A favorite feature of FriendFeed is the Like. You get to indicate your interest in an item with a simple click of the Like button.
The act of applying a Like does two things:
- Provides feedback to the content creator
- Reveals what your interests are
It’s that second point that is interesting. Amazon.com matches you to other shoppers based on what you buy in order to provide recommendations. Toluu matches you with others based on common RSS feeds. Diigo matches you based on common bookmarks and tags.
How about matching people based on common FriendFeed Likes? Call it the FriendFeed Likes Compatibility Index.
Curious about this, I went to my Likes tab on FriendFeed. I went back to my 50 most recent Likes, and tallied the number of Likes by others. By doing this, I figured I’d see with whom I had the most in common.
The top 29 people are shown below – I put the cutoff at having 4 Likes in common. Some of these folks I know, others I really haven’t interacted with yet.
Here are my top matches in FriendFeed:
- Atul Arora (13 likes in common)
- Louis Gray (13)
- Mitchell Tsai (11)
- Shey (11)
- Robert Scoble (10)
- Thomas Hawk (9)
- Julian Baldwin ( 8 )
- Jason Kaneshiro ( 8 )
- Mark Trapp (7)
- Charlie Anzman (6)
- Mark Dykeman (6)
- Bearded Dave (5)
- Bwana McCall (5)
- Mack D. Male (5)
- Mike Fruchter (5)
- Phil Glockner (5)
- Alejandro S. (4)
- Andrew Badera (4)
- Anthony Farrior (4)
- Dobromir Hadzhiev (4)
- edythe (4)
- Kenichi Matsumoto (4)
- Marco (4)
- Nikpay (4)
- Rob Diana (4)
- Ruth Ferguson (4)
- Shawn L Morrissey (4)
- Susan Beebe (4)
- Timothy Neilen (4)
One small observation – I’m not in sync with a lot of women, am I? What’s up there? FriendFeed Is from Mars, Twitter Is from Venus?
Now what I need to do is to subscribe to those on this list that I haven’t yet. Also of note – there were 241 different people with whom I shared a Like in this analysis. Really great how FriendFeed lets you come into contact with a wide range of people.
Would be cool if a script could automate the FriendFeed Likes Compatibility Index…
*****
See this item on FriendFeed: http://friendfeed.com/search?q=%22FriendFeed+%27Likes%27+Compatibility+Index%22&public=1
I could automate this….
…If friendfeed fixed this bug http://blog.yuvisense.net/2008/05/26/friendfeed-bug-makes-it-useless-as-an-archive/
Comment by Yuvi — May 28, 2008 @ 11:43 pm
Interesting point for analysis! If nothing else, social network graphs are a statistician’s playground.
Comment by Andrew Badera — May 29, 2008 @ 12:45 am
My major problem with the “Likes” is that judging from our blog’s stats like does not actually mean that you are really-really interested in the content. For example, for a post with over 30 likes only 20 people actually clicked through to see the post itself. To me it means that people only click “Like” to ensure their presence on FF, not to actually join the conversation or show their real interests. Pretty hard to graph something on such type of information, no?
Comment by Svetlana Gladkova - Profy — May 29, 2008 @ 4:59 am
FriendFeed is holding all sorts of information to characterize the user better. This is a good start to generating conversation over what types of features the community might like to see in the future. Good post Hutch!
Comment by Julian Baldwin — May 29, 2008 @ 5:11 am
Interesting post. I find that I sometimes use the Like button not to indicate that I really like something, but just so that I can find it again later on my Likes or Discussion pages.
Comment by Mack D. Male — May 29, 2008 @ 5:44 am
I could use you as an index to see how well I’m blogging.
Comment by markdykeman — May 29, 2008 @ 7:23 am
Svetlana, that’s one view, but you’re not seeing the whole picture.
If I saw a Profy story on my RSS feed, and liked it, or even shared it, and later I saw it on FriendFeed, I could click “like” and yet, I didn’t have to visit it again to read the story and show the referral click-through.
Also, it’s possible many folks “like” headlines. It’s not about FriendFeed presence, but more so about liking ideas they agree with.
Comment by Louis Gray — May 29, 2008 @ 8:24 am
[...] Carpenter had post about what he called the “Friendfeed Likeability Index”, calculated them for a tiny sample of his own likes, and found them pretty good [...]
Pingback by FriendFeed “Likes” Compatibility Index Pre-Pre-Pre Alpha - The StatBot - Fun stats. Visualizations. Leaderboards. — May 29, 2008 @ 9:22 am
http://thestatbot.com/2008/05/29/friendfeed-%e2%80%9clikes%e2%80%9d-compatibility-index-pre-pre-pre-alpha/ Louis and Scoble’s there…
If you want yours, just leave a comment, I’ll generate it and put it up…
Comment by Yuvi — May 29, 2008 @ 9:36 am
Your’s are here: http://thestatbot.com/2008/05/29/friendfeed-%e2%80%9clikes%e2%80%9d-compatibility-index-pre-pre-pre-alpha/#comment-205
Comment by Yuvi — May 29, 2008 @ 10:16 am
[...] came across this link on Friendfeed of someone who figured out how many people liked the same things he’s recently liked. a bit ago and was curious about it for myself. As ever too lazy to [...]
Pingback by Friendfeed Like Compatibility Calculator — May 29, 2008 @ 11:05 am
@Andrew Badera – yeah, FriendFeed is virgin territory for stats analysis.
