Dangers of E2.0 and Semantics

Did UCLA Profs Just Find Osama? Only if He’s an Idiot | Danger Room from Wired.com

As is so often the case in a media-driven world, the reality of the case is quite different from the headline. I chose this example because I'm a geographer by training, so it jumped out to me; but we could choose almost ANY science or statistics-based news story out there.

Where's the E2.0 beef? Well, we have to trust that people create content or tag content for altruistic reasons - not to get hits, kudos, or tenure. It's potentially a problem and we need to put procedures and incentives in place to reward good quality content rather than flashy headline content.

Update - based on Billy's comment
So this is where we revert to type. Billy is optimistic and believes in the wisdom of crowds, I tend to be somewhat more cynical about people's motivation. I do think that we can minimize to a certain extent the negative effects of headlining, exaggerating, and tagging by the methods Billy describes, but I'm not convinced it's all so simple. The trouble is (as with the example above, but perhaps I didn't explain my thought process clearly enough) that it's tough to control for cr*p that sounds convincing.

Don't believe me? Here's a perfect example of something worthless that bears all the indications of being useful. It's in a seemingly reputable place (although if you dug around, you'd find it's not peer-reviewed), it has graphs and math, it generates a lot of traffic and hits on google. But, it's completely wrong. Even if we assume the highest of principals to the author, the publishers, recommenders, and taggers - it's quite possible that every one of them wanted to be helpful - we still ended up more confused or misinformed after the fact than before.

Sometimes crowds are dumb.

Sometimes authors are misinformed or incorrect, or they seek to inflate their ego / rankings / visibility.

I'll hope, like Billy does, that these are exceptions rather than the norm, but we should be aware of the fact that our systems can be gamed and prepare for this eventuality.

Comments:

The problem of bad metadata as Bex says or non-altruistic tagging as you say is only a "problem" for half-a** implementations. If you allow tagging but don't aggregate, if you allow folksonomy and ratings but only show author/originator attributions then it's not really an E20 implementation. It's rollout of tagging software without realizing the benefits. In the recent InformationWeek Report on "The Web 2.0 Balancing Act" (worth a read - pdf and registration warning) George Faulkner of IBM's Media Library which heavily leverages E20 social software says that the Wisdom of the Crowds is enough to rein people in though it might take some time. I agree. With network effects, bad metadata (spam etc) become outliers, marginalized, shunned or called out by someone else. Of course that all presupposes a critical mass of participation within the topic domain. Still, I think that such issues are exceptions rather than the rule and should be treated as such.

Posted by Billy Cripe on February 24, 2009 at 07:11 AM CST #

Post a Comment:
Comments are closed for this entry.
About

Enterprise 2.0 and Content Management

Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today