Entries Tagged 'Community' ↓

On Facebook friend limits

In two recent posts (Facebook sucks, The you-don’t-need-more-friends lobby, Robert Scoble has complained about the 5000-friend limit in Facebook. He said,

[…] it isn’t scalable and falls apart at 5,000 contacts. It pisses me off more and more every day because of that scaling wall.

and

Someone asked why I keep pointing out the 5,000 friends limit. Why? Because I still haven’t gotten through and I’m still getting pushback from the lobby. So, let’s try one more time.

I agree with Scoble to some extent. The issue is important: scalability, both in terms of network size and application quality across different functions is where sites like Facebook will live or die. Plus, Scoble is an edge case, clearly, and I always think edge cases are important for webapps to take into consideration as they grow and develop.

There is a problem, however, and it is quite simple. The problem is this: Facebook seems to have been explicitly designed to NOT be a rolodex, to NOT scale to thousands of friends per user, and to NOT be an application that scales to the needs of highly-networked businesspeople as a function of their job. I don’t think it’s just a difficult engineering problem that will be overcome with time - it’s intentional. And, more than that, I think that accounts for the success of Facebook.

Facebook’s success isn’t just dumb luck (though there was almost certainly some luck involved). There are lots of other social network sites out there - including MySpace, which is still (arguably) bigger than Facebook. Of course there’s also Friendster, Orkut, Tribe.net, and many more. I’ve been on all of these, and a whole bunch more to boot (more task-oriented sites like Flickr and Shelfari). Facebook, however, is much bigger (for me) that all of these. Why?

  • Facebook has always been popular outside the tech/internet/weblogging crowd. On Orkut and the rest of them I always had the same kinds of contacts/friends. Some YULBloggers, friends who I’ve been in online communities with since the mid-90s, tech/business folks involved in other social networking or web properties, etc. My college roommate? Never. The girl I dated for a few months in twelfth grade? Never.
  • In Facebook, you can’t “collect” friends very easily. This is a huge difference from almost all of the other social networking sites, which explicitly privilege friend collection.
  • It’s easy to say “no” in Facebook. There isn’t even a message to the requester when someone declines a friend request. In some of the older systems, it was a big deal to not accept someone’s request - I suppose they thought that the social pressure to accept requests would grow the network faster. Without that social pressure, though, a paradoxical thing happened: even though it might have slowed the growth of the network, the whole system was non-threatening to non-professional users, who stuck around (in droves).
  • In Facebook, your friends list isn’t the most prominent part of the interface as it was for almost every similar system in the past. Rather, it’s what your friends are doing that’s front-and-center. Having more or fewer friends is simply a way to have more or less “stuff” parading up the screen from hour to hour, day to day. The “social graph” isn’t fundamentally about the list of friends - the social graph is just the conduit to the real content: interaction opportunities with other people.
  • There’s a cost to the user for each friend that they add. Each friend makes it that much more difficult to follow what any one friend is doing. This social cost is important, because although it puts negative pressure on the number of friends a user has in his or her list, it also helps most users to keep their friends list uncorrupted by non-friends who are simply collecting names (like they do in MySpace).

In all of these, there is a common thread: Facebook is expressly NOT there for collectors, it tries to privilege real life relationships as opposed to fleeting acquaintance. Scoble says,

[…]a “friend” in Facebook is NOT a “real friend.” (Let’s define “real friend” for now as someone who you’d invite over to your house for dinner). In social networking software a “friend” is someone you want in your social network. Period. Nothing more. The fact that people assume that you should only have “real friends” in your social network is just plain wrong.

In fact, I think that the distinction between Facebook and the other social networks - and the primary reason for the success of Facebook - is that is about “real friends” or at least is trying to be. MySpace and the rest were about collecting as many “friends” as you could. They were about the scope of the network above all things. Facebook isn’t about the network, but what the individuals in your network are doing: events they are going to, photos they are posting, games they are playing, chumps they’re biting, and all the rest.

Maybe the Facebook engineers were being honest with Scoble and it is just an engineering issue. If so, then they’d better watch it - because anything they do to remove the negative pressure against “collecting” instead of privileging the content over the network for its own sake could be Facebook’s downfall.

I just noticed that

I haven’t yet blogged about one of the most significant things I’ve been involved with on the net for a while: the move of the Café Utne to a new organization and a new server: The New Café. The Café is an online community that opened in 1995 and in the move, we moved over the entire archive of almost 6.5 Million posts over to the new server. There’s quite a bit of work to do yet - branding, capacity-building, and such, but that’s in the hands of a very capable General Manager so I expect all to go well in the coming weeks, months, and years.

Kottke on comments

and busy or high-profile websites: High volume flow. This is a problem with weblogs in general. Everyone has been touting them as the next generation in online community, which is true, but not because of the comments sections, though ironically this is the feature that most resembles traditional online communities such as the Well or Cafe Utne. The weblog world is community-like because of the give-and-take between and among “peer” weblogs via links-and-commentary posts, and it is in that respect that the weblog world is and encourages online community. The existence or not of the ability to comment has little to do, in my opinion, with the community-ness of weblogs.

The solution to the problem is simple and has existed for much longer than weblogs have existed: old-school online communities such as those mentioned above, with mandatory registration, active, consistent moderation, and persistent membership over time such that individuals have a social motivation to behave.

Curious:

Does anyone know how/where to download a copy of MetaPhilter? There’s a project I want to do and I am curious to see if it will do the trick or if I should try something else.

Adaptive Path,

the user experience agency, has always been in the “Web 2.0″ mix, and last week Brandon Schauer published an interesting article, “experience attributes: crucial dna of web 2.0“. It’s a pretty good piece, but I always wonder - what about the fact that there was a TON of very “Web 2.0″ stuff going on well before anyone actually wanted to live South of Market St.? The WELL? Very Web 2.0. The original HotWired? Very 2.0. I don’t think it’s a coincidence that Steadman owns Plastic, which is clearly a proto-2.0 space. For almost everyone who was involved - professionally or not - in the WWW back then knows that Web 2.0 is really simply things getting back on the course we set 12-15 years ago.

I wonder why everyone seems to ignore this?

Clay Shirky at Many-to-Many

Community != content. I have been fighting with librarian-types over this sort of thing for a while now, and it’s an uphill battle, not in the least because they have all the resources (and cred) afforded to an accepted academic discipline behind them. But make no mistake - understanding this is the beginning of understanding how to successfully do so-called “knowledge management”. [got there via Stewart, whose comments were right on as well]

About mikel.org

This is mikel.org, a weblog that has been published by Michael Boyle since January, 2000.

Links

Community

Friends etc.

Tag Cloud

Some Rights Reserved

This work is licensed under a Creative Commons Canada Licence.

Add to Technorati Favorites

Subscribe to mikel.org's syndication feed.
 Atom feed

YULBlog