I read Clay Shirky’s A Group Is Its Own Worst Enemy. The article was first published in 2003, but it’s still very interesting.

Some notes.

Shirky defines social software as software that supports group interaction. This is something new: the last technology that supported group interaction (and still does) was the table.

Some behaviors come from individuals and only appear as coordinated. Some others are group-related. So you can study a group as multiple individuals, nor as unique entity. You have to handle with individual behaviors and group effects.

The patterns in which a group interaction will evolve are three:

1. members will start talking about sex (eg. most of chats)
2. a common external enemy is identified and the group will cohalize against it (eg. open source people against Bill Gates)
3. the religious thing: people identify something that’s beyond any critique (eg. football team, movie director, book)

To avoid the disgregation, you should give a structure to the group. Note that this should be a technical AND social solution. Technics-only solution wont distinguish between normal behaviors and abuse.

The second part of the article try to explain why the web 2.0 thing is happening right now. Although the article was written in 2003, this part is still valid. Shirky points out that the technology needed for the web 2.0 was there since the first Mosaic was launched (~1994). We got Geocities instead of Blogger because we did not know what we were doing.

Now we see that this thing is going mainstream. This thing is different from everything we saw before: it’s web-native. It’s not something adapted to the web, this is really born on the web.

The last part is on social software design. Shirky think you have to accept three thing and deal with other four things.

The things you have to accept if you’re designing social software are:

1. you cannot completely separate technical and social issues
2. members are different than users. Ie. some users are more users than others. The core group, the every-day users, should have the means to protect themself against one-time-users that may corrupt the group
3. the core group has rights that trump individual rights in some situation. This is not something you may decide. This will happens, because it’s the group’s instinct of surviving. Citizenship should be different from ability to log on. The core group will find the way to protect against the tyranny of the majority, or the group will die.

The four things to design for:

1. the identity thing: identity should grant reputation. This does not mean that everyone have to use the real name. This means that user cannot change their nickname
2. you have to design a way for there to be members in good standing, such as identity (reputation), a sort of karma, a “member since” thing
3. you need barriers to participation. Eg in Slashdot: everyone can read, anonymous cowards can post, non-anonymous cowards can post with higher ranking. And it’s difficult to be able to moderate
4. you have to find a way to spare the group from scale

Shirky ends saying that “the act of writing social software is more like the work of an economist or a political scientist”.

PS: since my wordpress categories are a mess, I’m testing technorati tags

[tag]article, social web, groups, shirky[/tag]