But this post is about history of civilizations on the internet. While Google Plus is launched, MySpace was sold in last few days for a puny $30M and Friendster for $100M. Not to mention the once acclaimed AOL that was spun out of Time Warner in last few years. A $100B+ write down?
So why do social networks keep dying?
I think the answer lies in history of cities, and entropy of private information. Allow me to explain:
1. History of Cities: Social networks are like cities. They are born when small villages (Facebook at Harvard) expand due to a whole host of factors ranging from better cleaner layout, availability of resources (upload your pictures), ability to mingle with new kinds of people (dating & business relationships), etc. While all this is somewhat obvious – these social network cities need to be viral to expand. As more people join, we all derive more value — hence, most of them are designed to be viral. However, this virality means over time our social network cities become crowded, cluttered and polluted with noise that makes them less valuable. Many of us intuitively feel this – how many of us have had that moment where you go from – I love my new social network city to I don’t how all these people ended up being my friends? This extends beyond social network cities to many other communication tools – Yahoo! and AOL IM – small clean lists of people I wanted to talk to to a list of everyone who wants to bother us.
This happens to great cities too – London, New York, Calcutta, Mumbai – all have gone through there a growth spurts followed by the years/decades of being terrible places to live in. But the cost of building a new New York often outweighs the cost of cleaning it up and fixing it up. Where we can, we cheat and build a New Delhi (yes, New Delhi is literally a new Delhi built outside of Delhi; ditto for Jerusalem, and many other cities where you could geographically move out). Cities like New York, London, San Francisco end up spending money to fix the cities enacting laws to keep it livable and make it better over time.
Alas, with social network cities of MySpace and Friendster – no such luck. Its easier for the residents to move out gradually and live in the new suburbs of Facebook and Google Plus. And if your friends move with you, the move can be sudden and quite painless. You can live in 2 cities at once – and over time stop visiting your old city. This holds true for your old email and old IM too. Remember, Hotmail and Yahoo! Mail, or AIM – you don’t call her any more, do you?
2. Entropy of Privacy
Over time, private information that is shared becomes less private. Think about your Social Security Number, when you first got your SSN – it was a rather private number. If you have been using it for a while though, its probably in the databases of dozens of banks, mortgage issuers, insurance companies, employers, payroll managers, your dentist’s local office, and so on. This means its much more likely to have been disclosed to more people and therefore less private. Same is true for a new email address you acquire. For a while, very few people know your email address but as you start using your email address for its intended purpose i.e., email people – over time, email address gets shared by more and more entities and eventually is effectively public.This concept has been studied by academics with more precise definitions (see here).
The analog of this is your social network and how well managed it is. Disorder increases in your social network over time – you add more and more people, trust goes down, unless you manage your accounts as it were a full time job your social and business friends and acquaintances all get mixed – and sometimes you add strangers just to avoid offending someone.
So what do we do? We don’t have to live in the mess we co-create! Rather than clean house, we move. Of course, over time – we re-create the same mess. But for the first few months, may be even years you will get a lot of value from having your clean new Google Plus Circles which is a smart innovation but hardly path breaking. Facebook could easily help us manage our friend lists by creating 5 prepopulated lists with privacy settings and auto suggesting who goes where – they have plenty of data on my social graph to help partition it which is a nerdy way of saying they can help me create my ‘circles’ using information such as who do I chat with, share pictures with etc.
Fundamentally, Google Plus so far is just Facebook (Circles) & Twitter (Stream) rebuilt from scratch with certain solutions that are cleaner (by definition) and better integrated with rest of my Google life especially Gmail.
There are however new kinds of social networks that are fundamentally different. Yatown.com is focused on the opposite of a global social network – they are creating a local social network, a way for you to talk to, collaborate with and discuss with your neighbors. While Facebook and Google Plus help me connect with my family in Australia and India – Yatown does it for my neighborhood in San Francisco. This is different and interesting.
