Network effects are not indefensible. I don't understand why so many people in the bay area think that network effects are impossible to break in. Here is an example. Orkut. All my friends including myself were on Orkut in 2004/2005. By 2008 everyone moved to FB. Why? Orkut had horrible UX and buggy system. FB was breath of fresh air compared to that. Even though Orkut has superior privacy features (I still say they had best privacy features for any social app). FB is moving in that direction. It's quite buggy system and if FB must address the overall quality of application. It looks like Orkut in 2007.
The FB social network will die, in time, but it's still the world's largest email address list, plus all sorts of personal information that can be used to insidiously coax consumers.
Every FB user should gather up all the email addys of their friends and keep them in a safe place offline. This way you never lose contact.
FB allowed people to submit their email addresses to a central website and thereby connect/reconnect with friends, colleagues and so forth. This sharing of email addresses is not a new thing, but with FB it occurred on an unprecendented scale. Billions of email addresses (that work!). That is FB's contribution. Gather those email addresses and keep them offline. Soon you may be able to form your own social networks that are private, secure and more functional than FB. But you will need the email addresses of your friends to get it started. Don't believe that FB should be the safe keeper of your email address book. They will do what they have to in order to survive as their business winds down. Those email addresses are its most valuable asset.
One lesson from reading this text is: if your site has googleable content, it will not decay as fast as if you rely on (semi-)active following. Seems quite obvious when stated this way.
I don't think Facebook will die, for two good reasons:
1) there are many more people on facebook than networks like myspace ever had. This makes a transition much, much harder.
2) facebook is not only used for facebook, but for many third-party apps as well. People will keep their facebook accounts so that they can log in to external services.
You make a reasonable claim. Lets consider the evidence though.
" there are many more people on facebook than networks like myspace ever had. This makes a transition much, much harder."
Have you considered the cost to move a single community? Facebook has a lot of users but it also has a lot of communities. So the durability will be determined in part by how interconnected those communities are. So for example if everyone in your school is on Facebook, great but who else is on Facebook that you share with? If its nobody then the 'school' community is independent of the greater Facebook user base for you, it could move en masse to another service and the members would continue to be well served. Now do you have college friends and church friends and work friends? If so then you've got a more complex social web and moving to a different service for your college friends means you now check two services instead of one in order to keep up, that is a 'negative' on your part and if enough people in a community have this same constraint then the community is more tightly bound to Facebook.
Because of that reasoning I'd claim that it does not matter how many people currently use Facebook, rather it depends on how difficult it is for an arbitrary community to move to a new service. If movement is easy (or becomes desirable) then you could see mass migrations.
"facebook is not only used for facebook, but for many third-party apps as well. People will keep their facebook accounts so that they can log in to external services."
This is clearly a 'stickyness' mechanism that Facebook has implemented for the reason you state. If you log into a bunch of places using your Facebook account then having a Facebook account is important, and leaving Facebook is 'hard'. Clearly this is in the interest of Facebook and the reason they offer the account connecting APIs that they do.
There are two potential problems with this however, one is that Facebook is less willing to reciprocate with data access to web sites that use their APIs (see the Google / Facebook problems as exhibit A) which makes that linkage less desirable to the third party. If it is a big enough problem they will drop it.
The second problem continues to be the profiling problem which is that Facebook knows a lot about you and this simply adds to their knowledge. As people become uncomfortable with that they may choose not to use Facebook login even if it is available for just that reason.
It will be interesting to see how durable these defenses are against defection. I don't doubt for a minute that Facebook will die at some point. The question though is will it be be 10 years from now, or 20, or 5. If you look at companies that have spanned several generations there is a more fundamental value proposition at work than 'friends.'
In regard to 1 you may want to think of it in terms of percentage of internet users who are on Facebook rather than sheer numbers. My guess would be that Friendster, Myspace, and Facebook may have had similar percentages in their prime and as a result Facebook may be just as fragile.
The key driver seems to always be user frustration. Friendster was slow, MySpace had UX issues - for the masses (non-techies) is there something about Facebook that frustrates average Joe?
Regarding 2), I'd like to point out that just because you "leave" Facebook for another social network doesn't mean that you'd delete your Facebook account. In other words, the decision to use a new social network wouldn't preclude you from still logging in to other services with your FB login.
Only that Hotmail has been much improved recently and even got a completely new design with Outlook.com. Also, care to elaborate on what "much better ones" there are? I barely use Facebook while Twitter is essential for me, but I don't see a better social network for most people than Facebook.
e-mail account is different, as it gives you an address that others will use exclusively to communicate with you. On the other hand, Facebook friends will use your name and the search field to look up for you, unique address is not that critical in such infrastructure.
"TL;DR: Big social networks need to take advantage of the spotlight, and solidify their position to rely less on network effects. Otherwise, they are extremely risky investments."
In fact Quora ranks pretty high in Alexa for what I believe is a relatively small user base. They must have a significant number of lurkers and decent SEO.
Growing a company is one aspect while sustaining a company is another. Social Networks naturally take advantage of network effect while growing. It is not easy to sustain the company if perceived value provided by the company erodes quickly from its core customers' point of view. Whenever a company collapses usually it is because customers do not want its products for whatever reason.
There are various historical examples of various online networks. The one I began with in 1992 was the Prodigy commercial online network (which I had actually tried out as early as 1989). My paradigmatic example of an online network that faded away is AOL, which is still in business. My prediction about Facebook, posted previously here on HN, is "Facebook will go the way of AOL, still being a factor in the industry years from now, but also serving as an example of a company that could never monetize up to the level of the hype surrounding it." I could be wrong, but that's my sense of where Facebook is in the market.
