The backend had a problem too. Whenever there was an error, I didn’t know about it until there was something really odd happening with the stat numbers. I don’t like to do live debugging and I rarely get a chance to do it at home. So I added a bit of logic that if the branch of unknown (if something the system can’t handle happens) it’ll send down a parse report in json. It’s pretty neat. Also, the Curl system that was used to fetch the remote facebook/fluff friend pages has been updated too. It now randomizes the user agent and the referrer, which makes it harder for people to get mad at me.
I added a referrer catcher in the summer update. Since then, there has been about 370 unique (and just about everything makes it unique) referrer to the fluff grabber main-page. Most of clicks are coming from the fluff friend’s forum.
Since the summer update of 2008 (oh, rhyme!), there has been an explosion of fluff grabber growth. I’m not sure of the exact number of fluff books we had indexed then, but at this point we have about 232,000 fluff books in the database. Can you believe it? There is also about 1000 or so fluff books added daily when there isn’t much activity via the crawler. We’ve also surpassed the 10,000 removed mark which is great. It’s sad to see fluff friends leave but it’s good to keep the database clean of bad books. Currently, we’re around 1,502,780 clicks and that’s about 7,513,900 munny. It’s an unbelievable amount.
Fluff friends finally started their own super zoo. Of course, it’s slightly less friendly and much slower than the fluff grabber. It’s organized by newest fluffs, neglected fluffs and I believe famous fluffs. It’s slow too, since it’s on facebook. YSlow claims it took about 29 seconds for that page to load. YSlow claims too, that it took about 5 seconds to load the fluff grabber browse page. Big difference?
I want to make some improvements to the load and feel, I always do. It’s hard to change though because I suck at design and graphics. I’d like to add a static list for the system. Every single fluff book would be listed on a paginated list that went on forever. Google would probably like that. I want a new fluff grabblet. The current one can’t keep track of it’s users. I’d like to add a sessioning system addition that would also keep track of how many clicks a specific user starts. I’m not sure how to keep users separate though.
The fall update isn’t a large update but it is important in the ongoing evolution of the fluff grabber system. If you have any questions, comments or concerns, please, leave a comment on this post.. I’ll try to get back to you as soon as I possibly can.