Friday, May 05, 2006

Which reminds me of a porn empire I once tried to create

The incomparable Ms Fits recently received home-made amateur porn from a reader. Apart from the fact that she lives in a bizarre fantasy land where complete strangers make rude movies and give them to her, the story reminded me of the time I tried to establish a porn empire back in the heady days of dot-com wonder.

No, seriously.

Being the geeks that we were (there were four of us in on the plan), our idea revolved around a Content Management System or CMS for porn. I still find this funny. I have the data model design we created for the underlying database ... geeks working on porn, and we were entirely worried about the system architecture. I don't think any of us actually looked at porn.

One of the guys on our 'team' had actually run several successful porn sites (making $1000's a week), so we had a fair idea of what we needed to do, which was all very educational.

Turns out that there are only relatively few porn sites that charge money and have actual 'content'. Then there are thousands (millions?) of 'thumbnail sites' that offer a few (maybe a dozen) free images or movies (generally previews/segments of a larger film). The free thumbnail sites are basically ads for the actual pay porn sites. People look at the free stuff, and then maybe click an ad and get sent to an actual site where they sign up and pay lots of money to look at porn. At the time, if you referred someone from a thumbnail page you would get anywhere up to $50 US. To assist the punter in all this, there are several aggregators that basically have lists of all the free thumbnail sites around. People go to the aggregator, find thumbnails, sign up = profit.

Our idea was to completely automate the process of building thumbnail sites and posting them to the aggregators. The more sites you can generate, the more potential revenue you can make. At least that's the idea. And the sites can be pretty neatly targeted ... most porn is pretty spefically tailored to a niche market.

So ... our vision was to classify a massive porn library into categories (indeed, a poly-hierarchic taxonomy), and then you could just enter into the system 'teen hardcore asian' and it would generate a thumbnail site for you, with pay-per-click ads and everything. Our porn-industry friend had something like 50gb of royalty-free porn images (which you can just buy, I never knew that), which we were going to use.

It was at this point that we realised that it was about 2000 hours of work to classify the 50gb of porn.

So we gave up ... looking at that much porn would totally warp your mind. I'm not actually into porn (I know you're not going to believe that now) so like many of my hare-brained schemes the reality was all bit too much to deal with.*

I read in The Age a couple of years ago that one of the largest internet porn company had started making more money from the software they had built to run their empire than they actually did from the porn itself.

It's still a nice data model though.


* Earlier this year I started studying for my Water Fowl Recognition Test so I could get a duck hunting license. I lost enthusiasm when I realised that this meant I would have to kill cute ducks. I quite like ducks, used to have some as pets.


No comments: