We use, we bake and we eat cookies. By browsing our site you agree to our use of cookies. Okay!

Rankings create echo-chamber effect

Article-level metrics at PLoS
Image by dullhunk via Flickr

Over at Science Online 2010 I’ve asked Peter Bienfeld a question that was pretty of similar in spirit to Michael Barton’s recent tweet:

Can I get plos one article feeds for articles with greater than X page views/downloads?

Looking at the discussion over at FriendFeed, it doesn’t sound like a bad idea at all. And PLoS team is working on that feature too.  I couldn’t put my doubts about this concept into words on the conference, but it came naturally after reading Michael’s tweet. My first answer on the topic was:

Many services now provide lists of what’s popular. I’d rather see a list of what’s relevant to my work.

It’s way less work to put together popularity metrics. Rankings are easier than doing difficult text-analysis that would suggest something relevant to the topic I’m interested in. However, it’s only part of the story.

Current science is pretty much an echo-chamber. We do things that we can get funded, which can be published in “prestigious” journals – in both cases review process filters out ideas that are too far from the mainstream. Adding rankings on top of that is another way of making sure we won’t even have ideas that are too far from the mainstream. We won’t read anything weird, obscure, outside of our area. We won’t read anything that can inspire us to do something other than stamp-collecting.

Sounds bad, doesn’t it? Maybe not. Because rankings can be gamed (as most of automated algorithms), such development create an unprecedented situation in which we are given tools to create mainstream science. It would require lots of work, but it sounds perfectly doable to publish several articles in a short period of time, elevate their rankings immediately after publications, create viral campaigns and then let the old media to pick up the topic. As long as the idea sounds any good, it would reach mountains of hype within relatively short time.

One reply on “Rankings create echo-chamber effect”

  1. This is a great idea, Pawel. To me the main utility of a social network is the ability to go beyond what's popular to site visitors to what's popular among people with similar interests. This isn't to say PLoS needs to add social features to their site – they can use social data from other services (people on a specified twitter list, for example). I don't know why more sites don't do this, actually.

Comments are closed.