Monday, April 30, 2007

Digg Cheating

I suggested essentially the same algorithm for how an open-source search engine could work without being vulnerable to gaming even by those who understood all of its inner workings. The main difference, of course, is that Digg and reddit actually exist now. Digg declined to comment on the possible merits of such an algorithm; reddit's Steve Huffman said that the idea sounded interesting, although even if the idea got full buy-in, naturally any proposed change would take a long time to bring to fruition.But it seems that an algorithm similar to this one would be the only way to prevent cheating on sites like Digg that sort content based on user votes. So it's ironic that HotOrNot, the only site I know of that is using a variation of this algorithm and hence is probably the most secure against cheating, is also the one where cheating is least likely to be a problem. Getting a high placement on Digg might enable you to make some money, but getting a highly rated picture on HotOrNot isn't going to make you rich (unless it helps you meet a millionaire who is using the site to find his third wife). Also, making HotOrNot meritocratic doesn't give people an incentive to improve the "content" that they submit, because up to the limits of what can be done with hair and wardrobe, you can't make yourself that much more attractive. With Digg and reddit, on the other hand, I might work harder at submitting a good story, if I knew that it worked in a perfectly meritocratic fashion that pushed good stories right to the top.

Slashdot | How to Stop Digg-cheating, Forever

I don't know about using any type of algorithm to stop digg cheating.  It seems somewhat impossible to do.  If you implemented this type of algorithm, how would you allow affiliate sites to have a "Digg This" button on them?  The users would be in essence voting on a single story by going directly to the URL.  It is impossible to come up with any type of algorithm, no matter how clever to stop people from cheating on popular sites.  Someone out there with way more time, and sometimes more money than you will always defeat it. 

I think that Digg has done a good job in that their users can bury a story.  So the fact that the story was quickly buried is an example of this.  And as the linked to wired story indicates, the story was quickly buried once it got to the front page.  So what if she got a ton of traffic, 100,000 hits is nothing, probably less than 1% of the total traffic that Digg forwards each day.  Saying that it is a valuable way of gaming adsense is clear.

As I have said before, I don't know what the answer is, but technical solutions such as clever algorithms won't work, there is just too much money at stake, so someone will always reverse engineer it. 

Perhaps a genetic algorithm would work, but having tons of clones that all needed to learn and sync would hurt performance and initial efficacy.  You'd have to install genetic an genetic algorithm, train it for years, and then deploy it to make it effective, and do the syncs when search traffic were at a minimum.  Its tough.

technorati tags:, , , , ,

Blogged with Flock

No comments: