Remember when I asked you all what you thought about the idea of turning over the reins and letting you post technology, search and media-related stories to Searchblog? Inspired by Piers Fawkes, an FM author, who used Pligg to do Marktd, I’ve been working on a new feature at Searchblog I’ve come to call SearchMob. Ok, maybe not the best name, but at least there’s alliteration. The idea is simple – you can register and submit stories to the site. Then readers vote the stories up the ranks, similar to Digg, Reddit, and Newsvine (yes, all FM sites).
While I’d love to work with one of those sites, making Searchblog better isn’t their focus (though Reddit does help other sites, like Lipstick.) For this test anyway, I’m going with Pligg, an open source application. No doubt there will be bugs and incomplete features, but I’ve found the platform pretty reliable so far. I’ve asked a few folks who send me mail suggesting links to use Pligg and play around with an early version, and so far it’s been a lot of fun. So take a look. I’d love for you all to bang on it, read it, and tell me how to do it better.
Many thanks to Jonathan Schreiber, FM’s amazing author services engineer, for helping me make this happen.
Hi John,
Have you (or your developers) made any changes to the algorithm Pligg uses to determin what is most popular? I had a look at Pligg some months back but dropped it when I saw in the code it ranks stories simply by the date they were published (ie obtained the threshold number of votes to appear on the front page of the site).
Any further significant voting after the story is published to the main page doesn’t effect the weighting of the story. Said story will simply cascade down the rankings as new stories are voted up. (Digg, for example will repromote stories if they are heavily dugg etc etc)
Pligg is a nice ‘shell’ for a Digg-like site, but it lacks an algorithm and for that reason I don’t believe it’s mature enough to use.
If you’ve done any work on this area of your Pligg install could you let the community have a copy of your algorithm?
Many thanks
Ben