Comment by Hutch Carpenter — May 29, 2008 @ 1:25 pm
@Svetlana – that certainly is the glass-is-half-empty view of it. But I wouldn’t damn Likes based on that point of view. There’s still a lot of information there.
Most people are still clicking the links. And suppose people are clicking Like not for the content itself, but the ensuing discussion in the comments? That’s still good to know.
Comment by Hutch Carpenter — May 29, 2008 @ 1:28 pm
@Julian – thanks. I suspect there’s a rich vein of analytic-type of features that could be rolled out for FriendFeed.
Comment by Hutch Carpenter — May 29, 2008 @ 1:29 pm
@Mack – the Like as bookmark is a use case of mine as well. But the fact that I wanted to come back to it later still counts a form of interest, which is what the Like is.
Also, I created my own private room where I re-share things I want to remember. Rooms as a bookmarking vehicle.
Comment by Hutch Carpenter — May 29, 2008 @ 1:31 pm
@markdykeman – I love Broadcasting Brain (http://broadcasting-brain.com/)
Comment by Hutch Carpenter — May 29, 2008 @ 1:32 pm
@Louis – good point on Likes for headlines. Sometimes a headline is THAT good.
Comment by Hutch Carpenter — May 29, 2008 @ 1:33 pm
@Yuvi – love the work you did on automating this. Check out his blog for more info: http://tinyurl.com/6ba7oa
Comment by Hutch Carpenter — May 29, 2008 @ 1:34 pm
@hutch – that’s the spirit!
Comment by markdykeman — May 29, 2008 @ 3:20 pm
Great idea, and great to see that it’s already provoked a response! These kind of app developments and additional functionality are what really make services like Friendfeed invaluable.
I must admit that I don’t use the “Like” button as much as I perhaps should. I’m too wary of Friendfeed becoming too scattergun in terms of things being rated, as many other services are. But once I extend my usage then the pre-pre-pre Alpha that Yuvi is working on is well worthwhile!
Comment by Robin Cannon — May 29, 2008 @ 3:22 pm
I use my Like very sparingly. There seems to be this want for FriendFeed to be THE platform for everything: conversation, bookmarking, message board, feed-reader… whatever happened to doing one thing well? If I want to bookmark something, I’m still using ma.gnolia; as Yuvi has shown, stuff isn’t in there forever, so I’m going to lose it if I rely on FF to save it for me. For me, the Like is one of two things: “Hey, cool thing you did here/linked here” to the originator and/or “Hey, this thing is cool and I want to bump it back to the first page so more people see it.” That’s why I think MY likes have more value.
Comment by Cyndy Aleo-Carreira — May 29, 2008 @ 5:03 pm
Hutch – Just wanted to let you know I really ‘like’ this
Comment by Charlie Anzman — May 29, 2008 @ 7:53 pm
[...] FriendFeed ‘Likes’ Compatibility Index :: I’m not actually a geek – Hutch takes a look at the "Like" feature on FriendFeed and why that is good way to find other folks you might like to follow. [...]
Pingback by WinExtra » From the Pipeline - 5.29.08 — May 29, 2008 @ 10:03 pm
[...] utility looking at the last 30 300 likes. Mark Trapp posted about it here, and Hutch posted here. a2a_linkname=”My Pre-Pre-Pre-Alpha FriendFeed Compatibility Index”; [...]
Pingback by [scribkin] My Pre-Pre-Pre-Alpha FriendFeed Compatibility Index — May 30, 2008 @ 5:57 am
[...] FriendFeed ‘Likes’ Compatibility Index « I’m Not Actually a Geek- FriendFeed statistics are geeky and cool. [...]
Pingback by Linked June 1-3 2008 | FPettit.com — June 4, 2008 @ 8:11 am
[...] stole the concept for the “FriendFeed Likes Compatibility Index” from Hutch Carpenter (you can find a working implementation here courtesy Felix (aka the Topless [...]
Pingback by Flickr Faves Compatibility Index - The StatBot - Fun stats. Visualizations. Leaderboards. — June 13, 2008 @ 6:34 am
[...] FriendFeed ‘Likes’ Compatibility Index [...]
Pingback by My thoughts on social networking applications. » Blog Archive » Friendfeed.. How do you like it? — July 1, 2008 @ 7:11 pm
[...] FriendFeed ‘Likes’ Compatibility Index [...]
Pingback by I’ve Joined Connectbeam, and Social Media Got Me the Job « I’m Not Actually a Geek — August 13, 2008 @ 5:29 am
[...] FriendFeed ‘Likes’ Compatibility Index: I start with a blog post that achieved success within my favorite haunt, FriendFeed. And no surprise. It was about FriendFeed, and it caught a vibe that was powering FriendFeed. We were all meeting really cool people there, and this was a way to find even more users with similar interests. Between my feed and Louis Gray’s share of the post, there were 79 Likes and 68 comments. Plus action around many other shares of the post. And the post led to a really cool app produced by felix. [...]
Pingback by Getting Overly Focused on Your ‘Regular’ Blog Audience « I’m Not Actually a Geek — August 20, 2008 @ 6:44 am
[...] could get pretty busy. The only real discussion that caught my attention was around the “FriendFeed Likes Compatibility Index” – using commonalities between FriendFeed likes as a way to identify other users that [...]
Pingback by Take time out – gain perspective. » Walker Media — June 19, 2009 @ 10:39 pm