And I was talking about the value of that 'more business' which comes from Game Mechanics. Whether you're trying to sell sneakers, services or page impressions, changing the user's focus on the site will necessarily change their behavior. Which will cast doubt on the applicability of your old knowledge of how the site works.
Suppose you're trying to sell shoes and you notice that shoes with reviews sell better. So you add a karma reward system to incentivize reviewers.
The very act of soliciting those reviews with rewards will be reflected in the average review. And the relationship you formerly noted between reviews and sales, may no longer hold true. It may have been that the former relationship was truly between informative reviews and sales. And your reward system has created a flood of preference reviews which don't have the desired effect.
The goal is just to design the right game, then hopefully the right behavior will just sort of plop out of it.
Google has a similar problem regarding SEO. Their search engine is a game for SEOs to play. If they can set up the game well enough (ie if the referee can detect relevant content with high accuracy), then the "ordinary" providers of content and the SEO folks will merge. Then you don't care if somebody says they've used SEO to get on top of the search rankings, because they legitimately belong there.
Until they really have a bulletproof relevance algorithm (which is an AI complete problem, I suppose), they have to keep a lot of the rules, the proxies for relevance, hidden.
Of course, when you win the Google game, you get cold hard cash. So there is a legion of obsessive players trading tips and trying to figure out the game. I wonder if you could make a fun game for a site with hidden mechanics and unpredictable rewards, without that pot of gold at the end, that could still draw customers...
She spoke a lot about human psychology,so I wondered what her background was. I had wrongly assumed that she was a "techie"-turns out she has a PhD in Behavioral Neuroscience from University of Washington, and a BA in Experimental Psychology from UCSD! What a fascinating lady and incredibly interesting interview.
For the most part they have a pretty good system. There are definitely artists that shoot to the top of the list every time they upload something new which causes a lot of artists to be overlooked. That said, the community itself tries to make up for it by finding those underloved of songs and making them visible.
It also seems like the current system is showing some strain with a growing userbase. Things that worked a year ago don't seem to function so well anymore. Specifically, there aren't enough hearts given per day to spend on the growing number of songs uploaded. And since a song's visibility is directly tied to the number of hearts it receives, a lot of good songs go unnoticed because nobody enough hearts to go around.
Nits aside, I still love the site. It's done more for me as an artist than anything else I've done so far.
> Specifically, there aren't enough hearts given per day to spend on the growing number of songs uploaded. And since a song's visibility is directly tied to the number of hearts it receives, a lot of good songs go unnoticed because nobody enough hearts to go around.
What you've just described is a hearts recession. Paul Krugman will tell you that the obvious solution is to increase the liquidity of the hearts market. :)
E.g. Karma/Moderating systems glorify/reinforce consensus opinions to the detriment of actual discussion.
Friend counts destroy the real value of a friend connection.
Upvoting of articles leads to 'popular' articles inevitably taking the fore from on-topic articles.
etc.