How to test Social News sites
Step 1: Find a blog article that you find interesting and insightful
Step 2: Submit it to reddit/digg/newsYC (the title of the submission should probably just be the blog title)
Step 3: Wait 24 hours, has it been upvoted more than a dozen times?
The submission is almost always ignored entirely. Because there are just too many submissions. They can’t all be on the front page, so readers “vote up” content that is:
- Attention Grabbing (eg, GOOGLE TAKES OVER DIGG!)
- From a mainstream site (eg, Breaking on BBC..)
- Gimmicky (eg, kittens)
These aren’t bad things. But chances are your well written, interesting blog post is more interesting to at least some readers. What are sites doing to counter this?
Digg - Social Recommendation System (see here)
Reddit - Sub reddits & relying on users to a) submit to correct subreddit and b) assume readers are subscribed to it. (for example)
Mixx - Tags (for example)
Is this enough for readers? Should sites be doing more to link readers with interesting and relevant articles?
Where is Kevin Rose taking Digg?
Kevin Rose gave a fascinating talk about where Digg is headed at The Future of Web Apps.
He clearly believes the future of news is in its recommendation engine. They’ve had some serious growth thanks to their engine:
Hang on a second. Digging activity is up 40% from launch? Of course it is, the site is growing — not necessarily because of their recommendation engine.
Commenting is up 11%? Maybe because they implemented their new commenting engine under a month and a half before their recommendation post?
I’m not doubting their recommendation system isn’t fueling growth - I’m sure it is. But how much growth? Do users of Digg want recommended news?
So where will Kevin Rose take Digg next? Digg is in an ironic situation where as it grows faster and faster, the signal to noise ratio of good : crappy submissions increases spectacularly. Submissions are based on their “coolness” primarily on how well written the Digg title is — not the quality of the article is.
And thats the inherit flaw with Digg - users can’t be trusted to submit reliable news, because its a game of getting on the front page, look at how great reddit was 18 months ago - and how it is declining.
Some of the services we’ll be focussing on are sites that rely on algorithms to decide what is “good” and “relevant” news - making users simply readers, rather than editors.
Recommended News Sites Recommended by news.YC
After our first post about news recommendation engines, we got some amazing feedback. Firstly a list of all the recommended sites to look at:
http://jaanix.com
http://windycitizen.com
http://www.newscrawler.it
http://www.buzzup.com
http://techmeme.com
http://newsbrane.com
http://outside.in
http://meehive.com
on a side note: its weird this service won’t alphabetize (is that a word?) the above list.
We’re going to go through these sites one at a time and give a good thorough review. Then we’ll write up a comparison and be able to give the low down on who is where in this fascinating web area.
“The most valuable asset at a newspaper is still the journalist - this is even more true now. News aggregators like reddit, and the Internet in general, have diminished the brand power of media companies. Now we’re only a click away from leaving one newspaper for another. We don’t need to buy a newspaper or even change a channel. What matters now is content; sites like reddit are brand agnostic.”
