Comments on: The Long Tail of Popularity

By: Nuno Bastei

Nuno Bastei — Fri, 12 May 2006 13:01:19 +0000

It reminds me the Athenian democracy, when Aristotle banned himself from the City Council after founding himself too much popular to make relevant decisions and enlighten the daily life of the Athenian.

If blogs are boosting their author’s popularity outside the Net, and that probably will happen, will you trust Michael Arrington if he runs for the US presidency?

By: Pinoy Reading Lists - Migs Paraz - Random Takes

Pinoy Reading Lists - Migs Paraz - Random Takes — Tue, 07 Mar 2006 17:00:32 +0000

[...] What posts are important? For insights, see The Long Tail of Popularity by Joshua Porter. [...]

By: Kevin

Kevin — Mon, 06 Mar 2006 05:11:03 +0000

“But I think there is much more to popularity than unwarranted attention”

Spot on! Great entry….

By: Clark Valberg

Clark Valberg — Thu, 02 Mar 2006 15:38:01 +0000

Blog comments are a perfect example of where a filtration mechanism might be put into place to enhance the systems overall value to the reader. What if each post had a “star” rating system (think netflix) the lower the average rating, the less prominent (eg: lighter text color) the post. Great comments would stick out - and BS would just fade away. I dont know about you, but when I hit the bottom of an article I dread the amount of time i’m going to spend reading the 50 or so comments below… It becomes frustration when every other post adds little value to the conversation, yet takes just as much time to read. I guess you might call this “Display by Committee”

By: Michal Migurski

Michal Migurski — Wed, 01 Mar 2006 20:36:23 +0000

Ah, apologies for the misunderstanding, Josh. =)

I have perceived a bit of algorithmic messianism in the various attempts at good meme trackers, I think as a result of Google’s faith in mathematical models of social interactions after the first few years of PageRank’s success. It’s probably fine if a tracker’s user population is known or trusted - the votes matter more if I already know the bouncer at door checked everyone out.

By: Josh

Josh — Wed, 01 Mar 2006 19:25:41 +0000

Michal, I believe we are agreeing. When I say “model” I mean “give tools that allow us to model online how we do it offline”.

The Amazon feature of “Was this review helpful?” is a *great* example of what I’m talking about. It’s a model of how we can judge whether a review was helpful or not, thus helping others who come after us. Offline, we would disagree in some other way. Online, this way seems to work.

So, I’m definitely not saying that we can create memetrackers that take away our editorial control. In fact, that’s what they should optimize for.

Sorry for the confusion.

By: Scott Karp

Scott Karp — Wed, 01 Mar 2006 19:14:21 +0000

Great thought piece. A few things:

- Most bloggers get satisfaction from achieving popularity because they think what they’re writing IS important

- There are clusters in the network where popularity and importance overlap — these are created by “tribes” with common interest. Digg is a cluster of people who think geekily random information is important

- Popularity is “good” when popular information is also important to the people who are driving the popularity

By: Michal Migurski

Michal Migurski — Wed, 01 Mar 2006 19:11:11 +0000

You lost me at “we need … to model that in our software.” Why? Why, why, why? People are great at filtering, and it’s easier to write software to support human communication than to supplant it. In six months, this drive towards automagical meme tracking is going to look like a hardcore boondoggle, as I mention in a recent post: http://mike.teczno.com/notes/death-to-search-engines.html