Digg and Sensationalism
It annoys me that stories are being written for Digg.
There are sites that tell you how to write your articles for Digg, including such gems as Make A List and Make up outrageous statistics. I've commented on the list aspect before.
I'm breaking (or, I guess, following) rule #2 now: Write about Digg. But it does seem to work.
And that's the problem - they all seem to work. They work really well. Being on the Digg front page gets you a lot of traffic, and all you need to do to get on the front page is tailor your content for Digg.
I'm sure this wasn't the intention when Digg was created. It's supposed to be a way for users to submit existing web content, and then for other users to vote that content up so more people see it.
But we've got techniques on Digg like cliffhangers (having the summary end with a colon so you need to click it to see the interesting bit), and "pictures included" (which seems to be a big draw - hey everyone likes pictures) and people making up top n lists of simply irrelevant things (seriously - make up a list of the top 7 songs with the word 'tomato' in them, and you'll get traffic).
Here's my point:
User-generated content works, but user-submitted content does not.
Millions of users around the world are generating content. Digg could be a great system for finding them, but it's turned into something much less useful. Users who have figured out the system are using it to promote their own sites or blogs, at the expense of everyone else.
We need a new site that's different from Digg in two significant ways:
1. Users can't submit stories. The site needs to include sites based on some criteria other than users submitting it. I have a couple of ideas here. There's my LongTail site that randomly presents articles that you have to decide on without knowing who wrote them, and I think that works pretty well, but we need something more people will use.
2. Top content is per-user. The idea that I'm going to like the same content as every other Digg user is just wrong - But there are other people out there who will have similar likes as me.
Most of the good links I find these days, I find through the blogs I read. I use Google Reader, and when I see an article I think is interesting, I click the Share button. I've never given out the link to it (until now), but this leads to the possibility of using Google Reader or other blog data, or attention data from a browser, as input.
Sounds good until you remember that this, again, gives users a way to submit content. Unless we're only going to take content from people we trust, we can't take any data that users submit.
So either we're only going to take trusted content, which can lead to an echo chamber sort of effect where only a small number of links make it "into the circle", or we need some sort of efficient content triage for all new content posted to the Internet.
The LongTail site is currently throttled to just one post per minute, because it doesn't get a lot of users, and frankly pulling every new piece of content posted to the Internet all the time is just more work than my little server wants to do. But imagine if this were hosted at Google, and if there was some incentive for users to help triage the content.
And maybe that's where Google has a unique capability again. It's in their interest to drive good content, and especially, good content monetized with AdSense. Would offering users a tiny percent of revenue generated from pages they triaged be possible? And would it help?