I posted this because there is a persistent belief among technorati that social networks benefit from large network effects that prevent users from switching, even if the underlying software sucks. When I've said "No, users just use the social network as a venue to tell everyone where to find them, and then whole groups of people migrate en masse," people have asked me "Well, do you know of any cases of this actually happening." Well, here's one.

This is not unique at all - LiveJournal has had persistent problems with whole cliques of users splitting off to form other sites, eg. JournalFen, Plogs, uJournal, and GreatestJournal, because of the ease of forking their codebase. Hacker News started largely as dissatisfied Reddit users moved to a site where the quality of debate was higher. FictionAlley started after dissatisfied users migrated en masse from Fanfiction.net.

1) Encourage the uploading of lots of pictures without an easy way to download them again in a batch. Same thing goes for chat and mail messages.

2) Encourage high friend counts via the friend suggestion tool so that a small number of friends leaving as a clique have minimal impact on your overall experience.

3) Hide the defriending process so you aren't notified when friends leave the service. In fact the news feed, has so much in it now, you probably wouldn't notice anyone's absence even if they did notify you. You'd probably not see a "goodbye" group half your friends were joining, either...

Which is to say, Facebook may retain users a little better than Livejournal.

FB competitors seriously missed a trick when the new real-time feed was rolled out. A lot of my friends hate it deeply, but there wasn't really a good competitor ready and waiting, so people stuck around.

MySpace looks like it's trying, as it adds ever-more FB-esque features, but the brand is tarnished and the site is burdened under the weight of legacy accounts with their hideous custom CSS and auto-play music. The hordes of spammers don't help either. The solution there is either to clean house (a big risk), or set up on a new domain. Either way, the window of opportunity is almost closed.

1) In most cases, these people have 7 years of accumulated journal entries, all tagged and memory'd, which is a significant chunk of their life. And these are harder to get at than FaceBook photos, because you typically write a LJ post directly into the browser, yet you upload photos from your computer. I dunno about you, but I keep a backup of every photo I send to FaceBook.

2) Ditto LiveJournal via communities, at least among this user base. I've got 200+ friends on LJ, though most are inactive, having moved to FaceBook when they went to college. Some of the people in question have hit the LJ technical max of 500 friends (I think this is now 750, largely because of their complaints). Instead of their high friend counts preventing them from leaving, though, it makes the impact worse. They act like "superconnectors" in The Tipping Point, spreading the word to lots and lots of their friends.

3.) The defriending process on LiveJournal is also silent; it happens to be noisy here because people announce that they're leaving (which they can do on FaceBook too, through shared links, wall posts, status messages, etc.)

I think that the real reason FaceBook hasn't seen an exodus is:

a.) As mentioned elsewhere, they're really not that bad a site. FaceBook still actively develops the code, they've got a real corporate communications department, and they're basically a "serious business". That prevents them from making the sort of pissing-off-users fiascos that LJ has been plagued with.

b.) They've tried very hard not to cultivate passionate users. FaceBook users basically use it as a utility to keep track of friends and don't think too much about it, so when they leave, they just stop using it and don't try to convince all their friends to leave with them. While LiveJournal users tend to let their lives get very, very wrapped up in LiveJournal, so they have a large emotional investment in it and get very upset when that investment is betrayed.

It's not just that; tagging I'd argue is FB's killer photo feature (and weirdly something LJ could do easily, but never did). It's easy on FB to attend an event, then associate photos with that event and with the people and it's all cross-linked. Even if you download the photos, if you leave FB you lose all that linkage.

Hiding the defriending process is pretty common, if not universal across all social networks.

I suspect the lock in isn't really that great, people only really care about the last few weeks in most cases. The main lock-in is that your friends are still there even if you leave.

Ultimately I think people won't need to leave networks like this when they become active elsewhere, they will instead use aggregation to keep track with their friends on the old networks (for example in my product http://friendbinder.com) or possibly built into to new networks if they work how to do this properly.

At the very least, people will continue to use facebook to check their friends's updates. Some people use it for more than that, but it just depends.

Leaving any social network for something better isn't a huge deal because the "worse" one will still be there if you need to access your cache of photos or look up someone's phone number who isn't on the new network.

(Of course, I should note that I've been an LJ user for seven and a half years and have met lots of people who have been important to me through it, so I'm not exactly an unbiased observer.)

Freakonomics noted that baby names are usually introduced by wealthier folks and later adopted by poorer people. At that point, they are shunned by the wealthy.

Is there a reason that social networking sites should not exhibit similar patterns? Quite a few articles have been written about how the cool kids went to FB first. Are we simply witnessing the expected?

* Even if you hate going to these places, you probably are still aware of them.