Similarly, Quora is focused less on social networking and more on being a repository of the best answers on questions ranging from what cars to buy to how did company X get started. Its singularly focused on solving this problem with social being a side effect. Facebook Answers is somewhat similar but lacks the focus of Quora.
Summary
The mega trends towards people spending more time on social networks (Facebook, Facebook, Facebook) is a fundamental threat to Google. And Google Plus is a worthy response except it solves a problem that Facebook has already solved. It will be an interesting few months to see how this works out. In the meantime, I am much more keenly watching what’s not been solved yet ranging from local social networks, Q&A networks to enterprise collaboaration.
In the enterprise world, Salesforce.com’s Chatter is bringing private social networks to enterprises enabling Facebook like collaboration in a secure manner within the enterprise.
Exciting times to live and collaborate in.
(Disclosures: I am affiliated with Yatown and my opinions are therefore likely biased. My affiliation to Salesforce.com is also disclosed on my blog.)
![]()
Related articles
- Google+: A targeted response to Facebook’s shotgun approach (gigaom.com)
- The Evolution of Social Networks (cc-chapman.com)
- Google’s social network is the online equivalent of arranging wedding seating charts. (slate.com)
- MySpace sells for $35M, what went wrong? (lostremote.com)
(Cross-posted @ AnshuBlog)

Anshu -
I liked your post. It reminded me of one I did a while back:
http://dbmoore.blogspot.com/2009/07/moores-corrolary-to-metcalfes-law.html
The basic premise of this idea is that Metcalfe’s law described the benefits of being in a network, but did not describe the costs. My proposition was to describe the costs. As you pointed out, there are costs to being connected to certain folks, and there is a cost to being part of the network in general. The costs are related to bad behaviors costing your psyche, redundant messages taking your time, crime and abuse taking your time and perhaps other more valuable assets (money or identity), loss of privacy creating risk, etc. There are ways to contain the costs – limit membership, daily digests, rules, punishments for outrageous behaviors, moderators, etc. – things we’ve done in various social networks (or cities) to one degree or another.
Metcalfe’s Law describes the value of a network increasing with the square of the number of participants, because each participant brings her value to each of the other participants, but Metcalfe’s Law clearly is not complete. As you’ve described above, each participant can also bring costs to the network. If Metcalfe’s Law were complete, then all humans would live in one city, and there would be no need for a government.
I think I will revise my Moore’s Corollary post and add it to this interesting series. Now, if only I could drop my Google Buzz membership and trade it in for an invitation to Google+
- Dennis Moore
Dennis,
I concur with you. I enjoyed your blog post – yes, there is a certain maximum number after which the size of your ‘group’ becomes detrimental. And I think this idea has been brought up in other domains to find the right ‘team’ or ‘group’ size to maximize collaboration and productivity. Dunbar’s Number captures this and there is a whole lot of research around it. Here is a summary of Dunbar’s Number at http://en.wikipedia.org/wiki/Dunbar's_number
Anshu
Dunbar’s number doesn’t cover how to quantify the cost of adding new participants, nor how to minimize that cost to result in the best possible cost-benefit ratio for the group. In any case, great post!
- Dennis Moore
Yes, Dennis – The Dunbar’s number is a somewhat generic, broad concept while you are post does a great job of assessing the impact of adding members to your social network/circle.
Which do you prefer? Facebook “Like” or Google “Plus”? Just make you choice here, http://likevsplus.thewebsys.com/. Enjoy it
If public socnets like Facebook and Google+ are analogous to public cafes and bars (where camaraderie and “tummeling” are the benefits), there is a need for a “socnet” that is analogous to living room social gathering (where “strong ties” and close sharing are the benefits). Just in physical life, these private socnets should not place restriction on potential friends.
I also feel that there is no hard limit on “Dunbar number”. That number is a function of time and occasion. For example, an Indian may have 10s of friends, but for a family wedding, the number will increase to hundreds. Similarly, private socnets must allow for this kind of variation.
These and other ideas are captured in an implementation, EnThinnai that can be run by individuals in a device (like wifi router, Cable/DSL modems or residential gateways) running in their homes.