≡ Menu

Are your friends a filter or a firehose? Some musings on Twitter and FriendFeed

I was playing around with FriendFeed – again – and not seeing the point – again, when a thought came to me which I think encapsulates why I don’t get on with it, and why, in fact, I’ll never get on with it.

To start with, consider that social networks like FriendFeed, Twitter and Facebook are all being required to perform a very similar sort of task. The end product is a set of links, comments and conversations which should be relevant to me. That’s the key thing: Social networks should ensure, somehow, that things which reach me are relevant to me.

In FriendFeed, “friends” are actually a firehose. They’re the raw, unedited stream of information, which you then apply all the powerful filtering and management tools that FriendFeed has to. The end product, if you’ve set up your filters right, is all the stuff from that firehose that’s relevant to you.

That’s why FriendFeed perfectly suits Robert Scoble, who is, if he’ll forgive the phrase, the master of the firehose. Robert currently follows over 24,000 people on FriendFeed, and a lot of groups (which means he gets stuff from lots of people he’s not even following). That’s a firehose of information. The filtering, for him, comes after that – not before.

But there’s a different way of performing the same process: Make your friends the filter. On Twitter, I currently follow around 300 people, most of whom I either know from meatspace, have sparred with online for a while, or who are names within my industry who I trust.

Having a selection criteria, rather than just following anyone and everyone, turns my friends into my filter. I trust them to bring me information that’s important: I don’t need to filter them, because they never turn into a firehose.

This is why I look at FriendFeed and go “Huh?”. My friends, my human editors and curators of information, accomplish the same thing as the filtering tools in FriendFeed – they bring me information that’s important to me, filtering our crud. I don’t see that Patrick Swayze has (not) died: I do see the price and release date of the Palm Pre.

The question of which is a better approach makes no sense. For Robert, FriendFeed does the job. He wants to be able to take that firehose, slice and dice the massive stream of data that flows out of it, and find interesting stuff. FriendFeed is a data analysis tool, with “friends” as the source of the data.

My approach works for me. It brings me the information that’s important to me, in a timely fashion, because my human editors act as a filter, not a firehose. The big pool of data is their experiences, their lives, and they filter that for me.

Which approach will win out? Neither. I suspect that the firehose will be more popular with a small crew of geeks, while the “friends-as-filter” will prove to be more popular with everyone else. But we’ll see.

(Photo by Seattle Municipal Archives – http://flic.kr/p/59nVfL)

Reblog this post [with Zemanta]

Comments on this entry are closed.

  • http://messel.typepad.com/ Mark Essel

    A well developed perspective, we may find mixing of the two filter techniques and a little further in the future perhaps even artificial intelligent filters acting like friends but with full access to your online history-> in effect virtual versions of yourself as a filter.
    We just have to ensure that they don't miss out on serendipitous discovers that can sometimes be incredibly effective.

  • Pingback: Mark Cuban sums up why Rupert Murdoch doesn’t care about Google | Technovia()

Next post:

Previous post: