Nice idea, good work on that, and best of luck with it too.

Seems like you are looking at a few more of those ‘quality signals’ that people like Matt Cutts and Tim Maher allude to.

Just have to be careful not to throw babies out with bathwater.

Not every good website has an SSL cert, not every good website bothers with Yellow pages or Business.com or the ODP. Much of ODP data is centuries old or has been bought out and sold on. How many sites admited to ODP have since expired and been rebought and filled with content of dubity I wonder.

Serious spammers also spend serious money too in getting to where they need to be. There are varying flavours and degrees of spam too, some of better quality than others. Its increasingly more difficult to differentiate too. Data is bought and sold and exchanged and rehashed. Spammers activiley seek out new data sources and aren’t particulalry fussy what its about either. Monetisation programs like adbrite , ypn, adsense, kontera, bidvertiser etc etc all help make the enterprise a worthwhile venture too.

Don’t get me wrong. I think yours is a great approach to a problem that affects us all. I think the main players in search might even look at similar signals too, or at least should be testing various filters like you describe.

Our system, SiteTruth, accomplishes much the same goal, but from the other direction. We try hard to find the company behind the web page, and if we can’t, we downgrade its search position in our system. This derates most “doorway”, “referrer”, “directory”, and “direct navigation” pages. That covers most web spam.

SiteTruth is hard on some sites, but all you have to do to get a decent SiteTruth rating is put your business name and address in a format that would work on a mailing label somewhere obvious on the site, like an “about” page. Or get a good SSL cert. Or a BBBonline membership. Or get into Open Directory. Or put your URL in a Yellow Pages ad. If we can find a solid indication of legitimacy from a non-web source, we’re happy. Everybody else in search is endlessly grinding on web pages, but not looking off the web for hard data. We do that, and it works.

Try it at “sitetruth.com”. We’re in alpha test.

db

I agree, and if it were owned by anyone other than Google then perhaps it really could be a serious problem. It’s certainly illustrative of what could befall one, guilt by association bad neighbourhoods etc.

I’d advise anyone to go the self hosted route, especially if they are serious about this stuff and are in it for the long haul. Just makes sense.