I Could Just As Well Have Blogged About This Thursday

The New York Times reports today that it tends to take 36 hours for half the total readership of an online article to read it--and, by the definition of the study in the journal of the American Physical Society, to become "old news." The research was based on readers at a Hungarian web site origo.hu, so perhaps Americans are different. The researcher does say he was surprised by his results, since "traditional ideas about the way people use the Internet would have led researchers to expect a much shorter half-life, more like two to four hours." Certainly, the comment pattern on Hit and Run seems to indicate a much shorter half-life for posts here, closer to that two to four hours than 36. Of course, we are such bloggin' machines ovah heah that our posts rarely stay on the front page that long anyway.

Comments

[Tallocaust]

Of course, we are such bloggin' machines ovah heah that our posts rarely stay on the front page that long anyway.
Agreed. Whenever I actually go that long without reading H&R (not often, sadly), I'm behind on so many updates that my productivity is killed for at least a few hours...a few hours I'd spend doing something more productive like...watching TV.

[crimethink]

No offense, but H&R also has a much, much smaller readership than the NYT.

[joe]

Well, you're a blog, and they're a newspaper.

I rarely look at a blog post and say, whoa, too much for right now. I'll come back to this.

[thoreau]

I rarely look at a blog post and say, whoa, too much for right now. I'll come back to this.

What, you've never noticed the posts where Julian Sanchez discusses philosophy?

🙂

[jf]

I don't find this surprising at all, and it's yet another example of the "long tail". Thanks to the blogosphere and Google, people will be linked to articles years after they've been published online.

[Sphynx]

For myself, if a given blog item on H&R has more than 30 - 35 responses, I tend to skip reading any of them. This allows me to catch up on the rest of the H&R items and comment on ones where my "voice" isn't lost in the crowd.

But then, Hit & Run is the only blog I read daily due to time

[crimethink]

Sphynx,

But the 300 comment threads are the most fun!

[Brian Doherty]

Perhaps this ambiguity is my fault, but the research, as I thought I stated, while REPORTED in the NY Times, has nothing to do with the pattern of readers OF the NY Times; it is based on readers of the Hungarian Web site origo.hu.

[keith]

Last week, Jakob Neilsen published an article about how weekly distribution of existing content use continues to fit a zipf distribution. I don't know if Dr Neilsen has studied the "half life" of news content recently, but if he has, I'd expect an analysis of a deeper pool than a single foreign-culture news site with only 250,000 unique monthly visitors.

http://www.useit.com/alertbox/traffic_logs.html

I read what may be a review draft of the actual paper here:
http://mokk.bme.hu/archive/15fame_pr2005

This is based on a set of user logs from 2002, and while the half-life concept is a great addition to understanding internet content, I'm concerned by some of the (very mechanical)assumptions on user behavior. There's almost nothing in the references to research in the HCI and usability domains, which could have addressed some of the assumptions otherwise glossed over.

[Ship Erect]

This story disproved itself! I saw this posted on Slashdot a week ago.

[me]

I just went to origo.hu. Within 2 clicks I was viewing their party boat 2006 pictures (http://partyfoto.origo.hu/party/3505/page1.html). I think that if hit and run were publishing pictures of scantily clad, partying women, that the half life of those pages here would be a bit longer than a few hours. I hope that the picture pages were left out of the study, but my point is that I think content of the "news" matters.

[APL]

Old news.