protozoa?”. The web has always been notable for ever deeper specialization of groups (newsgroups, chat rooms, web sites, web rings, blogs, now social networks). I suspect the next step after tags will be sub-grouping. Why should you see the most popular tags from the great unwashed masses when you can see the tags from people who are more like you (and likely still unwashed ;-)
But, that does not mean you have to have a new site, just sites that allow you congregate better with your own kind. The link/connection/buddy/friend model is supposed to do that, but there is a larger grouping beyond that. Whether the social sites would agree to a portable description model (some XML blob) that allows you and your friends to set up shop in lots of social networks at the click of a button. Not sure, I suppose if advertisers see an opportunity to target you and your friends better, then it will happen.
There is a critical shortage of ionfrmiatve articles like this.
I’m glad I have no friends. ;)
I hate to say, but this is actually what we’re trying to build with OpenID + Microformats like XFN and hcard. Sure, we’re talking about a simple protocol (OpenID) and easy-to-use data formats, but that kind of interoperability and foundation in basic technologies that are available today make a huge difference.
You should take a look at what ClaimID and Videntity are doing… they’re getting close to a “more portable identity” — one where, instead of rhizomatically reincarnating myself every time I join a new social network, the network can literally create a “subscription” to me, that I can turn on and off at my choosing — guaranteeing that sites have the latest info that I want people to know about me *and* that I’m able to choose which networks are able to actively follow me.
Or at least that’s the idea. It’s coming, and sooner than you might think.
“I hate to say, but this is actually what we’re trying to build with OpenID + Microformats like XFN and hcard.”
Why should you hate to say? It’s clear that is where this stuff is headed, and it’s a great thing.
(this response posted here and on Tony Stubblebine’s Stubbleblog)
“Social networks are a big enough improvement in communication that the majority of people aren’t phased at all by the friction of joining new ones.”
In this era of transparency and measurement, we should be able to ask and answer the question: Of the folks who belong to any networks, how many belong to 2, 3, 4, 5, or as in your case, 6 or more?
Common screennames across services should enable the necessary data to be gathered, shouldn’t they?
The answer lies in the below statement.
“Microsoft to Work With the OpenID Community, Collaborating With JanRain, Sxip, and VeriSign”
As it happens, this gives me an opportunity to make a whole new class of contacts who would be completely inappropriate to cultivate and keep track of by any other methods. The solo efforts of writing down steps every night are replaced with a social effort of trying to keep up with a friend in London who walks everywhere or reading along on some friends who have left Ann Arbor but who stay in touch.
when you want to take the social network to the next level, u can try using mobile social networks like peekamo that allow you to communicate by text without giving ur cellphone number out.
Yeh I am agreed with you!
Could VOIS.com become another Facebook?
Since the advent of social networking sites in 1997, the phenomenon has taken the world by storm. Once called a passing fad social networking is now a thriving business, in 2006, alone it garnered over $6.5 billion in revenue, while the three biggest players, connected over 280 million subscribers in a way never known before to society. This form of connection has drawn the globe closer together than anyone ever predicted.
Just a few years ago, MySpace.com, solely dominated the social networking site market with almost 80% of the social networking site market but now websites like Facebook entered the social networking site race becoming the 8th most viewed website in the U.S. according to web measuring traffic site Alexa.com. Facebook.com which originally started at Harvard University , later extended to Boston area schools and beyond has mystified many naysayer’s with its explosive growth over the last three years and an astounding asking price of $10-$15 billion dollars for the company. But who will be next?
Who will carry the torch into the future?
With the rapid growth of the likes of MySpace and Facebook the burning question on everyone’s tongue is who is next? As with any burgeoning field many newcomers will and go but only the strong and unique will survive. Already many in the field have stumbled, as indicated by their traffic rankings, including heavily funded Eons.com with its former Monster.com founder at the helm, Hooverspot.com and Boomj.com with its ridiculous Web 3.0 slogan. There are many possibilities but it is a dark horse coming fast into view and taking hold in the social networking site market at the global level that has us interested the website – Vois.com. Less than a year ago, this newest contender directed at 25 to 50 years olds graced the absolute bottom of the list with its website ranked at a dismal 5,000,000. With not so much as a squeak this rising star has come from the depths of anonymity growing an eye-popping 10,000% in less than one year to make itself known worldwide now sporting a recent web traffic ranking in the 5,000 range.
