Why does everyone link their exit pages to google.com?
 

I'm working on a little program that crunches various data on all known sites and returns a number that specifies the chance of each site having adult content. A big part of this comes from whether other adult sites link to it. (A bit like PR but instead links pass "adult content juice")

This falls flat immediately with google: since so many porn sites link to it my algorithm will pretty quickly conclude that google.com is a porn site. :1orglaugh

I wont be hiring you for programming ...

i guess they are all hoping for a recip link back ;-)

lol...but still a good idea I think. Google and a couple other search engines will probably be a few anomalies that you can manually cut out, yeah?

or if not one porn word is on the page???

because if the site isnt what the visitor is looking for google will help them find it...

Yep there are also negative heuristics which will pull the "adult content" score down. Google does not have any adult terms on its page (AFAIK) nor does it link directly to any that do.

Do people reeaaallly think that in 2008 linking to google.com will better their SE positions? That's the kind of rumour that does benefit google. :winkwink:

And some people still actually believe linking to google will help their ranking.

I wonder if google is the #1 mainstream site with the most links from filthy adult sites. Don't tell their shareholders! :Graucho

I'll probably be able to answer that question definitively in a few days.

Clearly, as an industry we need to take some initiative here, and link all our exit pages to toysrus.com instead of google.

why doesn't your algorithm take into account %'s, not just raw number of sites. I'm sure by %, the number of adult sites linking to google is small.

Anyway, many SEO "experts" a few years ago said (incorrectly I'm sure) that linking to google increases your ranking on google.

hmm, what good is figuring out if the site is adult or not anyway? It's probably useful to google, and few other companies, but it seems like a pretty useless algorithm, or am I missing something?

Yeah i hear google puts links on the main page

LMAO!:1orglaugh:1orglaugh:1orglaugh

I usually link to my mainstream sites and/or forums.

Plenty of applications where you'd want to filter out (or against) a list of adult sites. I've got the raw data here so I may as well do something with it. :)

I doubt adult sites are linking to google outside of their splash page. So just spider a few pages, and tell the algorithm to not rate the splash page.

I suppose you could flag a site as adult from it's 2257 linkand flag a site that links to adult content from scans of sites the site links to.

However, lots of already well-established sites out there do this kind of job. Europe gave a directive earlier this year on what software ISPs should use for offering filtered content. British Telecom has been filtering sites for over 3 years, and Australian govt is pushing hard for mandatory adult filtering.

But good luck

I am not processing this data specifically for creating block lists. I wouldn't dare offer a computer generated list as an authoritative source since it would most likely contain a number of missed sites and also false positives.

There are plenty of other applications. For example, if a webmaster wanted to find similar adult sites for link trades, and his niche was "teen" it would not always be possible to determine whether it was an adult site based solely on the domain name.

I have always linked exit traffic to my own mainstream site:thumbsup