Understanding the Market
When people in the United States hear about Facebook and other services such as MySpace the widely held belief is that these websites are globally used and are as synonymous as Google or Yahoo in regards to having a global market presence. This idea is completely misguided. Now it is true that both of these social networking giants are geared to service the western industrialized cultures but when it comes to the markets of the future, the emerging markets, they have virtually no presence. The sites themselves are heavily Anglicized, and Facebook in particular has an extremely complicated web interface that eludes even those familiar with the language, making them virtually inaccessible in other parts of the world even where English is the main language.
Our interest in Vois is global and geopolitical. Simply, Vois understands this lack of market service and is building its provision model on a global research concept developed by Goldman Sachs a few years ago. The concept is basically predicated on the belief that beginning now using current economic models and continuing those models over the next few decades will lead to a major paradigm shift in the world regarding nations who are current economic leaders like those being the USA and the other members of the G-7 and those who will become dominant in the world economy mainly the BRICs. In the Goldman research report Goldman highlights the fastest growing nations and has dubbed them with the two acronyms BRIC’s and N-11. BRIC standing for ( Brazil, R ussia, India and China) representing the fastest growing economies and N-11 or what are being called the Next-11 representing the next 11 countries to emerge as future important economies such as Bangladesh, Egypt, Indonesia, Iran, Korea, Mexico, Nigeria, Pakistan, Philippines, Turkey and Vietnam. This approach has already been implemented with some success with companies like Orkut, who has over 80% of the market share in Brazil and large holdings in India and Eastern Europe . Other providers such as Hi5 have the world as their focus and are making great strides in global market share while Facebook builds itself into a niche provider wholly unready to take on the world.
A Growing Presence
As Vois breaks new ground in the world market pursuing previously ignored demographics, they afford themselves the opportunity of tremendous growth unfettered by the giants such as Facebook and MySpace. While cultivating this new user base, Vois will also be able to monopolize on their business revenue strategies, creating an area of commerce that will make their site increasingly attractive to business and users the world over. This concept, dubbed sCommerce, allows the subscriber to promote themselves in both personal and a professional fashion while giving them the option of setting up shop on the site. This approach will allow business owners to target their market in a way never before allowing them to focus on interested groups of individuals while providing follow-up without having to commit to wasteful blanket campaigns that are typically the order of the day. This newfound border will allow Vois to explore new revenue models while provide a tremendous service for both their regular subscribers and business subscribers alike. With all this going on, rapid traffic growth to the site, we pose the question – is Vois the next Facebook, it sure looks like it but only time will tell….
Social bookmarking will surely strengthen the traffic needs of any site in whatever language it is.
[…] by Meetup’s Scott Heiferman about the AOL/Facebook parallel. But I also caught echoes of Jon Udell’s post back in February about “social network fatigue”: Recently Gary McGraw echoed Ben Smith’s 1991 observation. “People keep asking me to […]
Social networking is a fun way for teenagers to torture each other and for adults to stay somewhat current in their friends’ lives without having to call them. It’s also an interesting way to discover indie bands. But as a web-of-the-future platform I think it’s largely a fad.
My idea of what critical mass means is when things start to happen exponentially.
It seems like something valid in theory,
let web traffic grow organically and eventually you will reach critical mass(aka a lot of users/visitors).
but there are too many variables on the net
to keep that from happening. there is too much media diversion going on craigslist,myspace,facebook are just a few you can’t read an article or newspaper without a reference to craigslist, other websites that actually have value to pass on don’t get any media attention. hmmmmm internet starting to resemble cable tv?
A great way to counter this diversion is for everybody on the planet to get their own website. start using professional networks to help spread the word about their websites and services. I found a great one: Merchantcircle. you can actually email people you dont know to try and make a connection…what a great idea. come on folks, how many friends do you really have?
done selling your family yet?
ok myspace is not professional so business owners can use bizpup.com to get a business website. There is an internet real estate
war going on. big media is still trying to keep down anything they don’t have an interest in. Wake up America!! see nitpickit.com
you all….thanks for sharing
very interesting thoughts.
though we all have the Internet already, many are dividing it up for their own interests…
http://www.asianme.com/join_inv.php?member_id=239
Years later and I am still pointing people at this blog entry…when they send me yet another linkedin request.